List updated languages and completed batches in Upcoming. Record the current untranslated counts, next batches and review requirements under TODO Later.
Thanks to xet7 !
Fill 46 English placeholders in Kurdish and Central Kurdish. Preserve existing translations, exact placeholders, numeric limits and literal filter query syntax; extend archive/date regression coverage to both locales. Seven relevant suites and 21 human-preference checks pass, with the affected regression rerun after wording refinements. Preservation audits confirm only English placeholders changed.
Kurdish and Central Kurdish wording is provisional and would benefit from speaker review.
Thanks to xet7 !
Fill 69 English placeholders covering automatic archiving, recency, inclusive date ranges, due dates and time in a list. Preserve existing translations and exact query examples. Add regression checks for placeholders, source order, numeric limits, query syntax and negative behavioral guidance.
Eight applicable suites, 21 human-preference checks and per-locale preservation audits pass. The affected regression was rerun after refining Somali template terminology. Turkmen, Tatar and Somali wording is provisional and would benefit from speaker review.
Thanks to xet7 !
Fill the final three Cherokee placeholders, reaching all 234 non-English catalogs. Assert exact catalog coverage so new locales cannot silently escape the import/reordering checks. Preserve existing translations, key order, exact placeholder inventories and import identifiers. All 458 translation-related suites, eight focused suites, 21 human-preference checks and the preservation audit pass.
Cherokee wording is low confidence and needs speaker review. Drag and paste vocabulary follows the Peoples Paths Cherokee word lists: https://www.thepeoplespaths.net/Cherokee/CherokeeWordLists/WordList-D.htm and https://www.thepeoplespaths.net/Cherokee/CherokeeWordLists/WordList-P.htm
Thanks to xet7 !
Fill three English placeholders and extend exact-token, key-order, import-identifier and keyboard-direction coverage to 232 locale variants. All 57 relevant suites pass, with the affected regression rerun after terminology refinement; 21 human-preference checks and a preservation audit pass.
Tigre wording is low confidence and needs speaker review. Vocabulary checked against the BeitTigreAI parallel corpus dictionary: https://beittigre.github.io/tigre-multilingual-dictionaries/english/index.html (pull, arrow, key, line and step).
Thanks to xet7 !
Fill three English placeholders, preserving import identifiers, source key order and exact placeholder inventories. Extend import/reordering regression coverage to 231 locale variants. Eight relevant suites and 21 human-preference checks pass; the preservation audit confirms no existing translations changed.
Inuktitut wording is low confidence and needs speaker review. The pull verb follows the uniaq-/uniar- examples in Viewpoint Aspect in Inuktitut, pages 98 and 102: https://www.collectionscanada.gc.ca/obj/thesescanada/vol2/OTU/TC-OTU-32898.pdf
Thanks to xet7 !
Fill three English placeholders in Tifinagh and extend import exclusion, keyboard direction, key order and placeholder checks. Nine relevant suites, a preservation audit and 21 human-preference checks pass. These technical translations are low confidence and need speaker review; dictionary vocabulary informs the drag action.
Thanks to xet7 !
The Compose files run FerretDB in its own container, where WeKan cannot see
the SQLite file, so continuous backup could stream only files there. As
decided, FerretDB now starts with --repl-set-name=rs0 and keeps
local.oplog.rs, and WeKan connects with directConnection=true: FerretDB
advertises its listen address 0.0.0.0 as the member, which a discovering
driver would follow out of the container. WeKan itself still polls.
Running the oplog engine against FerretDB found that it asked for
noCursorTimeout, which FerretDB refuses, so the tail never started there; it
no longer does, and the continuous backup integration tests pass against
FerretDB with an oplog. FerretDB did not record dropped collections; that is
fixed in wekan/FerretDB 3b111eac for its next release.
Thanks to xet7 !
A rule that moved a card to another board and then sent an email sent nothing:
the email source check refused a card that had left the activity's board, in
both engines. As decided, the email now reads the card on the destination
board, but only when this run of the rule made the move, the destination
opted into Sync effects, and the card is still there. The ordinary engine
remembers its run's own move (RulesHelper.emailActivity), durable Sync proves
it from a saved move of the same plan (ruleEmailActivity, ruleCardNow) and
rechecks it before sending. The stored source binding (version 6) records the
board the card left. Any other card off the board is refused as before, and
an email may now follow a move to another board in a durable plan.
Tests: binding positive and negative cases, a shape guard that nothing else
can claim a followed move, and a server test of both engines that sends once
from the destination and refuses a card nobody's run moved, a move to another
board, and a destination that opted out.
Thanks to xet7 !
Fill three English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Eight relevant suites, a preservation audit and 21 human-preference checks pass. These technical translations are low confidence and need speaker review; Greenlandic educational sources inform movement vocabulary.
Thanks to xet7 !
Fill three English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Eight relevant suites, a preservation audit and 21 human-preference checks pass. These technical translations are low confidence and need speaker review; dictionary vocabulary informs the drag action.
Thanks to xet7 !
Fill three English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Nine relevant suites, a preservation audit and 21 human-preference checks pass. These technical translations are low confidence and need fluent-speaker review; lexical references distinguish pulling along from tugging a handle.
Thanks to xet7 !
Fill three English placeholders and replace the Esperanto comments label with Volapuk. Extend import semantics, direction, key order, placeholder and corrected-label checks. Eight relevant suites, a preservation audit and 21 human-preference checks pass. These technical translations and the replacement are low confidence and need native-speaker review.
Thanks to xet7 !
Fill three English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Seven relevant suites, a preservation audit and 21 human-preference checks pass. These technical translations are low confidence and need native-speaker review; dictionary vocabulary informs the drag action.
Thanks to xet7 !
Fill six English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Eight relevant suites, two preservation audits and 21 human-preference checks pass. Both technical translations are low confidence and need native-speaker review; dictionary references inform drag vocabulary.
Thanks to xet7 !
Fill three English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Seven relevant suites, a preservation audit and 21 human-preference checks pass. These technical translations are low confidence and need native-speaker review; dictionary vocabulary informs the drag action.
Thanks to xet7 !
Fill three English placeholders and replace Zulu import-source and Venda linked-subtask text with Veps. Extend import semantics, direction, key order, placeholder and corrected-label checks. Eleven relevant suites, a preservation audit and 21 human-preference checks pass. These technical translations and replacements are low confidence and need native-speaker review.
Thanks to xet7 !
Fill three import and reordering placeholders and correct five Zulu seed values in Venda labels, checklist, comments, lists and import action. Extend regression coverage for these terms and import semantics. Seven relevant suites, a preservation audit and 21 human-preference checks pass. The technical translations and replacements are low confidence and need native-speaker review.
Thanks to xet7 !
Fill six English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Eight relevant suites, two preservation audits and 21 human-preference checks pass. Both technical translations are low confidence and need native-speaker review. Dzongkha computer terminology and Tibetan lexical references inform the drag action.
Thanks to xet7 !
Fill nine English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Ten relevant suites, three preservation audits and 21 human-preference checks pass. All three technical translations are low confidence and need native-speaker review; dictionary vocabulary informs the drag actions.
Thanks to xet7 !
Fill nine English placeholders and extend import exclusion, keyboard direction, key order and placeholder checks. Eight relevant suites, three preservation audits and 21 human-preference checks pass. All three technical translations are low confidence and need native-speaker review. Dictionary evidence distinguishes Sakha sos (drag) from sohuy (startle).
Thanks to xet7 !
Fill nine English placeholders and extend import semantics, keyboard direction, key order and placeholder regression coverage. Ten relevant suites, three preservation audits and 21 human-preference checks pass. These low-confidence technical translations need native-speaker review; dictionary references informed Manx straneys and tayrn and Cornish gnas and draylya.
Thanks to xet7 !
Add import and reordering guidance while preserving existing translations, source tokens, commands, formats, import exclusions and keyboard directions. Ten relevant suites, 21 human-preference checks and three preservation audits pass. Technical wording in these three languages has low confidence and welcomes native review.
Thanks to xet7 !
Add import and reordering guidance while preserving existing translations, source tokens, commands, formats, import exclusions and keyboard directions. Ten relevant suites, 21 human-preference checks and four preservation audits pass. Technical wording in these four languages has low confidence and welcomes native review.
Thanks to xet7 !
Add Uyghur, Arabic-script Uzbek, Konkani and Oromo guidance. Correct the Uzbek labels heading to Arabic script. Preserve existing correct-language translations, source tokens, commands, formats, import exclusions and keyboard directions. Ten relevant suites, 21 human-preference checks and four preservation audits pass. Technical wording has low confidence and welcomes native review.
Thanks to xet7 !
Add Polish, Czech, Slovak, Slovenian, Croatian, Romanian, Hungarian, Bulgarian, Ukrainian and Russian translations. Correct Italian labels in Romanian and Cyrillic description text in Croatian. Preserve existing correct-language text, commands and source tokens. Eleven relevant suites and 21 human-preference checks pass.
Thanks to xet7 !
Translate the remaining 35 backup messages and discover every non-English locale in the backup regression suite. Preserve existing translations and exact source tokens. Eight relevant suites and 21 human-preference checks pass. Cherokee technical wording has low confidence and welcomes native review.
Thanks to xet7 !
Create returns to the Labels popup with the card still open, so the test's
second click on the minicard closed the card and the label selector was never
found. It now checks the new label is listed, closes the popup and reopens the
label editor on the open card.
Thanks to xet7 !
Fill the remaining 35 backup messages in each locale. Preserve existing translations, time units, technical identifiers and restore safety wording. Cover all 43 backup messages in 233 locale variants and guard the Tamazight software execution term. All 31 relevant suites and 21 human-preference checks pass. Technical wording in both languages has low confidence and welcomes native review.
Thanks to xet7 !
Translate the remaining 35 backup messages in each locale, preserving existing translations and technical identifiers. Extend all 43 backup-message checks to 231 locale variants, including restore safety wording and time units. All 59 relevant suites and 21 human-preference checks pass. Technical wording in both languages has low confidence and welcomes native review.
Thanks to xet7 !
Previewing a Sync source saved with an internal host NAME before the SyncBleed
fix was recorded twice: as SyncBleed (medium) by server/listSync.js and as
DnsBleed (high) by the shared fetchSafe DNS check. The high one went through
blockOnSecurityEvent and disabled the account of the member who pressed
preview, who may not be the one who saved the address, and logged them out.
fetchSafe takes recordBlocked: false for a caller that records the refusal
under its own name; the request is refused all the same, and every other
caller keeps recording DnsBleed by default. tests/syncBleed.test.cjs pins the
option, the default, and that only List Sync turns it off; the Playwright
SyncBleed test checks the member stays logged in after the preview.
Thanks to xet7 !
scrum-chart-scope read the closed sprint's card rows from scrum.getBoardData,
which now carries the computed report and never the rows (they live in
scrumSnapshotRows). It now asserts the rows are absent, the committed report
counts only the visible card, and the hidden estimate is nowhere in the answer.
board-settings-rules-views expects the Scrum settings entry at the end of the
Board Settings menu, where it now is.
Thanks to xet7 !
Fill eight backup encryption, cloud and restart messages in Cherokee, Inuktitut, Tigre, Wolaytta and Standard Moroccan Tamazight. Synchronize eleven English variants without replacing their existing strings. Correct Swiss German spelling and the attested Tigre Files plural. Discover all non-English catalogs in backup regressions and allow the exact AES-256-GCM identifier in Arabic-script Uzbek checks.
All 455 translation-related suites pass, including reruns after the six failures were fixed. Technical wording in the five newly filled languages has low confidence and welcomes native review. Older untranslated backup messages and other catalog gaps remain.
Thanks to xet7 !
Add Cherokee, Inuktitut, Tigre, Wolaytta and Standard Moroccan Tamazight progress and retry messages. Discover all non-English catalogs in regression coverage and check this batch's declared scripts and exact source tokens. Preserve existing translations. Technical phrasing in all five languages has low confidence and welcomes native review. Backup translations and other catalog gaps remain.
Thanks to xet7 !
Fill Klingon, Volapuk, Northern Sami, Veps, Nahuatl and Greenlandic. Preserve existing translations and exact progress tokens with regression coverage in 229 locale variants. Technical phrasing in all six languages has lower confidence and welcomes native review.
Thanks to xet7 !
Fill Akan, Ewe, Wolof, Bambara, Fulah, Aymara, Aromanian, Ladin, Manx and Cornish. Preserve existing translations and exact progress tokens with regression coverage in 223 locale variants. Technical phrasing in all ten languages has lower confidence and welcomes native review.
Thanks to xet7 !
Fill Tigrinya, Tibetan, Dzongkha, Kashmiri, Chuvash, Buryat, Sakha, Venda, Guarani and Quechua. Preserve existing translations and exact progress tokens with regression coverage in 213 locale variants. Technical phrasing in this batch has lower confidence, especially Dzongkha, Kashmiri, Chuvash, Buryat, Sakha, Venda and Quechua, and welcomes native review.
Thanks to xet7 !
Fill Scottish Gaelic, Breton, Friulian, Venetian, Walloon, Samoan, Tongan, Fijian and Hawaiian messages. Preserve exact source tokens and existing translations, with regression coverage for 193 locale variants. Walloon, Tongan, Fijian and Hawaiian technical phrasing has lower confidence and welcomes native review.
Thanks to xet7 !
Fill Corsican, Sardinian, Sicilian, Neapolitan, Papiamento, Tok Pisin, Bislama, Maori, Cantonese and Hiragana Japanese progress and retry messages. Preserve source tokens and existing translations. Regression coverage reaches 174 locale variants. Sardinian, Neapolitan, Papiamento, Tok Pisin and Bislama technical wording has lower confidence and welcomes native review.
Thanks to xet7 !
Add progress and retry instructions with exact source tokens while preserving existing translations. Extend regression coverage to 164 locale variants. Igbo, Shona, Chichewa, Sesotho and Setswana technical phrasing has lower confidence and welcomes native review.
Thanks to xet7 !
Add progress and retry messages with exact source placeholders. Preserve existing translations and extend coverage to 138 locale variants. Faroese, Maltese and Romansh technical wording has lower confidence and welcomes native review.
Thanks to xet7 !
Add progress and interrupted undo/redo instructions with exact source tokens and regression coverage. Assamese, Odia and Sindhi technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add progress and interrupted undo/redo messages in Uzbek variants, Kazakh, Kyrgyz, Mongolian, Tajik, Turkmen, Tatar and Bashkir. Preserve existing translations and exact tokens. Technical terminology in Turkmen, Tatar and Bashkir is low confidence and needs native-speaker review.
Thanks to xet7 !
Translate progress and interrupted undo/redo instructions while preserving existing translations and exact tokens. Extend regression coverage to 106 locale variants.
Thanks to xet7 !
Fill progress and interrupted undo/redo instructions, preserving existing translations and exact progress tokens. Add regression coverage and verify all 47 changed catalog files.
Thanks to xet7 !
Add sixteen encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add eight encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add eight encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add eight encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs fluent-speaker review.
Thanks to xet7 !
All spec files with a baseline failure ran again in Chromium: 175 of 180
passed. Two need WEKAN_FILES_PATH and pass with it, one was timing and
passed on its re-run, and F14 checked the Board Settings / Card popup as it
was before rows were dragged and shown on both sides. F14 now checks the
same intent - a row with no place does nothing on Enter, the popup stays
open, those rows toggle on both sides, the two lists stack when narrow -
and the file passes in Chromium and WebKit.
Thanks to xet7 !
Add sixteen encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add sixteen encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add eight encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add 24 encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add sixteen encryption, cloud and restart strings with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add eight encryption, cloud and restart options with regression coverage. Technical terminology is low confidence and needs native-speaker review.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and the eight-string preservation audit pass.
Thanks to xet7 !
A copy checks access when it starts; a membership revoked while it runs
does not stop it, which is the documented atomicity property of the
backends rather than a missing check.
Thanks to xet7 !
Reviewing the legacy direct Rules writes: a board admin of their own board
could save a rule that named a trigger of a board they do not administer.
Only that board's own rules run on its activities, but the matcher took a
trigger's rule from Trigger.getRule() - whichever rule named it first - so
the other board's own rule could silently stop running. The matcher now
looks up the trigger's rule on the activity's board, and a DDP insert or
update of a rule naming another board's trigger is refused and recorded in
Admin Panel / Problems under RepointBleed (no client names one: the rule
editor creates the trigger on the rule's board first).
Tests: tests/repointBleed.test.cjs (the decision for insert, $set,
$addToSet and $push, the negatives, and the matcher's source), and a server
test that, without the change, accepted the foreign rule and lost the
victim board's rule.
Thanks to xet7 !
Add eight encryption, cloud and restart strings using the existing Uzbek translations as a reference. Extend placeholder and Arabic-script coverage. Preserve existing translations. Arabic-script spelling and technical wording are low confidence and need native-speaker review.
Validation: four focused suites, all 21 human-preference checks and the eight-string preservation and script audit pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
A client could insert a card on a board it may write to while naming the
list and swimlane of another, private board. The card stayed on its own
board, but its creation activity carried that list's and swimlane's titles
and the board's rules saw them, so a member could read the names of
another board's lists by id. The server's own copies already check their
destination (server/lib/cardCopyDestination.js); DDP inserts now do too,
and refuse such a card. No client sends one, so the attempt is recorded in
Admin Panel / Problems under BoardBleed (canary card.foreign-placement).
Found while reviewing the client-side template copy writes, which go
through the same Cards insert rules: a template or linked card is always
created on the board that owns its list.
Tests: a server test that refuses another board's list and swimlane with
no card or activity written and allows the board's own, failing without
the rule; the canary catalogue and coverage suites; and the template,
linked-card, card-copy and card browser specs in Chromium.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and three eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation, script and placeholder audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
The Upcoming release leads with it; TODO Later and the card-copy boundary
audit record the assigned-only source and descendant review as done, with
what remains to review.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
A copy of a card took every subtask of it. An assigned-only member who
copied a card assigned to them got copies of its subtasks assigned to
others, and any member got copies of its subtasks on boards they cannot
read - a parent may have subtasks on another board (#3626). Card.copy and
the durable rule copy now carry only the subtasks the copier could read and
copy themselves (models/lib/boardCardScope.js copyableSubtasks); a copy
with no acting user, made inside the server, keeps them all as before.
Not recorded in Admin Panel / Problems: copying a card with subtasks is
ordinary use, and the copy simply leaves out what the copier may not have.
Tests: tests/copySubtaskScope.test.cjs (the decision, and a scan that every
place copying subtasks filters them) and a server test through copyCard,
which without the fix gave an assigned-only member copies of another
member's subtask and of one on a private board.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass; Amharic script was checked.
Thanks to xet7 !
TODO Later said a sent message could not be taken back; but email is
durable on the card's own board. The reason is the source-binding check of
server/lib/ruleEmailSource.js: both engines refuse to email a card no
longer on the board where its activity happened, so the email sends
nothing either way, and changing that is a security decision.
Thanks to xet7 !
The ordinary engine takes a move-all's list from the board the rule's card
is on now (RulesHelper.ruleBoards); the durable move-all onto another board
took it from the plan's board, so it could not follow a move to another
board. It now takes the list from where the card is, saves that board as
`fromBoard` - only when it is not the plan's, so saved commands still
validate - and each unit leaves from it. A move-all onto the board the card
is already on is refused. Only an email now keeps such a plan on direct Sync.
A server test moves the card from A to B, then every card of B's list to C,
durably and with the ordinary engine, with the same result.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology needs native-speaker review, particularly the lower-confidence Yiddish wording.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation, script and placeholder audits pass.
Thanks to xet7 !
TODO Later no longer lists the multi-server noise of the background jobs,
fixed by the lease, and its remaining cross-board rule limits are an email
or a move-all onto another board after a move to another board.
Thanks to xet7 !
A rule whose card had been moved to another board could not move it on to
yet another one through durable Sync: the cross-board move command took
the card from the plan's board only. It now takes it from where it is -
the plan's board, or where an earlier move of the same plan took it
(ruleCardNow) - and saves that board as `fromBoard`, written only when it
is not the plan's, so every command saved before still validates. The
board the card leaves, its dependencies and the deferred hook records
follow the saved move's own before-place. A move onto the board the card
is already on is refused.
Such a further move may now follow a move to another board; every board
the chain reaches must have opted in, as the first. A move-all onto
another board still may not follow, since it moves a list of the rule's
board. A server test runs A to B to C and a later colour, durably and with
the ordinary engine on a twin card, and refuses the chain while C has not
opted in.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology needs native-speaker review, particularly the lower-confidence Odia wording.
Validation: four focused suites, all 21 human-preference checks and three eight-string preservation, script and placeholder audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
The background rollover and the background undo or redo of a large batch
were resumed by every server that found them after a restart; two of them
on one job wrote alike, and the card writes' compare-and-set stopped one
with a conflict the Scrum view showed as a failure. Each step now first
takes or renews a lease on the sprint or the job, one minute long, that a
server which went away loses when it runs out. A live lease of another
server leaves the job to it: a retried close says the rollover is running
there, and undo and redo on the board wait. The lease is not History
content and is not sent to the browser.
Tests: a foreign live lease leaves both the rollover and the undo job
untouched, and an expired one is taken over.
Thanks to xet7 !
Record the translation batch, validation and terminology review needs in Upcoming. Both changelog suites pass. Global completeness suites still report missing backup and Scrum strings.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to Xhosa and both Zulu catalogs, and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and three eight-string preservation audits pass.
Thanks to xet7 !
Upcoming gains continuous backup's encryption, cloud upload and restart
restore, large sprints closed and undone in the background, rule actions
that work where their card went, and the applyListWidth test fix. TODO
Later records them as built and what remains of each: an email or a second
move elsewhere after a cross-board move keeps direct Sync, two servers
resuming one large-sprint job stop one with a conflict, and the upload was
not tested against a live cloud account.
Thanks to xet7 !
Since 2026-10-02 a list's width is a shared board setting that needs write
access, read from the board's members; the test's stub board had none, so
the positive case failed. It now has a writing member, and a read-only
member is refused without a write.
Thanks to xet7 !
As decided on 2026-10-03, after a rule moved its card to another board, a
later action that names the rule's own board works where the card is now,
in the ordinary engine and in durable Sync alike:
- RulesHelper.ruleBoards gives the board the card is on and the board an
action works on. A move to the top or bottom, a sort and a move-all
resolve their list and swimlane names there; the write-access check
covers that board. Before this, a move to "*" after a move to another
board took the card's new list and swimlane but the old board, leaving
the card split across two boards.
- The durable move, sort, move-all and archive commands find the card
through ruleCardNow and save the board it is on as `onBoard`, written
only when it is not the plan's, so every command saved before still
validates; the archive's cascade takes the children on that board.
- These actions may now follow a move to another board, so such a plan
stays durable. An email still may not, nor a second move to yet another
board.
The archive reservation and follower guards follow the decision; the
cross-board move test now expects sorting and archiving to follow and
refuses an email instead, and a new server test runs the whole chain
durably and with the ordinary engine on a twin card, with the same result.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale. Extend placeholder coverage and guard hiragana and Arabic scripts. Preserve existing translations. Sindhi and Uyghur technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and three eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs language review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and exact placeholders. Hawaiian technical wording is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
As decided on 2026-10-03, every per-card guard stays, and the browser
never waits on one call for the whole of a large sprint:
- A close of more than one chunk (2,000 cards) returns once the sprint is
closed and its plan stored; its rollover continues in the background,
chunk by chunk under the board's lock, recorded under the close's named
History batch. Its progress is on the sprint, a failure is kept there for
the Scrum view's resume button, and a restart resumes it. A smaller
sprint is still closed and rolled over in one History row in the call.
- An undo or redo whose batch has more than four rows left applies the
first row in the call and the rest as a background job (scrumBatchJobs),
as the user who asked and with the same derived request IDs, so a retry
or a restart resumes it. Undo and redo on that board wait for it.
- The Scrum view shows both as progress bars and polls quietly while one
runs.
The board-copy, native-import and Scrum menu specs read snapshot rows
through a helper now that they live outside the sprint, and the sprint
report's chart is no longer counted together with the daily and
change-by-change charts below it, which loaded in WebKit first.
Thanks to xet7 !
Add eight encryption, cloud and restart strings each for Kurmanji, Sorani and Pashto, and extend regression coverage. Preserve existing translations and exact placeholders. Technical terminology is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and three eight-string preservation audits pass.
Thanks to xet7 !
Record the translation batch, tests and terminology review needs in Upcoming. Both changelog suites pass; the global completeness suites still report untranslated backup and Scrum strings.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each locale and extend regression coverage. Preserve existing translations and placeholders. Wu technical wording is low confidence and needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and both eight-string preservation audits pass.
Thanks to xet7 !
Add eight encryption, cloud and restart strings to each of three catalogs; the Khmer alias shares the completed catalog. Extend encryption and cloud regression coverage. Technical terminology needs native-speaker review.
Validation: four focused suites, all 21 human-preference checks and per-catalog placeholder and preservation audits pass.
Thanks to xet7 !
Record the eight-string translation batch, regression coverage and terminology review needs in Upcoming. Both changelog validation suites pass.
Thanks to xet7 !
Add eight restart, encryption and cloud options and extend translation regression coverage. Preserve existing translations and exact placeholders. Technical Yoruba terminology is low confidence and needs native-speaker review.
Validation: four focused suites and all 21 human-preference checks pass; the eight-string audit passes.
Thanks to xet7 !
Add eight restart, encryption and cloud translations while preserving existing values and exact source tokens. Extend encryption and cloud regression coverage. Technical Hausa wording is low confidence and needs native-speaker review.
Validation: preservation and relative key-order audits, four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Add 16 restart, encryption and cloud translations while preserving existing values and exact source tokens. Extend the encryption and cloud regression suites to both locales. Technical Malagasy wording is low confidence and needs native-speaker review.
Validation: exact-change and relative key-order audits, four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Add 16 restart, encryption and cloud translations while preserving existing values and exact source tokens. Extend the encryption and cloud regression suites to both locales.
Validation: exact-change and relative key-order audits, four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Add 16 restart, encryption and cloud translations while preserving existing values and exact source tokens. Extend the encryption and cloud regression suites to both locales.
Validation: exact-change, script and relative key-order audits, four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Add 16 restart, encryption and cloud translations while preserving existing values and exact source tokens. Extend the encryption and cloud regression suites to both locales.
Validation: exact-change, script and relative key-order audits, four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Add 16 restart, encryption and cloud translations while preserving existing values and exact source tokens. Extend the encryption and cloud regression suites to both locales.
Validation: exact-change, script and relative key-order audits, four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Add 16 restart, encryption and cloud translations while preserving existing values and exact source tokens. Extend the encryption and cloud regression suites to both locales.
Validation: exact-change, script and relative key-order audits, four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Add 16 restart, encryption and cloud translations while preserving existing values and exact source tokens. Extend the encryption and cloud regression suites to both locales.
Validation: exact-change, script and relative key-order audits, four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Add 80 restart, encryption and cloud translations for Latin-script Uzbek variants, Kazakh, Kyrgyz, Tajik, Mongolian, Turkmen, Tatar and Bashkir. Preserve existing values and exact source tokens. Extend script, vocabulary and encryption/cloud regression coverage. Technical Turkmen, Tatar and Bashkir wording is low confidence and needs native-speaker review.
Validation: ten preservation audits, four focused suites and all 21 human-preference checks pass. The restart option still needs translation in 104 locales; the broader translation backlog remains.
Thanks to xet7 !
Add restart, encryption and cloud labels for Irish, Welsh, Luxembourgish, Maltese, Faroese, Frisian, Occitan, Asturian, Aragonese, Corsican, Sicilian and Sardinian, including Welsh and Frisian variants. Preserve existing translations and exact source tokens. Technical Faroese, Occitan, Asturian, Aragonese, Corsican, Sicilian and Sardinian wording is low confidence and needs fluent-speaker review.
Validation: 112 translations pass preservation and key-order audits; four focused suites and all 21 human-preference checks pass. Remaining locales and the broader translation backlog are unfinished.
Thanks to xet7 !
Translate five cloud upload and fetch labels in 105 catalog files. Preserve existing values, source key order and exact placeholders; add checks distinguishing local-only upload and completed fetch messages.
Validation: all 106 preservation audits, four focused suites and 21 human-preference checks pass. Global catalog completeness and continuous-backup suites still report untranslated locales outside this batch.
Thanks to xet7 !
As decided on 2026-10-03:
- Encryption: with Encrypt on, the built-in engines seal what they store -
segments, page deltas, bases, blobs and state - with AES-256-GCM from
Node's crypto. The key is a file the administrator holds outside the
target and everything the stream reads, never in the database: 64 hex
characters, or a passphrase stretched with scrypt and the stream's salt.
stream.json keeps the salt and an HMAC check, so a wrong key is refused
on save and before a restore; blob names are keyed HMACs; a generation
records whether it is encrypted and is refused without its key.
- Cloud upload: the target is mirrored to S3/MinIO, Azure or GCS with the
Attachments storage panes' adapters and credentials. Sealed files go up
once, and an index or generation file again after the files it names, so
the remote copy never lists a segment it lacks; retention removals follow,
a failed upload is retried, and Fetch from cloud fills a lost target.
- SQLite restore on restart: a rebuilt, checked wekan.sqlite is staged in
<db>/continuous-restore with RESTORE_REQUESTED "continuous". The Docker
entrypoint, start-wekan.sh and the Snap keep the live database and its WAL
in continuous-restore/replaced first, then copy the staged one in, before
FerretDB opens its files. A pending request is never overwritten.
Tests: tests/continuousBackupEncryption.test.cjs, tests/continuousBackupCloud.test.cjs,
tests/continuousBackupRestartRestore.test.cjs (each script's real restore
block in a sandbox), encrypted integration tests on a real replica set and
WAL database, the server methods, and the pane in Chromium and WebKit.
Thanks to xet7 !
Translate both encryption labels in 105 catalog files. Preserve the AES-256-GCM algorithm name, 64-character hexadecimal key and minimum 16-character passphrase requirements, external key-file location and warning that restoration requires the key. Add focused token, key-format and warning regression coverage.
Validation: 106 batch preservation audits, three focused suites and 21 human-preference checks pass. Global completeness and backup suites still report untranslated remaining locales and newly added cloud labels.
Thanks to xet7 !
Record completed locale variants, passing batch checks and remaining translation-suite failures under Upcoming. Both changelog suites pass.
Thanks to xet7 !
Translate the next-restart timing and retention of the replaced database in 105 catalog files serving 106 locale variants. Preserve all existing values, exact source tokens and relative source key order.
Validation: all 106 batch checks and 21 human-preference checks pass. Board-link and language-wiring suites pass. Catalog completeness and continuous-backup suites still fail because remaining locales need the restart option and newly added encryption labels.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source key order and exact tokens. Extend restore safeguard, pause, time-unit and vocabulary regression checks. Technical Nahuatl wording is low confidence and needs native-speaker review.
Validation: four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
- A rule's move, sort, move-all or archive after it moved the card to
another board resolves on the destination board, in both rule engines;
then those actions become durable too.
- Closing a very large sprint, and undoing or redoing the close, run as a
resumable background job with progress; every per-card guard stays.
- Continuous backup gains cloud upload with the storage panes' credentials,
encryption at rest with an administrator-held key, and a restored SQLite
file staged for the next restart.
- #2509 is closed: changing the order of fields and hiding fields on the
card and the minicard already customizes how a card looks.
Thanks to xet7 !
Fixes#2509.
Fill 35 English placeholders and replace English fragments in the older backup description. Preserve other translations, literal paths, source key order and exact tokens. Extend restore safeguard, pause, time-unit and vocabulary regression checks. Technical Greenlandic wording is low confidence and needs native-speaker review.
Validation: four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source key order and exact tokens. Extend restore safeguard, pause, time-unit and vocabulary regression checks. Technical Fulah wording is low confidence and needs native-speaker review.
Validation: four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source key order and exact tokens. Extend restore safeguard, pause, time-unit, script and vocabulary regression checks. Technical Kashmiri wording is low confidence and needs native-speaker review.
Validation: four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Record the translation batch, restore-label correction, regression coverage and review limitations under Upcoming. Both changelog suites pass.
Thanks to xet7 !
Fill 35 English placeholders, replace mixed-language backup descriptions and correct the Esperanto delete label in the restore control. Preserve other translations, source tokens, literal paths and key order. Extend restore safeguard, pause, unit and vocabulary regression checks. Technical Volapük wording is low confidence and needs fluent-speaker review.
Validation: four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Record the backup translations, wrong-language corrections, full translation suite results and review limitations under Upcoming. Both changelog suites pass.
Thanks to xet7 !
Fill 35 English backup placeholders and replace 15 Venda, Zulu or mixed-language timeline, role-status and error entries with Veps. Preserve existing Veps entries, exact source tokens and key order. Extend safeguard, pause, unit, vocabulary and negative regression checks. Technical Veps wording is low confidence and needs native-speaker review.
Validation: all 453 translation-related suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders, replace English prose in two older backup descriptions and correct the backup label. Preserve other translations, literal paths and source tokens. Extend restore safeguard, pause, unit and vocabulary checks. Technical Klingon wording is low confidence and needs fluent-speaker review.
Validation: four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Link the translation batch from Upcoming and record regression coverage, low-confidence technical wording and the remaining backlog. Both changelog suites pass.
Thanks to xet7 !
Fill 35 English placeholders and correct two Romanian-seeded backup descriptions. Preserve source tokens, literal paths and other existing translations; extend backup safeguard, vocabulary and catalog checks. Technical Aromanian wording is low confidence and needs native-speaker review.
Validation: four focused translation suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct three Italian-seeded backup entries, restoring literal archive paths. Preserve other translations, source order and exact tokens. Extend restore protection, pause, path and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Ladin wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Record the verified translation batch and broad validation under Upcoming with its implementation commit link. Both changelog regression suites pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend vocabulary, script, restore protection, pause and time-unit checks. All 453 translation-related suites and 21 human-preference checks pass. Technical Tigrinya wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders and correct three misleading backup entries. Preserve other translations, source order, exact tokens and literal paths. Extend vocabulary, script, restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Sakha wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend vocabulary, script, restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Chuvash wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend vocabulary, script, restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Buryat wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders and replace three Tibetan-seeded backup entries. Preserve other translations, source order, exact tokens and literal paths. Extend vocabulary, script, restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Dzongkha wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend script, restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Tibetan wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Venda wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Northern Sami wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Guarani wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders and replace three English-filled backup entries, restoring literal paths and product names. Preserve other translations, source order and exact tokens. Extend restore protection, pause, path and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Quechua wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders and correct three English-filled backup entries. Preserve other translations, source order, exact tokens and literal paths. Extend restore protection, pause, path and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Aymara wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Bambara wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Acehnese wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders and correct three Italian-seeded backup entries, restoring literal archive paths. Preserve other translations, source order and exact tokens. Extend restore protection, pause, path and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Neapolitan wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Ewe wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Cornish wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit regression checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Walloon wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Manx wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders and correct three unrelated or mixed-language backup entries. Preserve other translations, source order, literal paths and tokens. Extend restore protection, pause and time-unit regression checks. Four focused catalog suites and all 21 human-preference checks pass. Technical Akan wording is low confidence and needs native-speaker review.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Technical wording is low confidence and needs native-speaker review. Five focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct two mixed-English backup descriptions, restoring literal paths. Preserve other translations and exact tokens. Extend restore protection, pause, time-unit and language regression checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and replace three English-filled backup strings, restoring the literal attachment path. Preserve other translations and exact tokens. Extend restore protection, pause, time-unit and language regression checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and vocabulary checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct two Polish-mixed backup descriptions, restoring the literal data path. Preserve other translations and exact tokens. Extend restore protection, pause, time-unit and language regression checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and replace a Walloon backup label in the Waray catalog. Preserve other translations, source order and exact tokens. Extend restore protection, pause, time-unit and vocabulary checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct two Spanish-mixed backup descriptions, restoring the literal avatar path. Preserve other translations and exact tokens. Extend restore protection, pause, time-unit and language regression checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and replace three Czech-seeded backup strings, restoring literal attachment and avatar paths. Preserve other translations and exact tokens. Extend restore protection, pause, time-unit and language regression checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and vocabulary checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and vocabulary checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and vocabulary checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Record the fills, corrections, low-confidence wording and successful 453-suite translation regression run under Upcoming. Both changelog checks pass.
Thanks to xet7 !
Fill 35 English placeholders and replace two Mandarin-seeded backup descriptions with Wu wording. Preserve other translations, literal paths and exact tokens. Extend restore protection, pause, time-unit, script and vocabulary checks. Technical wording is low confidence and needs native-speaker review. All 453 translation-related suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit, script and vocabulary checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Ethiopic script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 70 English placeholders, correct three Italian-seeded Friulian strings and restore literal paths in both catalogs. Preserve other existing translations and exact tokens. Extend restore protection, pause, time-unit and language regression checks. Technical wording in both languages is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Record the Romansh fills, corrections, regression coverage and low-confidence technical wording under Upcoming. Both changelog checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct an Italian backup label and mixed-language restore prompt. Preserve other existing translations and exact tokens. Extend restore protection, pause, time-unit and language regression checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Record the Hawaiian fills, corrections, regression coverage and low-confidence technical wording under Upcoming. Both changelog checks pass.
Thanks to xet7 !
Fill 35 English placeholders and replace two garbled backup descriptions, restoring literal paths and brands. Preserve other translations and exact tokens. Extend restore protection, pause, time-unit and literal preservation checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Record the Bislama fills, mixed-language corrections, regression coverage and low-confidence technical wording under Upcoming. Both changelog checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct three mixed-language backup strings, including the literal attachment path. Preserve other existing translations and source tokens. Extend restore protection, pause and time-unit regression checks. Technical wording is low confidence and needs native-speaker review. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct three mixed-language backup entries, restoring the literal attachment directory path. Preserve other translations, source order and tokens. Extend restore, time-unit and mixed-language checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Konkani vocabulary and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Maithili vocabulary and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Bhojpuri vocabulary and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Darija vocabulary and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Bashkir vocabulary and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and replace three Turkish-seeded backup entries with Tatar, restoring literal archive paths. Preserve other translations, source order and tokens. Extend restore, time-unit, script and vocabulary checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Yiddish vocabulary and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct the backup label and two Italian-seeded descriptions, restoring literal archive paths. Preserve other translations, source order and tokens. Extend restore, time-unit and language checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct two Italian-seeded descriptions, restoring literal archive paths. Preserve other translations, source order and tokens. Extend restore, time-unit and language checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct the backup label and two Italian-seeded descriptions, restoring literal archive paths. Preserve other translations, source order and tokens. Extend restore, time-unit and language checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 70 English placeholders across both Western Frisian catalogs while preserving existing translations, source order and tokens. Extend restore protection, pause and time-unit checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and replace three unusable backup entries, restoring literal archive paths. Xitsonga technical wording is low confidence and welcomes review. Extend restore, time-unit, placeholder and filler regression checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct two mixed-language backup descriptions, restoring literal archive paths. Preserve other existing translations, source order and tokens. Extend restore, pause, time-unit and mixed-language checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct two mixed English-Shona backup descriptions. Preserve other existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and mixed-language checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 70 English placeholders across both Zulu catalogs while preserving existing translations, source order and tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Odia-script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Assamese vocabulary and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 70 English placeholders across two catalogs serving three locale tags. Preserve existing translations, source order and tokens. Extend restore protection, pause, time-unit and Khmer-script coverage. Four focused suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and Myanmar-script checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit checks. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Extend restore protection, pause and time-unit coverage. Four focused suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders and correct four mixed English-Latin backup and database entries, including literal archive paths. Extend restore, time-unit and placeholder coverage. Latin technical wording is low confidence and welcomes review; consult the Latin computing glossary at https://ephemerisnuntii.eu/lexico/computatralia.pdf. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 70 English placeholders using existing Mandarin terminology and Cantonese prose. Preserve existing translations, source order and exact tokens. Extend restore protection, pause, time-unit and script checks. Four focused suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 105 English placeholders in Valencian, Slovenian (Slovenia) and Flemish using the corresponding existing language translations. Preserve existing translations, key order and tokens; extend restore and time-unit regression coverage. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Tajik vocabulary and script. Four focused catalog suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Kyrgyz vocabulary and script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 70 English placeholders across two catalogs while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Sinhala script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Gurmukhi script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Malayalam script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Kannada script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Gujarati script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Telugu script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Tamil script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Devanagari script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Nepali wording and script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Mongolian vocabulary and script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Kazakh vocabulary and script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders and replace Latin-script restore and save labels. Preserve existing correct-script translations, source order and exact tokens; cover restore warnings, time units and Arabic prose with unchanged technical names.
Low confidence: Arabic-script Uzbek orthography and technical phrasing need speaker review. Reference: https://en.wikipedia.org/wiki/Uzbek_alphabet and existing catalog vocabulary.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders across three catalogs while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Uzbek Latin vocabulary and script. Arabic-script Uzbek remains for a separate pass.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Armenian script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Georgian script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders across three catalogs while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Azerbaijani Latin vocabulary and script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Belarusian vocabulary and script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection, time units and Thai script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 35 English placeholders while preserving existing translations, source order and exact tokens. Cover restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 70 English placeholders and correct the Esperanto restore label, which previously meant remove. Preserve existing correct-language translations, source order and exact tokens. Cover restore warnings, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders and correct Macedonian restore and save labels. Preserve existing correct-language translations, source order and exact tokens. Cover restore warnings, time units and language-specific vocabulary and scripts.
Validation: all 453 translation-related suites and 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders across three catalogs while preserving existing translations, source order and exact tokens. Extend regression coverage for restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 175 English placeholders across five catalogs while preserving existing translations, source order and exact tokens. Extend regression coverage for restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders across three catalogs while preserving existing translations, source order and exact tokens. Extend regression coverage for restore boundaries, live database protection, time units and Devanagari and Bengali scripts.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders and correct the Urdu backup label. Preserve existing correct-language translations, source order and exact tokens. Extend regression coverage for restore warnings, live database protection, time units and Arabic script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 175 English placeholders across five catalogs while preserving existing translations, source order and exact tokens. Extend regression coverage for restore boundaries, live database protection, time units and Arabic and Hebrew scripts.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 175 English placeholders across five catalogs while preserving existing translations, source order and exact tokens. Extend regression coverage for restore boundaries, live database protection and time units.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 455 English placeholders across thirteen catalogs, including hiragana Japanese and separate Simplified and Traditional Chinese. Preserve existing translations and exact tokens; cover restore warnings, time units and locale scripts.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders across three catalogs while preserving existing translations, source order and exact tokens. Extend regression coverage for restore boundaries, live database protection, time units and Greek script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders while preserving existing translations, source order and exact tokens. Extend regression coverage for restore boundaries, live database protection, time units and Latin script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 175 English placeholders across five catalogs, including regional variants and the existing Russian locale alias. Preserve source order, tokens and existing translations; check restore warnings, time units and distinct Ukrainian and Russian vocabulary.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 140 English placeholders across four catalogs and replace Italian restore labels in both Romanian catalogs. Preserve source tokens and existing correct-language translations; check restore warnings, time units and Bulgarian script.
Validation: four focused translation suites and all 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders and correct the Slovak and Croatian restore labels. Preserve existing correct-language translations, key order and source tokens; cover restore boundaries, live-file protection, units and Latin-script labels.
Validation: all 453 translation-related suites and 21 human-preference checks passed.
Thanks to xet7 !
Fill 140 English placeholders across four catalogs while preserving existing translations and source key order. Cover source tokens, restore warnings, live database protection and time units.
Validation: four focused translation suites and 21 human-preference checks passed.
Thanks to xet7 !
Fill 105 English placeholders without replacing existing translations. Extend restore boundaries, live-file protection, units and exact-token checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 455 English placeholders across thirteen catalogs while preserving existing translations. Use Brazilian and European Portuguese wording and extend restore, live-file protection, unit and exact-token checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 315 placeholders across nine German and French catalogs without replacing existing translations. Extend restore boundaries, live-file protection, units, engine names, Swiss spelling and exact-token checks. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 70 English placeholders, preserving existing translations. Check restore boundaries, live-file protection, units, engine names and exact tokens. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Upcoming gains the sprint snapshot rows and the lifted card limit, the
continuous backup pane under a new Backup group, Focalboard import and
export, durable board-level activities and rule actions that follow their
card to another board, and the LinkedWriteBleed test fix. TODO Later now
says what stays open of each: slow but not refused very large sprints, the
cross-board moves that keep direct Sync and why, and what continuous backup
leaves to the administrator.
The CodeQL header of Upcoming is one line again, so the In short check
finds where the paragraph ends, and an LDAP summary no longer repeats its
group's name.
Thanks to xet7 !
Fill 70 English placeholders without overwriting existing translations. Check restore boundaries, live-file protection, units, engine names and exact tokens. Four focused suites and 21 human-preference checks pass.
Thanks to xet7 !
Add the 35 new source keys in English order without replacing existing translations. Completeness and email-recovery suites pass. Translation fills follow separately.
Thanks to xet7 !
Admin Panel / Attachments / Continuous backup, beside Backup, streams the
database, filesystem attachments, avatars and logs to a target directory
within seconds, so less work is lost between scheduled archives and a
restore can return to a chosen moment. Design: docs/Backup/Continuous-Backup.md.
The built-in engines use only copyfree code - WeKan's own (MIT), Node.js
and its node:sqlite (SQLite, public domain) - and start no other program:
- oplog: tails local.oplog.rs like Meteor's oplog driver, on a MongoDB
replica set or FerretDB with its oplog; a base archive (the existing
wekan-instance-backup format), then whole documents and deletions,
continued from the saved position after a restart, a new generation
when the oplog has lost it.
- sqlite: SQLite's online backup API on FerretDB's WAL database at an
interval, storing only the pages that changed.
- files: attachments, avatars and logs by watching and by rescans, each
content once under its SHA-256, links never followed.
Litestream (Apache-2.0) is an optional engine for SQLite: WeKan writes its
configuration and supervises a binary the administrator installed.
Segments are written whole, flushed and listed in append-only indexes.
A restore checks every index line, segment and blob first and writes
nothing when one is missing or changed; a database restore replays up to
the chosen moment, a file restore puts files back (add-missing or
replace-all), and a SQLite restore builds a checked file beside the stream,
never over the live one. Site administrators only; failures of the
unattended stream show in Admin Panel / Problems.
Tests: tests/continuousBackup.test.cjs; tests/integration/continuousBackup.test.cjs
against a real replica set, a real WAL database, real files and a stand-in
Litestream; server/lib/tests/continuousBackup.tests.js through the methods;
tests/playwright/specs/admin-continuous-backup.e2e.js on Chromium and WebKit.
Thanks to xet7 !
The LinkedWriteBleed test ran the module in a VM after removing only the
first `export`; the second exported function, editableCardIds, left an
`export` in place and the test failed to parse the file. It now removes
them all, and asserts what it did before.
Thanks to xet7 !
A sprint over 10,000 cards was refused at start and close. With the
snapshot rows already outside the sprint document, the rest now is too:
- The rollover plan of a close goes to its own collection,
scrumRolloverRows, in chunks bounded by rows and bytes; the sprint keeps
only rolloverPending: true. A chunk is applied, its cards read at once,
and removed; the mark is cleared when none is left. A plan kept in the
sprint document before this is still finished as it was.
- A Scrum History batch too large for one row is recorded as several rows
sharing a batchId, each at most 1,000 records and 3 MB a side. A record
is in exactly one row, and undo and redo walk the whole batch, with or
without a request ID, so one undo still reverses a whole close.
- Daily burndown observations keep their rows in scrumSnapshotRows too.
- No 10,000-card limit is left: snapshot inputs, daily capture, scope
replay, the daily report and the transfer read every card.
Undoing a batch re-hashed its whole History row and sent its whole plan as
a selector before every record write, so a 1,500-card undo took four
minutes. The row is now checked whole when the operation begins and by
its identity fields after that, and the worker check uses the plan's
identity and the worker's random id - nothing ever rewrites a pending
plan. A 10,500-card sprint closes in 8 s and undoes in 44 s.
Thanks to xet7 !
Fill the final 23 English email-recovery placeholders in Cherokee. Add all-locale missing, untranslated and exact-token regression checks alongside cancellation, uncertain-delivery and retained-pause checks. Four focused suites and 21 human-preference checks pass in the aligned worktree. Cherokee specialist wording and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders and replace two Tigrinya labels in the Tigre catalog. Extend cancellation, uncertain-delivery, retained-pause, vocabulary, script and exact-token checks. Four focused suites and 21 human-preference checks pass in the aligned worktree. Specialist Tigre phrasing, grammar and corrected labels have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders in syllabics, preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause, script and exact-token checks. Four focused suites and 21 human-preference checks pass in the aligned worktree. Specialist Inuktitut phrasing and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders and replace four mixed-English pause and email labels. Extend cancellation, uncertain-delivery, retained-pause and exact-token checks. Four focused suites and 21 human-preference checks pass in the aligned worktree. Specialist Wolaytta wording and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders in Tifinagh, preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause, script and exact-token checks. All 452 translation-related suites and 21 human-preference checks pass in the aligned worktree. Specialist Tamazight wording, grammar and regional terminology have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders, preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause and exact-token checks. Four focused suites and 21 human-preference checks pass. Specialist Nahuatl terminology, grammar and dialect choices have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders while preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause and exact-token checks. Four focused suites and 21 human-preference checks pass. Specialist Greenlandic wording and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders while preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause, script and exact-token checks. Four focused suites and 21 human-preference checks pass. Specialist Dzongkha wording and grammar have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders, preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause and exact-token checks; review recipient and message agreement. Four focused suites and 21 human-preference checks pass. Specialist Fulah wording, agreement and regional terminology have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders without replacing existing translations. Add cancellation, uncertain-delivery, retained-pause, script and vocabulary checks with exact token inventories. Four focused suites and 21 human-preference checks pass. Specialist Kashmiri wording and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders while preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause and exact-token checks. Four focused suites and 21 human-preference checks pass. Specialist Veps wording and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill 23 English placeholders and extend cancellation, retained-pause and exact-token regression checks. Four focused suites and 21 human-preference checks pass. Specialist Volapük wording and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill 46 English placeholders, preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause, script and vocabulary checks with exact token inventories. Four focused suites and 21 human-preference checks pass.
Specialist workflow wording and inflection in Tigrinya and Sakha have lower confidence and need speaker review.
Thanks to xet7 !
Fill 46 English placeholders and extend cancellation, uncertain-delivery, retained-pause and exact-token checks. Review Klingon word order, capitalization and retry wording. Four focused suites and 21 human-preference checks pass.
Specialist workflow wording and grammar in Guarani and Klingon have lower confidence and need speaker review.
Thanks to xet7 !
Fill 46 English placeholders and replace three language-prefix filler labels. Extend cancellation, uncertain-delivery, retained-pause and exact-token checks. Refine read and resume wording during review. Four focused suites and 21 human-preference checks pass.
Specialist wording, grammar and regional vocabulary in Aymara and Quechua have lower confidence and need speaker review.
Thanks to xet7 !
Fill 46 English placeholders and extend cancellation, uncertain-delivery, retained-pause and exact-token checks. Refine resume and form-field wording during review. Four focused suites and 21 human-preference checks pass.
Specialist workflow wording and grammar in Bambara and Ewe have lower confidence and need speaker review.
Thanks to xet7 !
A sprint's start and close snapshots kept one row per card inside the
sprint document, so a sprint over about 15,000 cards could not be closed.
The rows now go to their own collection, scrumSnapshotRows, in immutable
chunks of 2,000 named by sprint, kind, snapshot time and chunk number, so a
retried start or close, an undo or a redo finds the same rows. The sprint
keeps a header and the report's totals, computed from all rows.
The board data computes each sprint's report on the server and sends only
snapshot headers: from the stored totals, or for an assigned-only reader
from the rows they may see. The scope replay, the transfer export and the
board copy read the rows from wherever they are kept; the transfer import
writes them back as chunks through the import journal. Snapshots stored
inline before this keep working.
Thanks to xet7 !
Fill 46 English placeholders and repair seven Indonesian, mixed-language or generic email and pause labels. Cover cancellation, uncertain delivery, retained pauses and exact tokens. All 452 translation-related suites and 21 human-preference checks pass.
Specialist workflow phrasing and inflection in Acehnese and Akan have lower confidence and need speaker review.
Thanks to xet7 !
Fill 46 English placeholders and extend cancellation, uncertain-delivery, retained-pause, script and vocabulary checks with exact token inventories. Refine Wolof hold and restart wording during review. Four focused suites and 21 human-preference checks pass.
Specialist workflow phrasing and inflection in Tibetan and Wolof have lower confidence and need speaker review.
Thanks to xet7 !
Fill 46 English placeholders, preserving existing translated values. Extend cancellation, uncertain-delivery, retained-pause, Cyrillic and vocabulary checks with exact token inventories. Four focused suites and 21 human-preference checks pass.
Specialist workflow wording and inflection in Chuvash and Buryat have lower confidence and need speaker review.
Thanks to xet7 !
Fill 46 English placeholders and replace three Spanish or Italian Aromanian email labels. Extend cancellation, uncertain-delivery, retained-pause and exact-token checks; correct read and cancellation wording during review. Four focused suites and 21 human-preference checks pass.
Specialist phrasing and inflection in Aromanian and Northern Sami have lower confidence and need speaker review.
Thanks to xet7 !
The ordinary rule engine reads the card by id for every action, so the
actions after a move to another board act on the card there. The
durable commands that can do the same now find the card where a saved
move of the same plan put it, and record their writes, History rows and
activities on that board: card fields, checklist toggles, adding and
removing checklists, linking, and the title variables of a new card or
swimlane. A copy refuses, as the ordinary copy does. A move to another
board is no longer required to be the plan's last action. It is lifted
when every action that can follow it in a plan, in its own rule or in a
rule of the same trigger type, is one of these. Moves, sorting and
moving all cards are resolved against the board the card left, and
archiving cascades to child cards there, so with any of those the board
keeps direct Sync.
Thanks to xet7 !
Fill 46 English placeholders while preserving existing translations. Extend cancellation, uncertain-delivery, retained-pause, language vocabulary and token checks. Four focused suites and 21 human-preference checks pass.
Specialist workflow wording and inflection in Northern Ndebele and Waray have lower confidence and need speaker review.
Thanks to xet7 !
Add the new source key in English source order to all 243 remaining catalog files, preserving every existing value. The key remains a tracked placeholder for translation; English variants intentionally retain English. All 452 translation-related suites pass in this aligned worktree.
Thanks to xet7 !
Fill 46 English placeholders and correct two shared pause labels containing filler or Zulu. Cover cancellation boundaries, uncertain delivery, pauses and tokens. All 452 translation-related suites and 21 human-preference checks pass in the aligned worktree.
Specialist workflow phrasing and inflection in Tsonga and Venda have lower confidence and need speaker review.
Thanks to xet7 !
A Focalboard board.jsonl, from a .boardarchive zip, imports with or
without its version line: Focalboard's own importer reads the same
text. The board view's group-by select property becomes the lists.
Other select and multi-select properties become labels, and a property
called Labels gives labels by their own name. The first date property
becomes the due date, or start and due for a range. Text, number, email,
url, phone, checkbox and later dates become custom fields. Text and
heading blocks become the description, checkbox blocks a checklist, and
comment blocks comments. Members, person properties, images,
attachments and card templates are reported in the loss report. Export
writes the same text: a Status property for the lists, Labels, Due and
the custom fields. The field names follow Focalboard's model.Board and
model.Block.
tests/focalboard.test.cjs covers both board forms, the round trip and
the negatives. A Playwright case imports a board.jsonl through the page,
and the export audit fetches the new format. New instruction text is in
English, pending Transifex.
Thanks to xet7 !
Fill 46 English placeholders. Extend cancellation, retry-warning, pause and exact-token coverage, and correct Manx queue terminology during review. Four focused suites and 21 human-preference checks pass.
Technical phrasing and inflection in Hawaiian and Manx have lower confidence and need speaker review.
Thanks to xet7 !
Fill 46 English placeholders and extend cancellation, delivery-warning, pause and exact-token checks. Correct the Ladin retry interval wording during review. Four focused suites and 21 human-preference checks pass.
Specialist workflow wording and inflection in Walloon and Ladin have lower confidence and need speaker review.
Thanks to xet7 !
Durable delivery identified every activity by a card and a list, so a
rule's createSwimlane activity, which names neither, and a cross-board
move's moveCardBoard activity, which names no list, were written once
but delivered the ordinary way: their notifications and webhooks were
not replayed after a crash. Durable delivery now takes activities
without a list, and without a card. Recipients and webhook senders are
checked against the board, plus the card when one is named. A card or
list the activity does not name is refused. An assigned-only member
never receives a board-level event, since its scope cannot be checked
against their assignments. The rule stage has nothing to run for these
types, because no trigger exists for them, and it refuses a list-less
activity whose type has one.
Thanks to xet7 !
Fill 46 English placeholders and cover cancellation boundaries, uncertain delivery, preserved pauses and exact tokens. Four focused suites and 21 human-preference checks pass.
Specialist workflow phrasing and inflection in both languages have lower confidence and need speaker review.
Thanks to xet7 !
Fill 92 English placeholders in Māori, Samoan, Tongan and Fijian, and replace the Tongan pause filler. Cover cancellation boundaries, retry warnings, source order and exact tokens. All 452 translation-related suites and 21 human-preference checks pass.
Specialist workflow phrasing and inflection in these four languages have lower confidence and need speaker review.
Thanks to xet7 !
Fill 92 English placeholders and extend cancellation, uncertain-delivery and retained-pause regression checks. Four focused suites and 21 human-preference checks pass. Specialist workflow phrasing and inflection have lower confidence and need speaker review.
Thanks to xet7 !
Fill 92 English placeholders and extend cancellation, uncertain-delivery and retained-pause regression checks. Four focused suites and 21 human-preference checks pass. Specialist workflow phrasing and inflection have lower confidence and need speaker review.
Thanks to xet7 !
Fill 92 English placeholders across four locale tags, including regional Zulu. Extend cancellation, uncertain-delivery and retained-pause regression checks. Four focused suites and 21 human-preference checks pass. Specialist workflow phrasing and inflection have lower confidence and need speaker review.
Thanks to xet7 !
Fill 92 English placeholders and extend script, cancellation and retry-warning regression checks. Preserve existing translations and pauses during retry. All 452 translation-related suites and 21 human-preference checks pass. Specialist workflow phrasing and inflection have lower confidence and need speaker review.
Thanks to xet7 !
Fill 92 English placeholders with native-script and warning-boundary regression checks. Preserve existing translated values and pauses during retry. Four focused suites and 21 human-preference checks pass. Specialist workflow phrasing and inflection have lower confidence and need speaker review.
Thanks to xet7 !
Fill 92 English placeholders while preserving existing translated values. Extend cancellation, uncertain delivery and retained pause regression checks. Four focused suites and all 21 human-preference checks pass. Specialist terminology and long workflow clauses have lower confidence and need speaker review.
Thanks to xet7 !
Fill 92 English placeholders in Kurmanji, Sorani, Tatar and Turkmen. Preserve cancellation boundaries, uncertain-delivery warnings and existing pauses during retry. Four focused suites and 21 human-preference checks pass. Specialist workflow wording and inflection have lower confidence and need speaker review.
Thanks to xet7 !
Fill 48 English placeholders and require feature-visibility coverage for every non-English catalog. Preserve tokens, source order and existing translated values. Tigre and Cherokee specialist wording and inflection have low confidence and need speaker review. All 451 translation-related suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill 24 English placeholders and cover data retention, unchanged permissions, future approval and Tifinagh prose. Specialist analytics terminology and inflection have low confidence and need speaker review. Four focused suites and all 21 human-preference checks pass.
Thanks to xet7 !
Fill twenty English placeholders and extend placeholder, script and workflow wording checks. Cherokee specialist phrasing and inflection have low confidence and need speaker review. All 451 translation-related suites and 21 human-preference checks pass.
Thanks to xet7 !
Fill twenty English placeholders and extend token, syllabics and meaning regression checks. Inuktitut specialist workflow phrasing and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill twenty Standard Moroccan Tamazight placeholders, replace the Arabic Discard label, and preserve established Tigre card terminology. Specialist Tamazight phrasing and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill twenty English placeholders and extend token, meaning and locale-prefix regression checks. Wolaytta specialist terminology and inflection have low confidence and need speaker review. Older prefixed English values remain for further audit.
Thanks to xet7 !
Fill twenty English placeholders and extend token, script and selected Tigre wording checks. Tigre technical terminology and inflection have low confidence and need speaker review. Older mixed-language values remain for further audit.
Thanks to xet7 !
Fill twenty English placeholders and extend placeholder and meaning regression coverage. Nahuatl specialist workflow phrasing and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill twenty English placeholders and extend token and meaning regression coverage. Greenlandic workflow wording and inflection have low confidence and need speaker review.
Thanks to xet7 !
Fill 20 English placeholders with dictionary-assisted Veps wording. Preserve existing translations, key order and token inventories. Extend incomplete-history, missing-weight and discard-only recovery checks.
Veps workflow phrasing and inflection have low confidence and need speaker review. Vocabulary references include the Veps-Hungarian dictionary and the New Russian-Veps Dictionary entries reproduced in Wiktionary.
Thanks to xet7 !
Fill 40 English placeholders, bringing scope history, GitLab estimates, paging and recovery to 226 locale tags. Preserve existing translations, key order and exact tokens. Extend warning checks for incomplete history, missing weights and discard-only recovery.
Use Aromanian preservation vocabulary for saved state instead of the tell/show root. Ladin recovery expresses possibility. Specialist wording in both languages has low confidence and needs speaker review.
Thanks to xet7 !
Fill 80 English placeholders in Acehnese, Aymara, Guarani and Quechua. Scope history, GitLab estimates, paging and recovery now cover 224 locale tags. Preserve existing translations, source order and placeholder inventories.
Add warning checks for incomplete history, missing weights and discard-only recovery. Aymara recovery is available rather than mandatory. Reject English-prefix filler in the new Aymara and Quechua scope messages.
Specialist wording in all four languages has low confidence and needs speaker review. Older wrongly seeded strings elsewhere in the catalogs remain outside this batch.
Thanks to xet7 !
Fill 80 English placeholders in Tibetan, Dzongkha, Tigrinya and Kashmiri,
bringing scope history, GitLab estimates, paging and recovery to 220 locale
tags. Preserve correct-language translations, source order and tokens.
Extend native-script and semantic checks for incomplete history,
discard-only recovery and clearing missing estimate weights. Keep
Dzongkha wording distinct from Tibetan, and Kashmiri from Urdu.
Specialist workflow wording in all four languages has low confidence
and needs native-speaker review.
Thanks to xet7 !
Fill 52 English placeholders, including Northern Sami paging and recovery. Preserve existing translations, source key order and exact tokens. Extend incomplete-history and missing-weight regression checks.
Specialist workflow wording in Ewe, Fulah and Northern Sami has low confidence and needs native-speaker review. References include Basic Ewe, the Peace Corps Fulfulde technical manual, and Northern Sami dictionaries and usage.
Thanks to xet7 !
Fill 36 English placeholders across Akan, Bambara, Wolof, Arabic-script
Uzbek and Klingon. Paging and recovery now cover 215 locale tags;
scope history and GitLab estimates cover 213.
Preserve existing translations, key order and exact placeholder tokens.
Extend script and meaning checks for recovery, incomplete history and
clearing missing GitLab weights. Review Klingon suffix combinations.
Specialist wording has low confidence in all five languages, especially
Klingon estimate terminology, and needs native-speaker review.
Thanks to xet7 !
Fill 140 English placeholders while preserving existing translations. Paging and import recovery now cover 210 locale tags. Scope history and GitLab estimates gain Manx, Cornish and Volapuk, reaching 212 tags.
Extend native vocabulary, script, placeholder and warning-meaning checks. Use the Sakha storage term rather than confirmation, distinguish finish from cancel, and retain the no-weight-clears-value warning.
Workflow wording in all 23 locales has lower confidence and needs speaker review: Bashkir, Buryat, Chuvash, Sakha, Upper Sorbian, Silesian, Venetian, Walloon, Manx, Cornish, Volapuk, Ndebele, Swati, Sesotho, Tswana, Northern Sotho, Tsonga, Venda, Kirundi, Oromo, Luganda, Ewe and Fulah. Dictionary references are documented in Upcoming.
All 451 translation-related suites pass before the final Ewe and Fulah fills; six catalog and batch suites pass with those included. Human-preference verification passes all 21 checks.
Thanks to xet7 !
Fill 280 English placeholders across 70 stored catalogs, preserving existing translations. The four messages now cover 187 locale tags, with 47 remaining. Extend native-script checks and guard the distinct finish/discard actions, exact count tokens and unchanged-board warning.
Use native restart phrasing in Romansh and Faroese and distinguish finishing from cancelling in Hawaiian. Specialist wording has lower confidence in Latin, Luxembourgish, Maltese, Turkmen, Tatar, Amharic, Shona, Chichewa, Kinyarwanda, Igbo, Yoruba, Malagasy, Papiamento, Faroese, Frisian, Romansh, Friulian, Yiddish, Bhojpuri, Maithili, Konkani, Maori, Samoan, Fijian, Tongan, Hawaiian and Waray; speaker review remains welcome.
All 451 translation-related Node suites and 21 human-preference checks pass.
Thanks to xet7 !
Record the resolved completion-suite failure, 412 new translations, current coverage, validation results and lower-confidence wording.
Thanks to xet7 !
Fill 412 English placeholders without replacing existing translations. The four messages now cover 116 locale tags. Keep finishing a stopped import distinct from discarding an incomplete plan, preserve the paging count token, and add script and warning-meaning regression coverage.
Resolve the remaining completion-suite failure by translating the missing Scrum messages. All 451 translation-related suites pass at 52 locale tags; four catalog and batch suites pass after extending to 116. Human-preference verification passes all 21 checks.
Specialist wording in Irish, Scottish Gaelic, Breton, Kashubian, Corsican, Sardinian, Sicilian, Neapolitan and Aragonese has lower confidence and remains open to speaker review. Use the documented Occitan reaviada rather than French restart terminology.
Thanks to xet7 !
Link the 13-locale Scrum and GitLab translation batch, the new Scrum catalog synchronization and additional translations. Record validation results, dictionary references, lower-confidence wording and the remaining completion-suite failure.
Thanks to xet7 !
Fill 208 English placeholders across Akan, Moroccan Arabic, Bashkir, Bambara, Buryat, Chuvash, Upper Sorbian, Sakha, Silesian, Venetian, Walloon, Wolof and Wu Chinese. Both message groups now cover 209 locale tags. Preserve existing translations and extend native vocabulary, script and placeholder regression checks.
Synchronize four pending Scrum source keys across locale catalogs and translate 48 newly added values in the Czech, Hungarian, Russian, Slovak, Ukrainian, Estonian and Hebrew locale families. Specialist workflow wording in the 13-language batch has lower confidence and needs speaker review.
Validation: catalog and batch suites pass. Of 450 translation-related suites, seven initially failed on the new Scrum text; six pass after translation. The broad Vietnamese/Bulgarian completion suite still requires the new messages in further languages, first Persian.
Thanks to xet7 !
An interrupted Scrum import could only be continued by the offline
recovery command, with every writer stopped, because nothing told a
crashed import from a paused one. The import's checkpoint now names its
writer and carries a lease the writer renews at each step. Once the
lease runs out, a board administrator can finish the import from the
Scrum view: the recovery takes the checkpoint over by compare-and-set
and replays the saved steps. A writer that was only paused is fenced out
at its next checkpoint update, and the steps are idempotent. An import
whose plan was never fully saved can be discarded, which writes nothing
to the board. Its contents cannot be rebuilt, because the source file is
not kept. Rollback stays offline. Checkpoints from before this change
are taken over at once, since their writer did not survive a restart.
Thanks to xet7 !
Opening a board's Scrum view checked every card's edit permission one
at a time, querying each card's links and then their boards. One batch
now answers the same for all of the board's cards, and a server test
confirms it agrees with the per-card check for writers, read-only and
comment-only members, and outsiders with links. The Product Backlog and
Sprints tables, whose rows each carry an edit form, show 100 rows at a
time in Scrum order, with a button to show more. New interface text is
in English, pending Transifex.
Thanks to xet7 !
As decided on 2026-10-02, a card or swimlane copied or moved to another
board now takes that board's sprint and release of the same name when
exactly one matches: a planned or active sprint, or a release that is
not cancelled. Otherwise they are dropped, as before. Past sprints and
backlog rank always go. This applies to card and swimlane copies,
subtasks copied with their parent, the server hooks behind every move,
and the durable rule copy and move. A move that sets the Scrum revision
itself, as the durable rule move does, is left alone by the hook.
Thanks to xet7 !
Fill 400 English placeholders, bringing both groups to 196 locale tags. Preserve existing translations, catalog order and placeholders. Add native-script and warning-meaning coverage, and use the documented Fijian estimate term instead of the verb for compare.
All 450 translation-related suites pass at 192 tags; five catalog and translation suites pass with the final four tags and wording corrections.
Specialist workflow wording in Bislama, Tok Pisin, Maori, Samoan, Fijian, Tongan, Hawaiian, Waray, Ndebele, Swati, Sesotho, Tswana, Northern Sotho, Tsonga, Venda, Kirundi, Oromo, Luganda, Bhojpuri, Maithili and Konkani has lower confidence and remains open to speaker review.
Thanks to xet7 !
Fill 688 stored English placeholders, bringing both groups to 171 locales. Preserve existing translations, current key order and token inventories. Extend native-script checks and distinct missing-history, truncated-replay and missing-weight warning coverage.
All 450 translation-related suites pass with the first 36 added tags; five catalog and translation suites pass with the final eight included.
Specialist workflow wording in Assamese, Odia, Sinhala, Burmese, Khmer, Javanese, Amharic, Malagasy, Hausa, Somali, Shona, Chichewa, Kinyarwanda, Igbo, Yoruba, Kurdish, Arabic-script Uzbek, Turkmen, Faroese, Frisian, Papiamento, Friulian and Romansh has lower confidence and remains open to speaker review.
Thanks to xet7 !
Fill 1200 English placeholders, bringing both message groups to 127 locale tags. Preserve existing translations and token inventories. Add scope-warning and change-cause regression checks and extend missing-weight semantics checks.
All 450 translation-related suites pass with the first 56 locales; five catalog and translation suites pass with the remaining 19. Specialist wording in Luxembourgish, Maltese and Latin has lower confidence and remains open to speaker review.
Thanks to xet7 !
Bind the scope-report exception to its chart markup and read-only Scrum method. Continue rejecting History tables, browser calls and edit handlers. The full 450-suite translation-related run and negative mutation checks pass.
Thanks to xet7 !
A card or swimlane moved to another board kept the sprint, past
sprints, release and backlog rank of the board it left, pointing at
that board's records. A copy there already dropped them. Moves now do
the same, keeping the issue type, acceptance criteria and swimlane
purpose. Server hooks do it, so client, REST and rule moves are all
covered. Scrum fields are not the client's to write, so the
client-callable move does not write them itself. The durable
cross-board rule move writes the same result.
Thanks to xet7 !
Fill 561 stored English placeholders for scope history, replay warnings and change causes. Preserve existing translations, catalog order and placeholder inventories. The seven affected language suites pass.
Specialist wording in Corsican, Sardinian, Sicilian, Neapolitan, Aragonese, Asturian, Breton and Kashubian has lower confidence and remains open to speaker review.
Thanks to xet7 !
Finish the three card-setting labels in all 234 non-English locale tags and remove them from the pending queue. Extend feature messages to 231 tags, and translate five new GitLab estimate messages in 52 tags. Synchronize the newly committed GitLab and Scrum keys across catalogs.
Correct mixed-language Wolaytta labels and the Tigre linked-card relative clause, with exact-value repair history and script/token regressions. Specialist wording in the six final card-label languages and less widely used estimate locales has lower confidence and welcomes speaker review.
Six current structural and translation suites pass. The earlier eight failures pass after corrections in a review snapshot of the GitLab source; the newly added Scrum group remains pending translation.
Thanks to xet7.
Sprint Report gains a section that replays every recorded change since
the sprint started: cards joining or leaving it, estimate changes,
completion, moves to done lists, and archiving. It shows scope, remaining
and completed work after each change; the daily observations sample the
same sprint only once a day. Each card's state at the start is
reconstructed backwards from its current state through those History
rows. The result is checked against the sprint's own start and close
snapshots and, for an open sprint, against the cards as they are now.
Any mismatch means History missed a write, and the report says so
instead of presenting the numbers as fact. Assigned-only readers see
their own cards, marked partial. New interface text is in English,
pending Transifex.
Thanks to xet7 !
List Sync from GitLab can now carry an estimate, as Jira's already
could. The Sync settings choose GitLab's weight (in points) or its time
estimate (given in seconds, stored in hours), and any numeric custom
field on the board to hold it. An issue without a weight clears the
value; local changes are reviewed as conflicts, as for Jira. The
durable journal accepts the GitLab mapping, and the source coverage
report counts the mapped attribute as used. GitHub, Gitea and Forgejo
issues have no estimate to sync. New interface text is in English,
pending Transifex.
Thanks to xet7 !
The Jira import page now offers the numeric custom fields that the
pasted export declares in its schema, story points first, as
suggestions for the estimate field. When no field is entered and
exactly one field has Jira Software's story points type, that field is
used, in points. It is identified by its type, never by a name. With
two such fields, or none, nothing is chosen.
Thanks to xet7 !
Fill 135 messages in Aymara, Quechua, Guarani, Fulah and Greenlandic. Replace mixed-language Aymara and Quechua feature, administrator and linked-card labels, and correct the Guarani board label and creation message.
Settings coverage reaches 228 locale tags. All 448 related suites pass after the first three languages; eight selected suites pass after Fulah and Greenlandic. Specialist workflow wording in all five languages has lower confidence and welcomes speaker review.
Thanks to xet7.
The Sprint, Rank and Epic Link fields are found by their schema type, or
by name when the export has no schema, never by customfield number.
Future and active sprints become planned WeKan sprints with their goals
and dates. An active sprint is reported, because the export has no
commitment snapshot. Closed sprints are reported, not invented. Fix
versions become releases. Jira's rank orders the backlog, and a legacy
Epic Link becomes the card's parent. The data is written as a native
Scrum import, and these fields are no longer also imported as text
custom fields.
Thanks to xet7 !
A moveAllCardsInList rule onto another board now saves one unit per
card of the list. Each unit is a whole move to another board, with the
same mappings and records as a single rule move there. As in the
ordinary action, each card keeps its sort and lands in that board's
default swimlane. Like a single move to another board, it must end its
rule plan, because the rule's own card may be among those moved. The
rule guard follows a card to where its unit put it. The single-move and
move-all runners share one apply step.
Thanks to xet7 !
Fill 81 settings messages and correct the Volapük time and board labels. Extend language, permission wording and placeholder regressions to 223 locale tags.
Specialist workflow wording in all three languages has low confidence and welcomes speaker review. Selected catalog and language suites pass.
Thanks to xet7.
A moveCardToTop or moveCardToBottom rule onto another board now runs its
saved command. The card is updated once, conditionally, with every hook
record deferred, and each record is written from the command: History
rows, the legacy undo row, re-homed checklists, the dependencies left
behind on the old board, attachment placement, re-pointed label
activities, the moveCardBoard activity and the custom-field activities.
The server test compares the result with the ordinary Card.move.
Every other durable command acts on the card on the plan's board, so
such a move must end its rule plan. The command refuses any other
position before writing anything. Eligibility counts the move only when
it is the last action of its rule and no other rule on that board fires
on the same activity type; the destination must have opted in too.
The rule guard follows the card to the board this plan moved it to. The
activity's notification and webhook stages judge recipients on the card
as placed for the activity. Before, they refused a card that a rule had
moved, which would have failed any run whose moved card had a watcher
or a webhook, also for moves on the same board.
Moving all cards of a list to another board stays on direct Sync.
Thanks to xet7 !
Fill 393 stored placeholders for card color, linked card and description badge. Correct mixed-language linked-card labels and extend token and locale regressions to 220 tags.
Specialist badge wording in Corsican, Latin, Shona, Hausa, Igbo and Yoruba has lower confidence and welcomes native-speaker review. All 448 translation-related Node suites pass.
Thanks to xet7.
Add 54 messages using each language’s existing card vocabulary. Verify script, data-retention wording, current keys and placeholders. Specialist workflow wording has lower confidence and remains open to native review.
All six selected catalog and language suites pass.
Thanks to xet7.
The saved command for a rule move onto another board is written
(server/lib/syncRuleMoveBoardCommand.js). It plans what Card.move and
its hooks write there: labels by name, the card number, mapped custom
fields, filtered members and watchers, cleared dependencies, the
History rows, the moveCardBoard and custom-field activities, the
re-pointed label activities and the legacy undo row. It accepts only
the last action of its rule plan. The History planner gains those
fields, and the recording scope can defer the dependencies row and the
label-activity hook, which runs while the card is still on its old
board. Nothing uses the command yet: such a rule still keeps its board
on direct Sync until its runner, guard and eligibility are added.
Thanks to xet7 !
Add 81 messages. Feature settings now cover 218 locale tags and card-setting labels cover 86. Specialist workflow wording, especially Veps, has lower confidence and remains open to native review.
All 448 translation-related suites pass after Northern Sami and Veps; all 11 selected suites pass with Tibetan included.
Thanks to xet7.
The before-update hook that re-points a moved card's addedLabel
activities at the new board's labels of the same name ran twice per
update, once in the members hook and once in the assignees hook. The
second run saw label ids the card had not had before the move and
deleted every activity the first run had re-pointed. It now runs once.
Thanks to xet7 !
A Sync step's rules run after its card write, and since rules may move a
card to another list or archive it, the card can afterwards match
neither the step's before-state nor its after-state. Two things then
failed:
- An interrupted run never recovered. The replay retried the step's
write, which could not be confirmed, so the run failed on every replay.
The journal now treats a stored planned activity as proof that the
write happened (activities are stored only after it is confirmed) and
finishes the effects, which are idempotent.
- Even an uninterrupted run failed whenever a durable rule moved the card
out of the list. The activity's notification and webhook stages ran
after the rule and refused a card that was no longer in the activity's
list. They now also accept the card where a saved move of that
activity's own rules put it, and nowhere else.
A failed replay now logs why. The durable Sync server test runs a rule
that moves each new card out of the synced list, crashes after the move,
and checks that the replay completes the run with one move.
Thanks to xet7 !
Add 108 messages in Konkani, Moroccan Arabic, Acehnese and Waray. Replace French board and card labels in Waray and Malay linked-card wording in Acehnese. Specialist workflow descriptions, particularly Acehnese, have lower confidence and remain open to native review.
All six selected catalog and language suites pass.
Thanks to xet7.
A copyCard rule that copies onto another board now uses the same saved
command, with that board as its target: the copy takes that board's
labels of the same name, its custom fields as Card.copy maps them, its
card number and the dependencies that stay on it. Its checklists, items,
subtasks, comments and attachments are all placed on that board. As
with links, eligibility requires the destination to have opted in and
to have only durable rules of its own, and the runner checks it again.
The server test compares the result with the ordinary Card.copy to the
same board.
Thanks to xet7 !
Add 135 messages in Buryat, Chuvash, Sakha, Bhojpuri and Maithili. Align the Buryat linked-card label with its existing card terminology. Specialist workflow wording, particularly Buryat, Chuvash and Sakha, has lower confidence and remains open to native review.
All eight selected translation and catalog suites pass.
Thanks to xet7.
As decided on 2026-10-02, a linkCard rule that links onto another board
keeps durable Sync only when that board has Sync effects enabled too and
the actor may write there. Eligibility now follows every board such a
rule reaches, because the link's creation activity runs that board's own
rules through the same stored stages; any action there without a durable
adapter keeps the source board on direct Sync. The runner checks the
destination again before each write.
A card-field rule acting on a linked card now writes the card it links
to, as the ordinary setters do through getRealId. Before, the durable
command refused a linked card, so a durable link followed by such a rule
would have failed its Sync run on every replay.
Thanks to xet7 !
Add 135 messages in Ladin, Aromanian, Cantonese, Wu Chinese and Yiddish. Correct Ladin linked-card terminology and Cantonese and Wu Save labels. Update the README completion count to the measured 165 languages.
Specialist workflow terminology, particularly Ladin and Aromanian, has lower confidence and remains open to native review. All nine selected catalog and language suites and three additional Chinese suites pass.
Thanks to xet7.
Add 108 messages and correct malformed Hawaiian labels and Italian-seeded Venetian linked-card text. Specialist workflow terminology has lower confidence and remains open to native review.
All 448 translation-related suites pass after the Hawaiian and Fijian changes; all 10 selected suites pass with Venetian and Walloon included.
Thanks to xet7.
The minicard's checklist due date and the awaiting-approval note use
margin-inline-start rather than margin-left, so they follow the language
direction. The issue-type guard now follows the badge into the ordered
minicard badges, where it sits behind its own board toggle that defaults
on. The Taskwarrior format's documentation link is allowlisted for that
file in the release source audit, as Leo's namespace URL is.
Thanks to xet7 !
TODO Later records the maintainer decisions of 2026-10-02 on the rule
guard and cross-board effects, and keeps only cross-board rule effects
and board-level activity delivery as remaining.
Thanks to xet7 !
The rule stage's guard now accepts the card where a saved move of the
SAME rule plan put it (maintainer decision of 2026-10-02), so the rule's
later actions act on the moved card once. Each move is a saved command
with Card.move's effects: the update, the position History row, the
legacy undo row, the attachments' placement and the moveCard activity.
moveAllCardsInList saves one unit per card. Moves, copies and links to
another board stay on the direct path.
Thanks to xet7 !
Add 135 messages in Bislama, Tok Pisin, Māori, Samoan and Tongan. Replace mixed-English feature headings and linked-card labels. Specialist workflow terminology has lower confidence and remains open to native review.
All eight selected catalog and language suites pass.
Thanks to xet7.
Add 162 messages in Breton, Cornish, Manx, Kashubian, Upper Sorbian and Silesian. Correct Breton linked-card meaning and replace Czech-seeded Upper Sorbian labels. Specialist flow-metric wording has lower confidence and remains open to native review.
All 12 selected translation and catalog suites pass.
Thanks to xet7.
Add 189 messages in Akan, Bambara, Ewe, Luganda, Tsonga, Venda and Wolof. Correct mixed-language feature headings and linked-card labels, and replace eight Zulu labels in Venda.
Specialist flow-metric wording has lower confidence and remains open to native review. All 19 selected catalog and language regression suites pass.
Thanks to xet7.
Fill 135 messages in Sesotho, Setswana, Northern Sotho, Swati and Northern Ndebele. Correct the mixed English/Setswana feature heading. Specialist flow-metric terminology has lower confidence and remains open to native review.
Verify current catalog keys, placeholders, native vocabulary and linked-card terminology: all 448 translation-related Node suites pass.
Thanks to xet7.
Everything the LDAP module logged about what it sends and hears - the
client and TLS options, the search options and base DN, the group
filters, why an authentication failed - went to Meteor's Log.debug,
which never prints when Meteor.isProduction. The snap and every bundle
run in production, so turning on ldap-log-enabled showed none of it,
and the search options printed as "[object Object]" in any case. They
now go through the LDAP logger, which prints when LDAP_LOG_ENABLED is
true and redacts secrets, as objects. A failed bind now logs the DN and
the directory's own answer (Active Directory's "data 52e" and the like),
never the password, before it is rethrown.
Thanks to rholighaus and xet7 !
Fixes#6548.
Add 81 messages across three locale tags and preserve the distinction between Zulu and Xhosa vocabulary. Specialist flow-metric wording has lower confidence and remains open to native review.
snapcraft has no per-app ulimit key and snapd owns the systemd unit, but
a process may always raise its own soft open-file limit up to the hard
one without privilege. The snap's Caddy service now does that before it
starts Caddy, every time, which is what makes it persist across
restarts and refreshes. CADDY_NOFILE may ask for a lower number; it is
capped at the hard limit, a non-number is ignored, and a failure to
raise is logged without stopping Caddy.
tests/snapCaddyNofile6552.test.cjs runs the helper in bash with a
lowered limit. It has not been run inside an installed snap here.
Thanks to xet7 !
Fixes#6552.
An addSwimlane rule now has a saved command: the title is resolved once
by RulesHelper.ruleSwimlaneTitle, the swimlane is inserted once under an
id derived from the rule invocation, and the insert hook's
createSwimlane activity is written once, by a derived id, by the command
itself. A swimlane activity has no card, and WeKan's durable activity
delivery is keyed by a card, so that activity is delivered the way every
board-level activity is; what the command guarantees is that the
swimlane and its activity each exist exactly once across replays.
The stored-rule tests now use moveAllCardsInList as their example of an
action with no durable adapter.
Thanks to xet7 !
A linkCard rule that links onto the card's own board now has a saved
command: the linked card is built once as Card.link builds it (a copy
without id and labels, of type cardType-linkedCard pointing at the
original), under an id derived from the rule invocation so a replay
inserts it once, and its creation activity is saved and delivered
durably. The target is resolved by RulesHelper.linkCardTarget, which the
ordinary action now uses too. A link to another board keeps the board
on direct Sync.
Thanks to xet7 !
Feature settings now cover 171 locales and card labels cover 39. Correct Sicilian and Papiamentu linked-card vocabulary and keep Rundi distinct from Kinyarwanda. Specialist flow terminology in these ten languages has lower confidence and remains open to native review.
A copyCard rule that copies onto the card's own board now has a saved
command. What the copy contains is decided once, with the ordinary
copy's own helpers: the card as Card.copy builds it (the custom fields
the actor may read, its dependencies on the board, its Scrum metadata,
a new card number, the bottom of the list), its live attachments, its
checklists and items, its subtasks with their checklists, and its
comments. Every document gets an id derived from the rule invocation,
so a replay inserts each once. The copy's and its subtasks' creation
activities are saved and delivered durably; each attachment file is
copied once (copyFile can now be given the id) and its History row is
written from the stored attachment; checklists, items and comments are
direct inserts, as Card.copy makes them. The cover is remapped to the
copied attachment by its derived id. Like the ordinary copy chain
guard, a copy whose own creation triggers the same action is not copied
again. A copy to another board keeps the board on direct Sync.
The server test copies a card with all of those, replays it, checks the
chain guard, and compares the result with the ordinary Card.copy.
Thanks to xet7 !
A rebase rewrote the commits these entries link to, so the links named commits
that no longer exist and would have 404ed on GitHub. Repointed by
releases/fix-changelog-hashes.sh, which only touches a link no ref in this clone
can reach.
Thanks to xet7 !
Reuse established linked-card terms, remove the Odia separator artifact, and correct Italian wording in Friulian and Romansh. Add regressions for the corrected vocabulary.
Upcoming lists the three newly durable rule actions, and TODO Later
counts twenty-seven durable action types, with what is left and why:
moving all cards of a list waits on durable moves between lists,
linking a card is blocked on a decision about cross-board effects of a
Sync run, and copying a card is large, separate work.
Thanks to xet7 !
Feature coverage reaches 161 locale tags, preserving their existing translations. Add native-script and data-retention checks. Specialist flow-metric wording in this batch, especially Faroese, Frisian, Friulian, Romansh and Odia, has lower confidence and remains open to native review.
A createCard rule now has a saved command. Its list, swimlane and title
are resolved once by RulesHelper.createCardTarget - which the ordinary
action uses too, so the two cannot drift - and the new card's id is
derived from the rule invocation, so a replay inserts it once. The
creation hook's createCard activity is known before the insert and is
saved with the command, then delivered durably, which runs the new
card's own rules. A target that does not exist is refused, as the
ordinary insert fails at its activity.
Thanks to xet7 !
A sortList rule re-sorts the cards of the rule card's list, or a named
list on the same board, and gives each card its index as its sort; the
hook records one position History row per changed card and there is no
activity. It now has a saved command: the order is decided once, by the
same sort keys the ordinary action uses (moved to the pure
models/lib/ruleSortList.js so the two cannot drift), and each changed
card is a unit written conditionally on its saved sort with its planned
position row. Replays change nothing more.
Thanks to xet7 !
The checklist lifecycle command now also covers addChecklistWithItems:
the checklist and one item per comma-separated title, both with ids
derived from the rule invocation so a replay inserts them once. After
the inserts it records, from the stored documents and in the order the
hooks write them, the checklist's activity and History row and then each
item's addChecklistItem activity and 'added' row. The item hooks are
deferred only for items of the checklist the command names. The item
titles are resolved by the same RulesHelper call the ordinary action
uses.
Thanks to xet7 !
In-place rule moves and checklist creation and removal are durable now.
Moves out of the card's list wait on a maintainer decision about the
rule stage's guard, which binds the card to the triggering activity's
list; adding a swimlane waits on the durable activity pipeline learning
board-level activities, which have no card; the remaining structural
actions are listed as not started, with the commands to follow.
Thanks to xet7 !
Upcoming now lists the CodeQL #548 fix as a code-scanning security entry,
the Board Settings / Card work (Scrum rows, drag and drop, both sides for
every row, rows for what had none), Scrum settings' new place in the
menu, Taskwarrior import and export, durable in-place rule moves and
checklist creation and removal, and the #6737 date custom field tooltip.
The In short paragraph covers them, and over-long prose lines in the
section are wrapped at 80 columns.
Thanks to xet7 !
Feature visibility now covers 146 locales. Add the three new card labels in these fourteen locales and correct the Arabic Uzbek linked-card label. Synchronize the new keys elsewhere without overwriting translations. Specialist flow wording, especially Arabic Uzbek, Bashkir, Tatar, Turkmen, Uyghur and Kurdish, has lower confidence and remains open to native review.
The card color, the parent card path and prefix, the linked card
indicator, the rule card buttons, the description badge and the unread
comments marker were drawn with no Board Settings / Card row, so a board
could not turn them off. Each has a row now, with a card and a minicard
side. Where something was drawn before it stays on; the description
badge on the card, the card buttons on the minicard and the unread
marker on the card are new and off until chosen. The minicard's card
buttons read the board's rules from one board-level subscription, taken
only while that side is on, and run a rule without opening the card.
Not given rows, by decision: the archived banner and the upload progress
are status, not options; week-of-year is the user's own Member Settings
preference; card aging has its own board setting. The new labels are in
English, pending Transifex.
Thanks to xet7 !
Eighteen rows of Board Settings / Card had only one column. Each now has
both. On the minicard: Location, Requested by, Assigned by, Flowtime and
Pomodoro while running, the attachment names, the text-note count, when
the card last changed, the description title, and the checklist title
and due date. On the opened card: the collapse caret's own switch, labels
above the title, labels text (with the user's own override, like the
minicard's), the card's List title, the swimlane name and the comment
count. A side that already drew something stays on; a new element is
off until chosen. The new minicard elements are last in the minicard
order by default and reorder like the rest.
The rows' switches are plain board flags handled by one generic toggle
that reads each row's declared default. The List title row is offered
only when the popup was opened for a card, since from Board Settings
there is no card for it to change.
Thanks to xet7 !
Add Javanese, Haitian Creole, Latin, Hausa, Shona, Yoruba and Igbo messages. Correct the Latin feature heading. Specialist flow-metric wording in Latin, Hausa, Shona, Yoruba and Igbo has lower confidence and remains open to native review.
The up and down arrows of Board Settings / Card and Board Settings /
Board View are replaced by drag and drop. With drag handles on, a handle
sits where the arrows were; with them off, the row's icon and label are
what is dragged. The Up and Down arrow keys on the focused handle or
label still move a row one step, so the order stays reachable from the
keyboard.
A dropped Card row is walked to its place one step at a time through the
same moves the arrows made, so a pinned header stays first and a field
leaving its section takes the section along; head, tail and modifier rows
have no place of their own and are not draggable. A dropped Board View
order fills the slots of the shown rows, so a view the instance disabled
keeps its place. The DOM order is cancelled at the drop and the rows are
redrawn from what was stored. The new hint text is pending Transifex.
Thanks to xet7 !
Specialist flow-metric wording in these three languages has lower confidence and remains open to native review. Preserve existing translations and explicit data-retention and permission semantics.
GitHub CodeQL code scanning alert #548 (js/useless-regexp-character-escape)
flagged `\${` in a template literal in tests/templateRegexDos.test.cjs:
the text is a string-template regex directive, and the escape read as a
useless regular-expression escape. It is concatenated from plain strings
now, and so are the two regex samples in noIdentityReplacement.test.cjs
that had the same shape. A parsed guard fails when any template literal
in a regex context escapes `${` again; shell scripts quoted in template
literals, which need the escape, are not regex material and stay.
Thanks to xet7 !
The Scrum fields were a separate "Scrum settings" checkbox list under
each column of Board Settings / Card, and drawn as one block on the card
and the minicard. Each is now a row of both columns - "Scrum settings:
Sprint", "Scrum settings: Release" and so on - ordered with the other
rows, and drawn on the card and the minicard where the board's order puts
it; by default where the block was. The minicard gained Past sprints and
Acceptance criteria. The minicard's work item type row is the switch of
the work item type badge, which stays on unless turned off. The column
actions set the Scrum rows too, through scrum.configure so its lock,
revision and Scrum history apply.
Thanks to xet7 !
Translate all 24 feature visibility and approval messages in ninety locales, preserving existing values and regional terminology. Synchronize the new Taskwarrior import source key for the remaining work queue.
All 225 translation and i18n suites passed before the final locale extension; the expanded feature suite and full-catalog completeness checks also pass afterward. Regression assertions preserve the data-retention, permissions, approval and preview-audience distinctions.
Thanks to xet7 !
A Date custom field's tooltip always began with the Start date's "Starts
on" (card-start-on), so a field named "Initial due date" read as a start
date. The tooltip is now led by the field's own name, or is just the date
when the field has no name. The card's own Start date keeps its wording.
Thanks to rmb82 and xet7 !
Fixes#6737.
Scrum settings is now a group of its own between Rules and Change color,
with a rule between it and Change color. It stays board-admin only. The
menu order guard and the Board View docs diagram follow the new order.
Thanks to xet7 !
Taskwarrior's task export / task import JSON is the task format of the
Taskwarrior command line tool, and WeKan read and wrote neither. It now
imports from the import picker, as a JSON array or one task per line as
versions before 2.4 write it, and exports from the board's export menu.
A completed task goes to Done, a waiting one to Waiting and a started one
to In Progress; project and priority become labels, tags become labels,
entry, due, scheduled and end become dates, annotations become comments
and depends becomes blocked-by dependencies resolved by uuid. WeKan's
wekanlist and wekandescription attributes, which Taskwarrior keeps as
orphaned UDAs, carry list names and descriptions across a round trip.
Deleted tasks, recurring templates, bad dates and unmapped attributes are
reported in the loss report instead of guessed.
tests/taskwarrior.test.cjs covers the mapping, both input shapes, the
round trip and the negatives; the format guards learn the new format and
a Playwright case imports a Taskwarrior export through the page. The new
instruction text is in English, pending translation.
Thanks to xet7 !
addChecklist and removeChecklist rule actions now have saved commands,
like the other durable rule actions. addChecklist inserts a checklist
whose id is derived from the rule invocation, so a replay inserts it
once, and records the hook's addChecklist activity and History lifecycle
row from the document as it was stored. removeChecklist saves every
checklist the ordinary selector matches, one unit each, and for each
writes the removeChecklist activity before the removal and the 'removed'
History row after it, refusing a checklist that changed since capture.
The checklist hooks are deferred only for the checklist a command names.
Thanks to xet7 !
Fill four messages across all 234 non-English locales, preserve regional English sources, synchronize newly added feature keys and remove completed pending entries. All 224 translation and i18n suites pass.
Low-confidence specialist wording is recorded for native review, especially minority-language and constructed-language locales: ace, ak, ay, bi, bm, bo, bua, chr, csb, cv, dz, ee, ff, fj, gn, gv, haw, hsb, iu, kl, kok, ks, kw, lg, lld, mai, nah, nap, nd, nso, om, qu, rn, rup, sah, se, sm, ss, st, szl, tig, tlh, tn, to, ts, uz-AR, ve, ve-CC, ve-PP, vo, wa, wa-RR, wal, wo, wuu-Hans and zgh. Tests check structure and selected language distinctions, not fluency.
Thanks to xet7 !
A moveCardToTop or moveCardToBottom rule whose list and swimlane are '*'
on the card's own board only changes the card's sort. It now has a saved
command, like the card-field actions: the sort is computed once at
capture as the ordinary action computes it, the write is conditional on
the saved sort, the hook's position History row is planned and written
once, and the legacy UserPositionHistory row that Ctrl+Z reads is
written once with trackChange's redo-stack clearing. Replays apply the
saved command and change nothing more.
A move to another list, swimlane or board is reported to eligibility as
'<type>:elsewhere', so a board with such a rule keeps direct Sync.
Thanks to xet7 !
The Upcoming section now lists the Features group in Admin Panel /
Settings / Visibility, the opt-in act-editCard webhook event and Sync
lifetimes for old lists, beside the translation regression entry
already there. TODO Later no longer lists #4912 or the old-list item as
waiting.
Thanks to xet7 !
Admin Panel / Settings / Visibility has a new Features group. Each
optional group of board views (Table, Calendar, Time, Overview, Gantt,
Scrum, Flow charts, Map) can be turned off for the whole instance. A
disabled feature is removed from the Board View menu and from Board
Settings / Board View rather than greyed out, and a user who had it open
is moved to the board's default view. Its data and every permission stay
as they are.
Administrators can preview disabled features, and named pilot users see
them too. An opt-in policy keeps features added by later WeKan updates
disabled until an administrator enables them; the features present when
it is turned on stay as they were. Swimlanes and Lists are core and
cannot be disabled. The English texts are pending Transifex.
Thanks to nalilord and xet7 !
Fixes#6736.
globalWebhookEditCardActivity pinned the old webhook selection and call,
which the opt-in event replaced on purpose; it now pins the shared
per-webhook choice. notificationActivityGroups read act-editCard as a new
activity type; it is a webhook-only event name, so it is listed with that
reason.
Thanks to xet7 !
Restore current key order across locale files. Preserve the 2,417-string completion milestone while keeping newer feature work visible in ordinary translation reports. Retain full-catalog placeholder and feature-specific checks, with negative coverage for regressions and hidden backlog.
Thanks to xet7 !
Such a list had no incarnation, so it stayed on direct Sync for good.
Saving its Sync settings now assigns one in the same list update that
selects the new credential, and the credential is bound to it, so the
list becomes eligible for durable Sync. No migration assigns it, so an
upgrade leaves every stored credential matching.
Thanks to xet7 !
A webhook whose activity list names act-editCard now receives that event
for card title and description edits, with a `field` parameter saying
which one changed. A webhook on `all` or on the per-field events keeps
receiving act-a-changedTitle / act-a-changedDescription as before, and
every webhook still gets one delivery per edit, so no receiver sees the
same edit twice. Ordinary dispatch and stored Sync webhook plans make the
same per-webhook choice. The English event text is pending Transifex.
Thanks to Rishats and xet7 !
Fixes#4912.
Legacy lists get an incarnation when their Sync settings are next saved,
durable moves are the next structural rule action, #4912 becomes an
opt-in act-editCard event beside the per-field ones, and #2460 stays
open.
Thanks to xet7 !
Fill 27 Ladin placeholders for notification activity types, due date
reminders, duplicate dependencies, string template hints, saved filters,
import reports and the draggable label. Low confidence pending
native-speaker review.
The Ladin key-order check now skips keys pending Transifex, which live
in en.i18n.json only until the translating agent adds them.
Thanks to xet7 !
The login lockouts count per address and per (account, address), so a
guesser spread over many addresses was never slowed, and a hard per-account
lockout would let a stranger lock the owner out (JamBleed). After five
failures on one account within 15 minutes, from anywhere, the account now
accepts one attempt per slot - 1 s doubling to 30 s - over REST and DDP
alike; an early attempt is refused before the password is looked at, and
addresses the account has signed in from are not slowed. Refusals show as
BruteBleed in Admin Panel -> Problems, naming the account as the target.
LDAPBleed and BruteBleed move to the guarded list.
Thanks to xet7 !
An Organization's own admin could set its tenant hostnames to anything not
yet claimed, the instance's ROOT_URL host included, which put that
Organization's branding on everybody's sign-in page. The instance host is
now refused for everyone and an org admin's attempt is recorded under
TenantBleed. A new hostname from an org admin is stored as a request that a
site admin assigns by saving it; dropping hostnames still applies at once.
The three new strings are in English only, pending Transifex.
Thanks to xet7 !
The Docker image and the snap set LDAP_REJECT_UNAUTHORIZED=false, so an
LDAPS or StartTLS server certificate was not verified unless an admin
changed it, and an empty value disabled verification too. The default is
true everywhere now, and only the exact value false turns it off; a
self-signed or private CA belongs in LDAP_CA_CERT.
Thanks to xet7 !
Since Scrum records got a per-lifetime incarnation (20275418d4), the Scrum
transfer normalizer refused it as an unknown field, so board export and
board copy of any Scrum board with a sprint, release or event answered 500.
The transfer drops it, as Scrum History does.
Thanks to xet7 !
GET /api/boards/:boardId/attachments had no error handling, so its refusal of
an assigned-only member fell through to the app page; it answers with the
refusal's status now.
Thanks to xet7 !
TrayBleed (notification tray writes) and CasTokenBleed (a CAS callback from
a browser that did not start the login) get catalog keys of their own; each
new fix's test names its vulnerability. ReadOnlyBleed gets the permissions
test it was recorded without, pinning the six Custom Field handlers and
requiring a write-level decision on every mutating REST route.
Thanks to xet7 !
Nine catalog keys had no caller, so their Hall of Fame names could never
appear in Admin Panel -> Problems. Now recorded: a hostname resolving to an
internal address (DnsBleed) and a webhook refused by the SSRF guard
(IntegrationBleed), a Trello import carrying a non-http source URL
(SourceBleed), a client call of the OIDC-login-only org/team methods
(OIDCBleed), a sign-up with a wrong invitation code (InviteBleed), and a
read-only member writing over REST (ReadOnlyBleed). The three keys that
nothing can attribute are listed with their reasons in the test.
Thanks to xet7 !
The callback stored the validated CAS identity under whatever casToken its
URL carried, so an attacker could choose a token, get a victim signed in to
CAS to open a CAS login link carrying it, and then log in with that token as
the victim. The starting browser now sets a SameSite cookie with its token
and the callback must present it; a mismatch is recorded under CasBleed in
Admin Panel -> Problems. A ticket parameter on an instance without CAS is
left to its route.
Thanks to xet7 !
getProgress and getUnconvertedAttachments sent the full stored attachment
documents - storage paths, versions, uploader, names of files on cards the
caller cannot see - to anybody who could read the board. The client uses
only the ids.
Thanks to xet7 !
The attachment and avatar restore checked that an entry's name stays inside
the files directory, then wrote with fs.createWriteStream, which follows a
symlink already on that path. It now skips such entries and replaces the
file exclusively, as the full-backup restore already did.
Thanks to xet7 !
One try/catch wrapped every settings read at startup, so a single failed
read skipped accountsLockout.startup() and left the server without any
lockout. Each read now falls back to its default.
Thanks to xet7 !
Every visitor subscribes to 'setting' before signing in, and it carried the
directory server's host, port, base DN, bind account DN, search filter and
encryption mode. Only Admin Panel -> LDAP reads them.
Thanks to xet7 !
A chat incoming-webhook URL carries its secret in the path, and the board
publication sent it to every reader of the board, anonymous visitors of a
public board included. Other readers now get only the fields the
card-opened hook needs, and outgoingWebhooks finds the integration by _id.
Thanks to xet7 !
profile.notifications fell under the client-writable profile.*, and an entry
names an activity whose card, comments, checklists and attachments the
notification publications then send from any board. A client may now only
set an entry's read state and pull entries; anything else is denied and
recorded in Admin Panel -> Problems.
Thanks to xet7 !
METRICS_ACCESS_TOKEN set but empty let ?access_token= through, and the loose
early-exit comparison leaked how much of a guess was right through timing.
Thanks to xet7 !
The board publication shows an assigned-only member only the cards assigned
to them, but the attachment API (REST and DDP list, download and info), the
activities publication, Due Cards with all users, Global Search, the
structural move pickers and position history asked only for board membership
and so showed every card of the board. Each now applies the same rule; a
download or info request for another card's attachment is recorded under
AssignedBleed in Admin Panel -> Problems.
Thanks to xet7 !
importBoard and cloneBoard hand the client's currentBoard id to the creators,
which on Sandstorm archive that board. The id was never checked, so anyone
could archive any board by naming it. Only a board admin's board is
replaced now; otherwise the import runs and leaves that board alone.
Thanks to xet7 !
Trello and WeKan imports write the board through Boards.direct, which skips
the Boards.before.insert hook that enforces Admin Panel's private-only
policy, so an export marked public or instance arrived open. Both importers
now ask boardPermissionUnderPolicy() before writing.
Thanks to xet7 !
A legacy or imported rule action with no boardId made the move and link actions
look up lists and swimlanes with { boardId: undefined }, which matches by title
on any board, while the cross-board check compared against the activity's
board. performAction now gives such an action the activity's board once, before
any lookup or check; the cached document is not changed.
Thanks to xet7 !
checkUserId, checkAdminOrCondition and checkBoardWriteAccess refused only when
the admin lookup returned exactly undefined, and checkUserId refused only an
undefined user id; a null from the cache or the request would have passed. They
refuse any falsy value now.
Thanks to xet7 !
Every REST answer is the caller's - board exports as ZIP, PDF, Excel, CSV and
JSON, attachments as base64, board data - and none carried a cache policy, so a
shared cache was free to keep them heuristically; a public board's export is
the made-private-later case CacheBleed's fix names. The API gate now answers
every API request with Cache-Control: no-store unless a route sets its own. The
legacy /cfs/files/attachments route served files with no cache policy and
answered every HTTP method that the universal file server passed on; it serves
GET and HEAD only, with the private file policy.
Thanks to xet7 !
A History row keeps the board it was recorded on. Lists and swimlanes can move
to another board since, and restore/undo/redo then wrote them - and, for a
deleted list, its cards - wherever they are now, checking only the recorded
board; a row whose content moves something to another board was applied
without access there either. Restore, undo and redo now need write access on
the list's or swimlane's current board and on any board the row moves
something to. Not recorded, as for HistoryScopeBleed: a demotion while the
History pane is open reaches it. The test runs the real check with a moved list
and a cross-board card row.
Thanks to xet7 !
The outgoingWebhooks method checked that the integration exists and the caller
is a board member, then built the request from the CALLER's integration object
- whose type decided two-way, whose token was sent - and the caller's own
description and params. Any member, read-only included, could post arbitrary
text to the board's Slack, Matrix or other chat webhook as WeKan, without a
rate limit, or turn a one-way hook into a two-way one whose answer rewrites
comments. The request is now built from the stored integration; from a client
the only call allowed is the card-opened notification, with its params rebuilt
on the server from a card on that board. Anything else is refused and recorded
as HookBleed. Unit tests (including the forged call) and a browser test; two
tests whose stub integration lacked a url and type now stub the stored one.
Thanks to xet7 !
The api.attachment.copy and api.attachment.move methods wrote the target card,
list and swimlane ids as given, checking only write access to the target board,
so a file could be planted on a card of a board the caller cannot write - a
card's attachments are published by meta.cardId alone. They now require all
three to be on the target board, as the REST routes already did. Copying - by
method or REST - also follows the copy-scope rule: an assigned-only member
copies files only from cards assigned to them.
Thanks to xet7 !
With WITH_API off, the API gate refused req.url.startsWith('/api') - but
Express matches routes case-insensitively and on the decoded path, so /API/...
and /%61pi/... reached every API handler anyway. The gate now decides the
same way. The test covers the spellings Express routes and those it must not.
The isImpersonated method answered for any user id, to anyone, signed in or
not; it now answers only about the caller, or to a site admin.
Thanks to xet7 !
The ${"regex":...} token of a string-template custom field runs a pattern a
board member writes against a value another member writes. Rule e-mails format
card details on the server, so a pattern like (a+)+$ on 'aaaa...!' stopped the
one event loop for every user. The server now runs the regex in its own V8
context under a 50 ms limit, which V8 enforces inside backtracking, and pattern
and value are capped in length everywhere (the browser too, so one member's
format cannot hang the others'). A stopped regex leaves the token visible. The
negative test fails on any server call of the formatter without the bounded
runner.
Thanks to xet7 !
The Trello zip import checked entry sizes from entry.vars.uncompressedSize, a
field unzipper's directory entries do not have - the size is
entry.uncompressedSize - so every size check saw 0, and an archive of a few
hundred kilobytes inflated without limit in memory: any logged-in user could
exhaust the server. The WeKan zip import read wekan.json with no cap. A
declared size is the archive's own claim anyway, so entries are now read
through readZipEntryBounded, which counts the bytes that actually inflate and
stops past a per-entry and a per-archive limit.
POST /api/import/zip also allowed any active member to import into a board;
importing writes, so it needs write access, like the DDP import method.
The test builds a real small archive that inflates to megabytes and reads it
through unzipper; the negative test fails on any server code that inflates a
zip entry whole or reads entry.vars.
Thanks to xet7 !
GET /api/user and the attachmentMigrationStatuses publication found the
caller's boards with a dotted 'members.userId', which also matches boards the
caller was removed from, so a removed member kept seeing those board ids (and
the publication looked for public boards by a field that does not exist). Both
match active memberships now, and boards readable without membership by the
shared rule. POST /api/boards/:boardId/copy hand-wrote its admin check without
isActive, so a removed former admin could still copy the board; it uses
Authentication.checkBoardAdmin. Negative tests fail on any REST handler or
publication that finds the caller's boards by a dotted members.userId, or that
hand-writes the admin check.
Thanks to xet7 !
copyCard, saveCardAsTemplate, copySwimlane, copyList and moveList checked only
that the caller belongs to the source board, so an assigned-only member could
copy a card they are not assigned to (by its id) - or a whole list or swimlane,
with every card in it - into a board they control and read it there. One rule
(mayCopyFromBoard in models/lib/boardCardScope.js) now allows such a member to
copy a card assigned to them and no container; everyone else copies as before.
Not recorded: copying a list is ordinary UI. The test drives the rule and
checks all five methods ask it; the moveList harness gets the rule's module.
Thanks to xet7 !
146d58df27 created models/lib/boardBackground.js for isOwnBoardBackground - but
that file already existed: computeBoardBackground, which the client uses to
switch board backgrounds (#4978). Overwriting it broke background switching in
the UI and tests/boardBackground.test.cjs. The original is restored, and the
BackgroundBleed rule lives in models/lib/boardBackgroundOwnership.js, used by
both download paths and the Boards deny rule.
Thanks to xet7 !
The boardChartData method checked only that the caller can see the board, but
its loaders read every card, so Meteor.call('boardChartData', boardId,
'agingWip') returned the titles, dates and assignees of the cards an
assigned-only member cannot see - the data the chart export routes refuse
them. It now uses canExportBoardData, which allows such a member only the
charts that scope to their own cards. The negative test fails on any caller of
the chart loaders without the export rule.
Thanks to xet7 !
Authentication.checkBoardAccess, the check behind 32 REST routes that return
whole-board data - lists, cards, checklists, comments - kept its own
hand-written role list and let the three assigned-only roles through, so an
assigned-only member's API client read every card the UI hides from them. They
are refused there now (not recorded: such a member's client reaches it in
ordinary use) until those routes scope to assigned cards, which is recorded in
TODO Later. The test drives the real check per role; a browser test covers the
REST answers.
Thanks to xet7 !
BypassBleed (2022) reported, besides the authentication bypass, that two rules
undoing each other - label added: remove it; label removed: add it - recursed
without end: a rule's action writes a card, the write inserts an activity, and
the activity runs the rules again in the same awaited chain. That part was
never fixed. Rule runs started by a rule are now counted in async context, and
a chain deeper than five stops and is reported to Problems as BypassBleed
(detected, medium: usually an honest mistake). Separate activities are not
affected. The test runs the real executeRules with the reported loop;
BypassBleed moves from the recorded gaps to guarded and from pending to keyed;
two tests that load RulesHelper in a vm get the guard as a pass-through.
Thanks to xet7 !
A passwordless sign-in code was Meteor's default - 6 hex characters, 16.7
million values - valid for an hour. The account lockout counts failed
passwords, not codes, and the only limit was Meteor's per-connection rate
limit, which an attacker multiplies by opening connections: a thousand of them
gave about one chance in ten per code within its hour, for any account where
passwordless sign-in is enabled. Codes are now 10 characters (over a trillion
values) and live 15 minutes, which makes the same spray about one in three
million. The code field no longer asks for a numeric keyboard - a hex code has
letters - and takes the whole code.
Thanks to xet7 !
The fix makes HTML comment markers in rendered card text visible so nothing can
be hidden from readers, but used String.replace with a string pattern, which
replaces only the first occurrence - a second comment stayed hidden, and
DOMPurify then dropped it unseen. Every occurrence is replaced now. The test
applies the real replacement to text with two comments; the negative test fails
on any first-only comment replacement. InvisibleBleed moves from the recorded
gaps to guarded.
Thanks to xet7 !
The original fix refused markup in an attachment's new name on the server.
"Try to fix build errors" (2023-02-21) deleted that check, leaving only the
client's, so a direct renameAttachment call stored names like
<img src=x onerror=...>.png. Every display path escapes names today, so it did
not execute. The server refuses such a name again and records FileNameBleed
(medium), and stores every other name cleaned, as an upload's is. The negative
test fails on any server code that renames a stored file to an unchecked name.
Thanks to xet7 !
applyListWidth, setBoardAutoWidth and setStickyListHeaders change board-wide
state that every viewer sees, and refused only non-members, so read-only,
comment-only and worker members could change them. They now require write
access (not recorded as attempts: the menus may still offer them).
applyListWidth also awaits its write, which it had fired and forgotten. The
negative test fails on any server method that writes board content after a
membership-only check.
Thanks to xet7 !
ErrorBleed fixed boards.js and users.js. Thirty-four more REST handlers - lists,
swimlanes, card counts, integrations, org, team, settings, attachment storage
settings, rules and dependencies - still answered a refused request with HTTP
200 and the raw error object, or the raw object with its status, so a
non-member's request read as a success and an unexpected failure exposed its
internals. All now answer through publicErrorData(). The negative test scans
the whole server tree for the old shapes, single- and multi-line; a browser
test checks a non-member gets 403.
Thanks to xet7 !
The REST fix ignores owner and role fields, but Boards.insert from a client
accepted its own members array, so a user could create a board that names
somebody else as its admin, or adds people as members who never joined. The UI
sends no members (the schema makes the creator the sole admin), so a client
insert naming anyone else is refused and recorded as OwnerBleed - which now has
a Problems key and leaves the pending list.
Thanks to xet7 !
The /cdn/storage/avatars and /cfs/files/avatars routes in the universal file
server answer before avatarServer.js, and served an avatar inline under its
stored type unless that type exactly matched a short dangerous list -
'text/html; charset=utf-8' did not - with no CSP. An avatar's owner could set
that type from the client, so an image-looking upload could be re-labelled
and served as HTML on WeKan's origin. Only known raster image types are now
served inline, under their canonical name with a restrictive CSP; anything else
is a sandboxed download.
The client rules for avatars and attachments allowed updating type, extension,
size and meta - meta names an attachment's board and card, so a file could be
moved onto another board's card. No client updates these documents; only the
name may change now.
Thanks to xet7 !
The on-disk name of an upload starts with its file id, which the client sends
in opts.meta.fileId. Attachments validated it (UploadPathBleed); avatars - the
same Meteor-Files pipeline - used it verbatim, so an avatar upload with a
fileId like ../../x wrote outside the avatar store, and Avatars.onBeforeUpload
did not require a signed-in user. Both collections now share one rule
(models/lib/uploadFileId.js): an id that does not look like WeKan's own is
replaced with a fresh one and the attempt recorded; avatars need a signed-in
uploader. The negative test fails on any namingFunction in the tree without
the rule.
Thanks to xet7 !
Clients could insert position-history entries as long as the entry's boardId
and previousBoardId were boards they belonged to. Nothing tied the entry's
card, list, swimlane or checklist item to those boards, and redo had no check
at all, so a member could forge an entry naming another board's card - or a
newBoardId, or a delete/restore action - and undo/redo would move it into
their own board or soft-delete another board's list with its cards.
The server records history itself (trackChange), so client inserts are now
refused outright. Undo and redo first check that the entity is still on one of
the entry's own boards (a cross-board card move is recorded under the board it
left) and that the user can write there and on the destination. Refusals trip
the existing history.cross-board canary. Unit and browser tests; the test that
pinned the weaker insert rule pins the stronger one.
Thanks to xet7 !
The browser test found two faults in dd36db3641. The real door is
useraccounts' ATCreateUserServer - the method the sign-up form calls, which
picks the profile fields and passes the rest of the client's options to
Accounts.createUser - not Meteor's createUser, which useraccounts already
refuses to clients. Verified against the code before this change: with
registration disabled, ATCreateUserServer with from: 'admin' created the
account. Both methods are wrapped now. And the wrapper handed the original
method a copy, so audit-argument-checks refused every ordinary sign-up ("Did
not check() all arguments"); the client's own arguments are check()ed first.
The browser test now signs up ordinarily first, then makes both attempts the
way the form does (hashed password) on fresh connections, and resets the
shared unknown-user lockout of 127.0.0.1 before and after. The LinkedWriteBleed,
RepointBleed and BackgroundBleed browser tests use a fresh account per refused
attempt and check it is disabled - a refused attempt costs the account.
Thanks to xet7 !
With BROWSER_POLICY_ENABLED (the Docker and Snap default) WeKan is meant to be
frameable only by itself and TRUSTED_URL, so another site cannot load it in an
invisible frame and trick a signed-in user into clicking through it
(clickjacking). When browser-policy was removed on 2022-04-08, every line of
server/policy.js was commented out, and since then the app's pages carried no
framing header at all; only file responses did.
Every response now gets Content-Security-Policy: frame-ancestors 'self' plus
the TRUSTED_URL origins, and X-Frame-Options: SAMEORIGIN when no trusted origin
is set; a malformed TRUSTED_URL never widens it, and Sandstorm, which always
frames WeKan, and a disabled policy send nothing. build.sh and build.bat run
the test server with the default policy. FrameBleed moves from the recorded
gaps to guarded, with unit, negative and browser tests. A framed load cannot be
told from a normal one, so there is nothing to record in Problems.
Thanks to xet7 !
With an admin-defined activity e-mail body template (#2022), the HTML body was
the substituted template, and the values substituted into it - card, board and
list titles, the username, a comment - are written by members, so markup in a
card title reached other members' HTML mail. The template stays trusted and may
carry HTML; the HTML body now substitutes escaped values, and the text body is
unchanged. Rule e-mails and invitations are text only. The test drives the
substitution with hostile values; an assertion that had pinned the vulnerable
form now pins the fixed one.
Thanks to xet7 !
A blocked record of high or critical severity that names a userId disables
that account. Records on paths ordinary users reach therefore took accounts
away from legitimate users:
- authz.card-member, when a client wrote back assignees still listing a removed
member - now recorded only for ids that were never on the board;
- History restore after the user was demoted - no longer recorded, as the
HistoryScopeBleed deliberate omission already says;
- export by an assigned-only member, who reaches the export menus - recorded
only when the caller cannot see the board;
- a large legitimate import hitting a size limit - recorded at medium (cycles
and non-plain objects stay high), and the record can no longer throw;
- ssrf.attachment (a Trello board's link to an internal host), ssrf.redirect (an
identity provider's avatar) and authz.upload-path (a cached older client) -
still recorded, but blocksAccount: false;
- a REST client writing back a card with the parent it already had - only a
changed parent is checked. That PUT now also answers 404 for a card that is
not on the named board and list instead of acting on the id alone.
tests/problemsFalseBlocks.test.cjs drives the blocking decision and the real
assignee filter, and pins each condition; four tests that pinned the old calls
follow them. MegaBleed moves to guarded (its Translation sibling test).
Thanks to xet7 !
GET /api/attachment/download-background/:boardId and the
api.board.downloadBackground method returned whatever attachment
board.backgroundImageId named, checking only board membership, and a board
admin could set that field to any attachment id. Anyone could create a board,
point its background at an attachment on someone else's private board, and
download the file.
Both download paths now serve only a live attachment of the same board, and a
Boards deny rule refuses setting backgroundImageId to anything else. Refusals
record BackgroundBleed (the UI only offers the board's own backgrounds). The
negative test fails on any server code that streams a backgroundImageId
without the check; a browser test covers both the update and a planted pointer.
Thanks to xet7 !
The Translation collection's client rules let any logged-in user insert,
update or remove a document whose _id equalled their own user id - the same
mistake TenantBleed removed from Org and Team. Custom translations replace UI
and e-mail text for everyone and several render as HTML
({{{_ 'board-public-info'}}}, the legal notice), so a member could plant markup
that runs in an admin's browser. The MegaBleed fix had guarded only the
methods. All three rules now use the shared site-admin rule, which records a
refused attempt in Problems; the negative test fails on any permissions file
that authorizes by document id. Browser test included.
Thanks to xet7 !
Accounts.onCreateUser trusts from: 'admin' (admin or REST creation, which
skips the registration checks) and ldap: true (directory account, allowed while
registration is disabled) as server-only signals, and Meteor's createUser DDP
method passes the client's options through unchanged. With "Disable
registration" on, anyone could still create an account by adding either.
The createUser method is wrapped so a client's options lose the server-only
keys; the server's own callers - the admin method, POST /api/users, the LDAP
package - call Accounts.createUserAsync directly and keep them. An attempt is
recorded as SignupBleed. The negative test fails if onCreateUser trusts another
option as an early-return signal without it being stripped from clients; a
browser test signs up both ways with registration disabled.
Thanks to xet7 !
POST /api/boards/:boardId/members/:userId/add and /remove read the target user
into the variable they answered with, so any action other than add or remove
sent the whole user document back - password hash and login-token hashes
included - to any board admin, which anyone becomes by creating a board. An
unknown action is now refused before the user is read, and the user is held
apart from the answer.
The negative test fails on any REST handler in the server that sends a
variable read from the users collection without withoutSecrets(); a browser
test calls both routes with an unknown action.
Thanks to xet7 !
The per-user layout methods keep maps keyed by the ids a client sends -
collapsed lists, cards and swimlanes, list widths and heights, board sort
indexes, workspace assignments - and wrote them as
map[boardId] = map[boardId] || {}; map[boardId][listId] = value. With boardId
'__proto__', map[boardId] is Object.prototype, so any logged-in user could set
a property on every object in the server process until a restart
(Meteor.call('setListCollapsedState', '__proto__', 'isAdmin', true)).
Every function in models/users.js and server/models/users.js that writes such
a map now checks its keys with assertSafeMapKey (models/lib/safeMapKey.js):
never __proto__, constructor or prototype, nor a '.' or leading '$' MongoDB
refuses anyway. A refusal is recorded as PrototypeBleed (no client sends one).
The test runs the real method bodies in their own vm context and shows the
pollution before the guard and none after; the negative test fails on any
id-keyed map write without the check. A browser test makes the reported call.
Thanks to xet7 !
Triggers, Actions, Rules and Integrations were allowed to a board admin of the
board the document was on BEFORE an update; nothing refused the update itself
changing boardId. Any registered user, admin of a board they created, could
move a rule trigger to '*' or unset its board - which the rule matcher matched
on every board - and have a send-email action mail them cards, comments and
attachments from every private board; or move a webhook integration onto a
private board and receive its activity, or rewrite its comments through a
two-way hook. Custom fields accepted a board list if the caller could write to
ANY board in it, so a field could be planted on (or pulled from) a board
without access; comment reactions could be moved onto another comment.
A shared deny rule now refuses any client update that changes those
documents' board, custom fields need write access on every board they are put
on or taken off, and the rule matcher runs a rule only for activity on its own
board. Refusals record RepointBleed in Admin Panel -> Problems (high: no client
changes these fields). Unit, negative (every board-owned permissions file) and
browser tests.
Thanks to xet7 !
The lockout reporter recorded brute.lockout (high, blocked) with the GUESSED
account as its userId, and a high-severity blocked event with a userId
disables that account. So three wrong passwords from anyone who knew a username
disabled its owner until an admin re-enabled it - worse than the per-user lock
GHSA-rf3w-rj48-jxcc removed - and a user who mistyped their own password three
times was disabled the same way.
The reporter is now one module for startup and reload and names the account as
targetUserId; a catalog key can declare blocksAccount: false, and brute.lockout
does. An account already disabled keeps the reason it was disabled for instead
of later events overwriting it. tests/jamBleedVictimBlock.test.cjs drives the
real blocking decision with the real catalog; its negative test fails if a
lockout is recorded anywhere else or a login target is recorded as an actor.
Thanks to xet7 !
POST /api/boards/:boardId/lists/:listId/cards/bulk wrote each entry's parentId
with Cards.direct, which skips the DDP deny rule, and never checked that the
caller may see the parent card's board, so a writer on one board could attach
cards to a card on a private board. Each entry is now checked like the single
create; a refused entry gets a per-index error and the rest are created.
The authz.parent catalog key existed but nothing recorded it: a refused parent
now shows in Admin Panel -> Problems as ParentBleed (the UI only offers
visible parents, so a refusal is an attempt). The negative test fails on any
REST handler that writes a parentId before the check.
Thanks to xet7 !
Inviting someone to a board also added them to - or reactivated them on - the
board's subtask deposit board, checking the inviter's rights only on the board
they invited to. A deposit pointer needs only write access, so a writer (an
assigned-only member, or a normal member where only board admins may invite)
could point their own board's deposit at any board and invite anyone into it,
their own second account included, or reinstate someone removed from it.
The deposit board now admits the invitee only when the inviter may invite to
that board too (canInviteToBoard, the rule invitations use). It is not recorded
as an attempt: an inviter without rights on the deposit board is also an
ordinary configuration. The test drives the reported attack through the real
rule, and the negative test fails on any member write to a deposit board
before that check, anywhere in the server.
Thanks to xet7 !
POST /api/boards/:boardId/lists/:listId/cards with linkedId checked only READ
access to the source card's board and cloned the whole source card with
Card.link(), so a read-only, read-assigned-only or comment-assigned-only member
could link and copy cards - including unassigned ones and their descriptions -
into a board they control, which the createLinkedCard method refuses.
Both now call one createLinkedCardFor(): write access to the source board, an
assigned-only member links only cards assigned to them (a check the method
lacked as well), the list and swimlane must belong to the destination board,
and the link holds the title, not a copy. Refusals record LinkedWriteBleed with
source rest:card-link. The tests that pinned the method's body follow the
shared function; the negative test fails if the REST branch checks only read
access or clones the card again. A browser test covers the REST roles.
Thanks to xet7 !
Both were an async access check called without await, whose Promise never
stops the handler: 48 REST endpoints (BFLABleed) and the Excel export route
(ExcelBleed). The test reads which Authentication checks are async and fails on
any call to one, or to an exporter's canExport(), that is not awaited or
returned anywhere in the server tree. All 124 checks and every canExport()
call are awaited today. Both move from the recorded gaps to guarded.
Thanks to Fredrik Dietrichson and xet7 !
EmailBleed (2021, SMTP password readable in the Admin Panel) is published as
Emailbleed, which the case-sensitive Hall of Fame match skipped, and a comment
had called it an unpublished alias. Both coverage tests now match in any case,
EmailBleed is guarded by tests/adminMailSettings.test.cjs (passwords are never
published), and it is a deliberate Problems omission. RelayBleed is guarded by
tests/relayBleed.test.cjs.
Thanks to Georg Krause and xet7 !
The live Trello import downloaded attachments, the board background and member
avatars with the importing user's Trello API key and token in an OAuth
Authorization header, whatever host the URL named. A link attachment is a URL
anyone who can edit the Trello board chooses, and one renamed to look like a
file was downloaded like an upload, so its host received the importer's token.
The background and avatar downloads retried with the credential after a public
download failed, with the same effect.
The header is now built in one place, only for https://trello.com and
https://api.trello.com on the default port; every other download goes without
it, and fetchSafe still drops it on a redirect off the origin.
tests/relayBleed.test.cjs runs the real import code with the network stubbed
and checks which hosts receive the credential, including Trello lookalikes, and
fails if the header is built anywhere else. The live import calls the fixed
api.trello.com, so there is no browser test. A link elsewhere is ordinary board
content, so withholding the credential is not recorded as an attempt.
Thanks to xet7 !
GHSA-5q84-p3vr-f3xv: a board member with write access chooses the List Sync
server URL for Gitea, Forgejo, GitLab and Jira. It was fetched with the
platform fetch(), so it reached 127.0.0.1, private networks and the cloud
metadata address, and up to 200 bytes of a failed response were copied into
the error the preview returns and lastSyncError stores. An open and a closed
internal port also answered differently.
Every request now goes through fetchSafe: internal addresses refused, DNS
pinned, redirects refused, response size and time bounded. An error carries the
origin and HTTP status only, a non-JSON answer says so without quoting it, and
a guard refusal is one message that does not name what a host resolved to.
Saving Sync settings refuses an internal address too, so none is stored. Both
refusals are recorded as SyncBleed in Admin Panel -> Problems.
A self-hosted tracker on the server's network is allowed only by the
administrator, with LIST_SYNC_ALLOWED_PRIVATE_HOSTS (exact hosts); the test
server allows 127.0.0.1 for the List Sync specs' mock tracker. The browser test
found that an IPv6 literal such as [::1] kept its brackets and was looked up as
a name at save time; the shared validator now checks it as an address.
tests/syncBleed.test.cjs reproduces the report on the real fetcher for every
provider, and fails on a platform fetch() to anything but a fixed or
administrator-allowed host, or a response body copied into an error, anywhere
in the server tree. list-sync-source.e2e.js saves and previews internal
addresses in the browser and checks no request reaches the mock tracker.
securityRegressionCoverage lists CacheBleed and SyncBleed with their tests.
Thanks to alham-rizvi and xet7 !
A signed-in non-member imports My Dependencies on an instance-wide board and
cannot choose the board layer. Passes in Chromium, Firefox and WebKit against
the source-loaded test server.
Thanks to xet7 !
canViewBoard in the dependency access rules allowed public boards and members
only, so a signed-in non-member of an instance-wide board could read its cards
but not its dependency lines. It now uses readableWithoutMembership, the rule
board reads share; tests/boardPermission.test.cjs had flagged the public-only
check.
Thanks to xet7 !
GHSA-w3qg-pf27-g68r: attachments were answered with
Cache-Control: public, max-age=31536000 and no Vary after the board access
check, so a shared cache (CDN, caching proxy) could store a member's copy and
serve it to anyone without credentials for a year, also after the member was
removed from the board. The same header was in setFileHeaders, both ?download=1
branches, both avatar routes, and Meteor-Files' own /cdn/storage route sent its
default public, s-maxage one-year header.
Every file response, its 304 and the thumbnail now send
Cache-Control: private, no-cache and Vary: Cookie, Authorization, X-Auth-Token
through models/lib/fileCacheHeaders.js, for public boards too, because a board
made private later would leave a public copy behind. Both FilesCollections pass
the same cacheControl. The ETag keeps revalidation a bodiless 304.
tests/fileCacheHeaders.test.cjs drives the policy against the shared-cache rules
the report used, checks each route, and fails on any public Cache-Control, a
FilesCollection left on the Meteor-Files default, or an ETag-sending file without
the policy, anywhere in the tree. The thumbnail Playwright spec checks the
headers on every route. Nothing is refused, so there is no Problems key.
Thanks to alham-rizvi and xet7 !
Fill 23 English placeholders while preserving query syntax and archive safeguards. Ladin technical phrasing is low confidence pending native-speaker review; existing mixed-language entries still need correction.
Fill 29 activity recovery and synchronization time estimate messages with token and recovery safeguard checks. Technical terminology remains low confidence pending native-speaker review.
Fill 27 remaining messages and preserve reference tokens, UTC observations, report limits and distinct sprint states. Specialized terminology remains low confidence pending native-speaker review.
Fill 28 remaining messages and verify template variables, reminder offsets, distinct actions and the recovery menu path. Technical phrasing remains low confidence pending native-speaker review.
Fill 25 remaining messages while preserving rule variables, markup and URI identifiers. Technical phrasing remains low confidence pending native-speaker review.
Fill 23 remaining messages and preserve filter query syntax, placeholders and source order. Technical phrasing remains low confidence pending native-speaker review; full Cornish completion is still pending.
Add 11 missing dependency messages with preserved placeholders and distinct ownership and actions. Tigre technical prose remains low confidence pending native-speaker review. Align the Esperanto completion check with the deliberately restored literal Pause key label.
A rebase rewrote the commits these entries link to, so the links named commits
that no longer exist and would have 404ed on GitHub. Repointed by
releases/fix-changelog-hashes.sh, which only touches a link no ref in this clone
can reach.
Thanks to xet7 !
Fill 34 messages for sprint reports, daily observations and Sync conflicts. Preserve reporting limitations, count placeholders and source-system boundaries.
Fill 37 messages for Scrum completion policies, sprint planning, events and work status. Preserve existing translations and extend regression coverage.
Fill 35 messages for rounding, powers, logarithms, trigonometry and editor controls. Preserve mathematical notation and the degree-versus-radian distinction.
Fill 36 math labels and tooltips for number properties, list statistics and random values. Preserve numbered placeholders and inclusive or exclusive bounds.
Fill 38 messages for list repetition, insertion, sorting, splitting and Boolean comparisons. Preserve placeholders and match the sort label to its argument order.
Fill 38 messages for list creation, indexing, retrieval, removal and ranges. Preserve numbered placeholders and the distinction between retrieval and removal.
Fill 38 remaining field, block, comment and warning labels. Extend source-order and placeholder regression coverage while preserving existing translations.
Fill 35 messages for rule variables, notification categories, due reminders, dependency directions and saved filters. Extend token checks to brace-delimited template variables.
Fill 30 untranslated messages with Cantonese prose. Add regression coverage for placeholders, query syntax, link schemes, archival exceptions and board visibility restrictions.
Fill the final 50 reported Kazakh placeholders and add Kazakh to full-locale completeness coverage. Key order, placeholder inventories and human-preference checks pass. The broad suite still fails at missing Cherokee dependency keys; native-speaker review remains outstanding.
Fill 35 remaining messages for Sync previews, source-field reports and run outcomes. Preserve report limits, retention periods and partial-change warnings.
Fill 45 Scrum messages for sprint planning, events, reports and lifecycle confirmations. Preserve count and estimate placeholders and the distinction between unknown and zero estimates.
Fill 32 remaining messages and recognize 28 literal keyboard legends, platform names and mathematical symbols as Kazakh-specific invariants. Verify the new prose, literal labels and source placeholders.
Fill 37 remaining editor messages for variables, workspace announcements, search and legacy block labels. Preserve numbered placeholders and keyboard instructions.
Fill 51 remaining editor messages for text search, replacement, extraction, formatting and input prompts. Preserve source placeholders and indexing semantics.
Fill 49 remaining messages for screen-reader modes, keyboard shortcuts, variable renaming and text operations, preserving numbered placeholders and existing translations.
Fill 48 remaining messages for trigonometry, variables, functions and editor controls. Preserve placeholders and use the existing block backpack terminology.
Fill 47 editor labels and tooltips for list statistics, random numbers, rounding, powers, logarithms and inverse trigonometric functions. Preserve placeholders and interval boundaries.
Fill 48 editor labels and tooltips for comparisons, Boolean operations, arithmetic and number properties. Preserve all source placeholders and extend the Kazakh regression coverage.
Fill 48 editor strings and preserve numbered placeholders. Match the sort label to Blockly TYPE, DIRECTION and LIST argument order. Kazakh regression and human-preference checks pass.
Fill 34 remaining editor strings and extend Kazakh token and script checks. The focused regression block and human-preference checks pass. The broad suite still stops at missing Cherokee dependency keys.
Fill 42 editor labels with Kazakh prose and extend placeholder coverage. Recognize the restored physical Pause key legend in nine locale-specific invariant sets. The broad completion suite now reaches the outstanding Cherokee key gap.
Fill eleven dependency messages and check source key order and placeholder inventories. Terminology and grammar, including the coined relationship noun, have low confidence and need native-speaker review.
Fill eleven missing messages in Tifinagh and extend key-order and placeholder checks. Terminology and grammar have low confidence and need native-speaker review.
Fill eleven missing messages in syllabics and extend key-order and placeholder checks. Terminology and grammar have low confidence and need native-speaker review.
Fill eleven missing messages and extend key-order and placeholder checks. The dependency terminology and grammar have low confidence and need native-speaker review.
Fill eleven missing messages using dictionary terminology and extend key-order and placeholder checks. Wording has low confidence and needs fluent-speaker review; older inconsistent translations remain.
Fill eleven missing messages and extend key-order and placeholder coverage. The dependency terminology and wording have low confidence and need fluent-speaker review.
Fill eleven missing messages per locale and extend key-order and placeholder coverage. Terminology and grammar have low confidence and need native-speaker review.
Fill eleven missing messages in each locale and extend key-order and placeholder checks. Terminology and phrasing have low confidence and need native-speaker review.
Fill eleven missing messages per locale with key-order and placeholder checks. Terminology and grammar have low confidence and need native-speaker review; older mixed-language values remain for correction.
Fill eleven missing dependency messages per language and extend key-order and placeholder coverage. Wording has low confidence and needs native-speaker review.
Fill eleven messages per locale with preserved placeholders and source key order. Terminology and wording have low confidence and need native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has low confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has low confidence, especially Veps terminology, and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. These translations have low confidence and need native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording and Uzbek Arabic orthography have lower confidence and need native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. The wa-RR locale is Waray according to the language registry. Walloon and Waray wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved placeholders and source key order. Wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages in Shona, Southern Sotho, Tswana, Swati, Tsonga and Northern Sotho. Wording has lower confidence and needs native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages in Hausa, Igbo, Chichewa, Kinyarwanda, Zulu and Xhosa, including the regional Zulu locale. These translations have lower confidence and need native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale, preserving source tokens and key order. These translations have lower confidence and need native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages per locale with preserved tokens and key order. These translations have lower confidence and need native-speaker review. Dependency checks and all 21 human-preference checks pass.
Fill eleven messages across seven locale tags, preserving source tokens and key order. Pashto, Sindhi and Odia wording has lower confidence and needs native-speaker review.
Dependency checks and all 21 human-preference checks pass. The Khmer completeness suite still fails on its remaining untranslated backlog.
Fill Breton, West Frisian, Friulian and Romansh messages, including the Frisian regional locale, and synchronize Welsh and Japanese variants. Add missing source messages to eleven English regional locales without editing en.i18n.json.
Breton, West Frisian, Friulian and Romansh prose has lower confidence and needs native-speaker review. Dependency key-order/token checks, Japanese completeness and human-preference checks pass.
Preserve newer reviewed wording, align the Asturian template title, and restore literal Pause key labels in nine locales. Match current checklist markup in its translation regression.
Translate 11 dependency messages in Welsh, Irish, Scottish Gaelic and Faroese. Scottish Gaelic and Faroese wording has lower confidence and needs native-speaker review. Focused translation and human-preference checks pass; overall completeness remains pending.
A rebase rewrote the commits these entries link to, so the links named commits
that no longer exist and would have 404ed on GitHub. Repointed by
releases/fix-changelog-hashes.sh, which only touches a link no ref in this clone
can reach.
Thanks to xet7 !
833 translation entries linked from the v12.12 section are commits made after
v12.12 was cut and contained in v12.13, so they are moved to the v12.13
section and its Languages updated line lists the languages they changed.
CHANGELOG.md then keeps October 2026 only; the 70 September releases move to
old-CHANGELOG/2026/09.md. The format test measured the newest release to the
next release heading, which does not exist when it is the month's only one.
Thanks to xet7 !
releases/check-upcoming-release.sh counts only <summary><a href> entries, and
Upcoming held plain bullets only, so release-all.sh refused to run.
Thanks to GitHub CodeQL and xet7 !
The CodeQL alert 547 entry had landed inside the published v12.13 section,
but its commit came after that release; it moves to a new Upcoming section.
The DOMPurify 3.4.15 to 3.4.16 update merged after the release is recorded
there too.
Thanks to GitHub CodeQL, dependabot and xet7 !
Upcoming gains a SECURITY ISSUES found by GitHub CodeQL code scanning
subsection for the Trello link check in a translation test.
Thanks to GitHub CodeQL and xet7 !
CodeQL js/incomplete-url-substring-sanitization flagged
tests/vietnameseBulgarianCompletion.test.cjs: the Kashubian trello-api-key
translation was checked with includes('https://trello.com/app-key'), which
also passes for https://evil.example/?x=https://trello.com/app-key or a
lookalike host. The test now extracts each link from the text, parses it,
and requires the hostname trello.com and the path /app-key.
A new tree-wide suite, tests/urlSubstringSanitization.test.cjs, fails on any
URL or host literal checked with includes(), or with indexOf()/search() used
as a yes/no test, in source and tests alike; it skips comments and template
literals with interpolation (not constants), proves it recognises the
reported line and the earlier #531 form, and checks the new form rejects a
link in the query, in the path and on a lookalike host. No shipped code had
the substring form - the alert was in a test - so there is no Hall of Fame
row or Problems key.
Thanks to GitHub CodeQL and xet7 !
The dependency import now combines (cff80a9d7b): a line already on the board
is kept and counted as skipped. The All Boards / Import / Dependencies result
reported only imported and unmatched lines; it now uses the same message as
Member Settings / Import.
Thanks to nalilord and xet7 !
Issue #6732: the board-header "Show dependencies" button wrote one shared
board field, so one user turning the lines on or off did it for everybody,
and only board admins could write that field, so a member - or anyone on a
public board - who could see the board could not turn them on at all.
The button is gone from the header. At the top of Member Settings are now:
- Show My Dependencies
- Show Board Dependencies
- Import
- Export
Both switches are off by default and saved in the user's profile, so each
user decides for themselves and anyone who can view a board may show its
dependencies.
Board Dependencies stay on the board's cards. They may now be edited by
every role that can edit OR move cards - a Worker included - and by an
assigned-only member on the cards assigned to them; one rule
(models/lib/dependencyAccess.js) is applied by new server methods, by the
card details, the line popup, drawing on the board, Import and the REST
routes. My Dependencies are a new layer in the user's profile: each user's
own lines between cards they can see, drawn dashed and seen only by them.
The client cannot write them directly; only the methods, which check card
access, can.
Import and Export work on the open board in the Board Settings / Export
formats (JSON, and SVG that re-imports). Importing Board Dependencies is
offered only to roles that may edit or move cards, and every import now
combines: a line already there is kept exactly as it is, and only missing
lines are added - also for All Boards / Import / Dependencies, which used to
overwrite an existing line's type and colour. The design is written up in
docs/Features/Editor/RedStrings/Board-And-My-Dependencies.md.
Tests: the role table, viewing, assigned-only scope, normalizing and
combining (unit), with guards that the header toggle and every reader of
the old board flag are gone and every write goes through the rule; a
full-app test drives the methods as admin, Worker, read-only, assigned-only
and a stranger; Playwright covers the menu order and defaults, a per-user
switch that leaves another member's view alone, a read-only member's My
Dependencies (dashed, unseen by the owner) with import and export, and a
Worker's combining board import - in Chromium and WebKit. Header, fold and
mobile guards that listed the old button are updated. 567 server tests pass.
Thanks to nalilord and xet7 !
Fixes#6732.
Reported by email against v12.12: "Preview of Images is gone". On the board
a card with an image cover showed a flat coloured band, and in the open card
the attachment row showed a small mark on a blue bar instead of the picture.
Two separate faults:
- The minicard cover() helper loaded the attachment and then returned only
a wrapper with link(). attachmentPreviewUrl() needs the attachment's _id,
so it returned '' and the cover became url('?dummyReload...') - a request
for the page itself. The helper now returns the attachment, as
models/cards.js cover() does, keeping the linked-card lookup and the
soft-delete check.
- The attachment thumbnail is a <button> (the keyboard control with its
aria-label), and forms.css fills every button with the theme accent at
rest and on hover, focus and active. The local override only cleared
padding, border and font. Every state is now reset: a fixed 144x96 frame
(96x64 below 800px) on #f4f5f7, the whole picture shown with object-fit
contain, a 2px focus outline. My Attachments, which uses the same classes
on a div, keeps its previous sizing.
The board cover and the open-card cover are now img elements of the
thumbnail, drawn only when there is a URL: cover-fitted 120px bands on the
board (100px on the modern theme) and 180px in the open card, which used the
full original before. A thumbnail that fails to load falls back to the
original once and stops, without logging. The invalid `border-radius: top`
is gone. The table page's attachment preview is out of the accent-filled
action buttons and gets the same neutral frame in every state. forms.css,
the cover options' defaults, the /thumbnail route and SVG handling are
unchanged.
Tests: the fallback executed as written (once, no loop, originals left
alone) and source guards for the cover helper, templates and every reset
state; two guards that pinned the old cover background and the preview's
accent fill are updated with the reason. A new Playwright spec checks a
real cover image, the tile's neutral colour and size at rest and on hover,
the open-card banner, the once-only fallback and the cover option off - it
passes in Chromium and WebKit, with the thumbnail, admin preview,
attachments and board-attachment specs (21 tests) still passing.
Thanks to xet7 !
Fill 20 English placeholders while preserving existing translations. Extend coverage for inclusive date boundaries, time ranges and the template archival exception.
Translate 37 remaining recovery, sign-in and history messages. Retain 36 keyboard legends, brands, code symbols and shared Latin terms as locale-specific invariants. Include Latin in full key-order, token and placeholder checks; older linguistic quality still needs review.
Fill 34 English placeholders while preserving existing translations. Extend checks for cancellation boundaries, SMTP failure categories, time estimate mapping and activity-preserving retries.
Fill 29 English placeholders while preserving existing translations. Extend checks for report retention, null semantics, uncertain delivery and paused retries.
Fill 34 English placeholders while preserving existing translations. Extend checks for local content retention, unchanged subcards, replacement reuse and omitted diagnostic values.
Fill 34 English placeholders while preserving existing translations. Extend checks for UTC observation limits, unknown estimates, sprint cancellation and one-way sync conflict handling.
Fill 49 English placeholders while preserving existing translations and tokens. Extend checks for unknown estimates, velocity comparison requirements, sprint actions and consistent backlog terminology.
Fill 49 English placeholders while preserving existing translations. Extend checks for keyboard shortcuts, screen-reader spacing, permissions and rule conflicts. Update the README count after Latin crosses the existing coverage threshold.
Fill 65 English placeholders while preserving existing translations. Extend checks for whitespace, text replacement, search results, title case and variable type conflicts.
Fill 46 English placeholders while preserving existing translations. Extend checks for screen-reader state transitions, variable renaming scope and opposite navigation directions.
Fill 33 English placeholders while preserving existing translations. Extend checks for function return values, disabled definitions and function-only block warnings.
Fill 42 English placeholders while preserving existing translations. Extend checks for random-number endpoints, numeric negation, rounding directions and angle units.
Fill 44 English placeholders while preserving existing translations and formulas. Extend checks for degree limits, inclusive bounds, statistical distinctions and ternary label references.
Fill 53 English placeholders while preserving existing translations. Extend regression coverage for copy semantics, case-insensitive sorting, Boolean negation and conjunction versus disjunction.
Fill 49 English placeholders while preserving existing translations. Extend coverage for list reads versus mutations, missing-item results, sublist copies and keyboard argument order.
Fill 66 English placeholders while preserving existing translations. Extend coverage for bitmap argument order, row and column labels, editor state actions and mathematical operands.
Fill 40 English placeholders while preserving existing translations. Extend checks for while/until conditions, branch fallback, loop argument order and variable deletion prompts.
Fill 43 English placeholders while preserving existing translations. Extend checks for variable deletion restrictions, color ranges, collapsed warnings and loop flow semantics.
Fill 43 English placeholders while preserving existing translations. Extend checks for import warnings, map access and movement directions with unchanged argument order.
Fill 35 English placeholders while preserving existing translations and variables. Extend checks for reminder day direction, notification exceptions, duplicate relationships and saved filter privacy.
Fill 35 English placeholders using existing Latin kanban terminology. Preserve date queries, variables and URL schemes. Extend regression coverage for access boundaries and automatic archival semantics.
Translate 26 remaining delivery, sign-in and history messages. Retain 27 physical keyboard labels, product names and code symbols as locale-specific invariants. Include Haitian Creole in full key-order, token and placeholder coverage; older linguistic quality still needs review.
Fill 38 English placeholders while preserving existing translations. Extend coverage for cancellation boundaries, SMTP failure types, source activity preservation and delivery access checks.
Fill 38 English placeholders while preserving existing translations. Extend coverage for retention periods, null handling, report limitations and email retry safeguards.
Fill 53 English placeholders while preserving existing translations and tokens. Cover daily observations, sync conflicts and source previews. Extend safeguards coverage and update the README count after Haitian Creole crosses the existing completion threshold.
Fill 63 English placeholders while preserving existing translations and tokens. Cover sprint planning, workflow states and reports. Extend checks for unknown estimates, matching units and policies, and closing versus cancelling a sprint.
Fill 89 English placeholders while preserving existing translations. Extend coverage for whitespace, text indexing, replacement argument order, title case and navigation directions.
Fill 71 English placeholders for mathematics, variables, procedures and accessibility. Preserve existing translations and placeholders. Extend checks for numeric negation, angular units, return values and screen-reader state transitions.
Fill 82 English placeholders, preserving human translations and code tokens. Extend checks for Boolean negation, numeric interval endpoints, constants, angle units and statistical operations.
Fill 84 English placeholders while preserving existing translations. Cover keyboard navigation, list indexing and mutations, sorting, splitting and Boolean values. Extend placeholder and operation semantics regression checks.
Fill 116 English placeholders for control flow, bitmap accessibility, variables and input labels. Preserve existing translations and executable placeholders. Extend regression coverage for flow semantics, numeric ranges and argument order.
Fill 86 English placeholders covering saved filters, import warnings, map placement and Blockly controls and announcements. Preserve existing translations and printed keyboard and product names.
Translation and registry suites and 21 human-preference checks pass. Extend coverage for numbered argument order, movement directions, filter privacy, partial import warnings and map placement. English remains at 3783 keys and no locale is missing keys. Browser review was not run. Wider translation work continues.
Fill 65 English placeholders covering archive policy, date filters, board visibility, rule variables, notifications and due reminders. Preserve existing translations, template variables, filter expressions, URL schemes and markup.
Translation and registry suites and 21 human-preference checks pass. Add coverage for variable syntax, signed-in access, archive exclusions and reminder direction. Browser review was not run. Wider translation work continues.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, author access and permanent discard. All 246 locale paths now contain the 3783 current English keys.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Older English placeholders and broader language-quality work remain.
Fill 44 missing source keys for todo.txt imports and email recovery. Replace three English labels prefixed with Wolayttatto: for rules, confirmation and refresh. Preserve other existing values and verify tokens, import syntax, duplicate delivery, author access and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery in Tifinagh. Preserve existing translations and verify tokens, import syntax, duplicate delivery, author access and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, author access and permanent discard.
Vocabulary checked against the Mini Gharo English-Tigre dictionary. Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery in syllabics. Preserve existing translations and verify tokens, import syntax, duplicate delivery, author access and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, author access and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys and 11 older email recovery placeholders while preserving existing translations. Verify placeholder tokens, import syntax, duplicate delivery, author access and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation work continues.
Fill 910 missing and English placeholder values with standard Dutch wording, preserving existing Flemish translations. Include 44 new import and email recovery keys, use the local Klaar label, and clarify mathematical wording.
Recognize shared Dutch technical labels for Flemish and update the verified README count to 139. Translation completeness, placeholder, import syntax, delivery safeguard, registry and 21 human-preference checks pass. Browser review was not run. Wider translation work continues.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing translations and verify tokens, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
A rebase rewrote the commits these entries link to, so the links named commits
that no longer exist and would have 404ed on GitHub. Repointed by
releases/fix-changelog-hashes.sh, which only touches a link no ref in this clone
can reach.
Thanks to xet7 !
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Longer technical wording is low confidence and needs fluent review. Browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing keys in km and km_KH, also serving the existing km-KH alias. Preserve existing values and cover placeholders, import syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys in each Zulu locale for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys in each locale for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys in each locale for todo.txt imports and email recovery. Preserve existing values and test placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys in each locale for todo.txt imports and email recovery. Preserve existing translations and add coverage for placeholders, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Translate 44 missing keys into Kurmanji and add the same missing source entries to 11 regional English locales. Preserve every existing locale value and leave en.i18n.json unchanged.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys per locale for todo.txt imports and email recovery. Preserve every existing translation and cover placeholders, syntax, duplicate delivery, access checks and permanent discard.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source keys while preserving every existing locale value. Cover placeholder inventories, todo.txt syntax, duplicate-delivery warnings and access checks.
Translation and registry suites and 21 human-preference checks pass. Browser review was not run. Wider translation completion remains in progress.
Replace 25 Latin-script values and distinguish due dates, permanent deletion and archive restoration. Preserve source placeholders and add script and meaning regression coverage.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 25 Latin-script values with Arabic-script Uzbek. Preserve placeholders, ISO 8601, distinct calendar variants and next versus previous month labels.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 25 Latin-script values with Arabic-script Uzbek. Preserve restoration scope and the no-deletion assurance, distinguish reporting metrics and correct the bucket-list example.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 20 Latin-script values with Arabic-script Uzbek. Correct Watch and Collapse meanings, preserve the zoom range and distinguish desktop, mobile and calendar views.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 25 Latin-script values with Arabic-script Uzbek. Translate privacy labels, correct assignee wording and preserve placeholders, HTML emphasis and membership scopes.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 25 Latin-script values with Arabic-script Uzbek. Preserve placeholders and distinguish permanent attachment deletion from recoverable removal with retained files.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 25 Latin-script values with Arabic-script Uzbek. Preserve placeholders, archive scope, administrator restrictions and loading warnings; correct Apply and Add After List wording.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 25 Latin-script values with Arabic-script Uzbek. Correct attachment terminology and preserve placeholders, list positions, shortcut toggle directions and positive integer swimlane heights.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 25 Latin-script values with Arabic-script Uzbek. Preserve placeholders, Markdown, the one-board limit and Home removal without deletion; correct due-date terminology.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 20 Latin-script values with Arabic-script Uzbek. Preserve placeholders, completion distinctions and checklist, date and card argument roles verified against activity templates.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 20 Latin-script values with Arabic-script Uzbek. Preserve percent placeholder counts and positional item, source and destination roles. Verify the checklist card reference against its template caller.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 20 Latin-script values for card movement, custom fields, membership and archives. Preserve placeholders and movement origins and destinations; prevent Latin prose from returning in this batch.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Replace 20 Latin-script activity and membership values with Arabic-script Uzbek. Preserve placeholder inventories, distinguish organization from team membership, and prevent Latin prose from returning in this batch.
Translation and registry suites and 21 human-preference checks pass. Arabic-script wording is low confidence and needs fluent review. Browser review was not run; older script corrections and wider translation completion remain in progress.
Fill nine final placeholders per Uzbek locale and record 26 reviewed technical invariants. Add all four locales to full source key-order, token and placeholder checks.
Translation and registry suites and 21 human-preference checks pass. Source coverage is complete, but language quality is not: older Latin-script text in uz-AR remains to correct and new Arabic-script wording needs fluent review. Browser review was not run; wider translation completion remains in progress.
Fill 20 placeholders in each Uzbek locale. Preserve existing translations and tokens; distinguish pause, permanent cancellation and mail-server acceptance.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 20 placeholders in each Uzbek locale. Preserve existing translations and tokens; cover hours, explicit null, distinct recovery states and no activity recreation on retry.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 20 placeholders in each Uzbek locale. Preserve existing translations and tokens; cover temporary versus permanent rejection, cancellation boundaries and distinct delivery failures.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 20 placeholders in each Uzbek locale. Preserve existing translations and tokens; cover diagnostic retention, explicit null semantics, uncertain delivery and pause behavior.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 20 placeholders in each Uzbek locale. Preserve existing translations and tokens; cover retention, report limits, distinct outcomes and the absence of resume or undo behavior.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 20 placeholders in each Uzbek locale. Preserve existing translations and tokens; cover preview limits, distinct changes, unchanged subcards and reuse of replacement cards.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 20 placeholders in each Uzbek locale. Preserve existing translations and tokens; cover observation limits, unknown estimates and local-only conflict resolution.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 25 placeholders in each Uzbek locale. Preserve existing translations and format tokens; distinguish unknown estimates from zero, report scope and sprint lifecycle states.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 25 placeholders in each Uzbek locale. Distinguish results review from process retrospection and committed from completed work. Update the README coverage count to 138 as all four Uzbek locales exceed 90 percent translated strings.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 25 placeholders in each Uzbek locale. Preserve existing translations and format tokens; cover roles, estimate sources and units, completion policies and sprint actions.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 22 placeholders in each Uzbek locale. Preserve existing translations and format tokens; cover rule validation, administrator permissions, save conflicts and consistent Blockly terminology.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 19 placeholders in each Uzbek locale. Preserve existing translations and format tokens; cover block counts, comment fragments, search results and keyboard instructions.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 25 placeholders in each Uzbek locale. Preserve existing translations and format tokens; cover replacement operands, trimming directions, prompts and variable assignment.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 25 placeholders in each Uzbek locale. Preserve existing translations and format tokens; cover substring boundaries, search failure values and empty-text checks.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 25 placeholders in each Uzbek locale. Preserve existing translations and format tokens; cover scroll directions, letter case and character positions.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 25 placeholders in each Uzbek locale. Preserve existing translations and format tokens; cover opposite directions, navigation targets and move completion versus cancellation.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 30 placeholders in each Uzbek locale. Preserve existing translations and format tokens; cover function output distinctions and screen-reader toggle actions.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 24 placeholders in each Uzbek locale. Preserve existing translations, format tokens, distinctions between inverse functions and ordinary functions, and the use of degrees rather than radians.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 23 placeholders in each Uzbek locale, preserving existing translations and format tokens. Cover inclusive integer bounds, exclusive random-fraction upper bounds and distinct rounding directions.
Translation and registry suites and 21 human-preference checks pass. Wording needs fluent-speaker review, especially low-confidence Arabic-script orthography. Browser review was not run; wider translation completion remains in progress.
Fill 15 English placeholders in each of the four Uzbek locales, preserving token inventories and existing translations. Cover distinct statistical operations and consistent minimum and maximum labels.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 18 placeholders in each of four Uzbek locales, preserving mathematical constants, inclusive bounds, number properties and remainders.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 14 placeholders in each of four Uzbek locales, preserving conditional labels, arithmetic operations, coordinates and angle ranges.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 19 placeholders in each of four Uzbek locales, preserving equality, inclusive comparisons, negation and conjunction semantics.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 20 placeholders in each of four Uzbek locales, preserving insertion/replacement distinctions, sort direction, case handling and list/text conversion.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 20 placeholders in each of four Uzbek locales, preserving list boundaries, missing-item results and copy semantics.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 20 placeholders in each of four Uzbek locales, preserving read/remove semantics and sub-list boundary markers.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 25 placeholders in each of four Uzbek locales, preserving keyboard shortcuts, text operation distinctions and empty-list semantics.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 22 placeholders in each of four Uzbek locales, preserving numeric roles, coordinate names, repeated counts and text boundaries.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 24 placeholders in each of four Uzbek locales, preserving condition ordering, list boundaries, split/join distinctions and warning controls.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 20 placeholders in each of four Uzbek locales, preserving bitmap coordinates, keyboard boundaries and editor open/close distinctions.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 18 placeholders in each of four Uzbek locales, preserving while/until semantics, clipboard actions and variable deletion counts.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 25 placeholders in each of four Uzbek locales, preserving colour ranges, loop control semantics and conditional branch distinctions.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 23 placeholders in each of four Uzbek locales, preserving block state distinctions, singular/plural inputs and variable deletion warnings.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 24 placeholders in each of four Uzbek locales, preserving movement directions, add/remove actions, expand/collapse distinctions and degree placeholders.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 22 placeholders in each of four Uzbek locales, covering saved filter outcomes, partial imports, map placement and Blockly movement announcements.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 22 placeholders in each of four Uzbek locales, preserving reminder offset signs, duplicate dependency directions and template syntax.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 20 placeholders in each of four Uzbek locales, preserving signed-in access restrictions, rule variables, link scheme names and notification distinctions.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Fill 24 placeholders in each of four Uzbek locales. Preserve query syntax, inclusive date boundaries, archive exclusions and list-age semantics.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review, especially low-confidence Arabic-script orthography; browser review was not run. Wider translation completion remains in progress.
Add 44 import and rule-email recovery translations in Arabic script. Preserve todo.txt syntax and substitution tokens, and check that the new prose contains no Latin-script text.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Southern Uzbek wording and orthography need fluent-speaker review; browser review was not run. Older mixed-script entries and remaining placeholders still need correction.
Add 44 import and rule-email recovery translations to each of uz, uz-LA and uz-UZ. Preserve todo.txt syntax, substitution tokens and existing translations.
Translation and registry suites and 21 human-preference checks pass. Uzbek wording needs fluent-speaker review; browser review was not run. Other placeholders and the Arabic-script Uzbek locale remain for subsequent translation.
Translate 45 remaining prose placeholders and retain 25 physical key legends, operating-system brands and math symbols. Add Hausa to full source-order, placeholder and untranslated-prose regression checks.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Source coverage does not establish translation quality. Wider translation completion remains in progress.
Fill 35 English placeholders for queued delivery controls, failure diagnostics and Jira time estimates. Preserve cancellation scope, pause-aware retries and missing-versus-null behavior.
Translation and registry suites and 21 human-preference checks pass. Hausa technical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 49 English placeholders for sync conflict recovery, source omissions, previews and run diagnostics. Preserve missing-versus-null behavior, retention limits and existing-card protections.
Translation and registry suites and 21 human-preference checks pass. Hausa technical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 35 English placeholders while preserving report limitations, local/source choices and duplicate-card retention. Update the README count to 134 languages above the source coverage threshold.
Translation and registry suites and 21 human-preference checks pass. Hausa technical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 60 English placeholders for backlog planning, sprint lifecycle, estimates and reports. Preserve state distinctions and unknown-versus-zero estimate guidance.
Translation and registry suites and 21 human-preference checks pass. Hausa technical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 34 English placeholders for workspace announcements, search controls and rule editing. Preserve composition whitespace, substitution tokens and validation warnings.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 54 English placeholders for Blockly screen-reader mode, clipboard actions and keyboard navigation. Preserve direction, state and action distinctions and all substitution tokens.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders, preserving distinctions between comparisons, roots, logarithm bases and inverse trigonometric functions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Hausa mathematical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 25 Blockly placeholders while preserving existing translations, shortcut tokens and distinct text operations and comparison bounds.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 31 Blockly input placeholders while preserving existing translations, source tokens and distinct positions, operands and numeric bounds.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders for bitmap controls, editing and conditions while preserving existing translations and source tokens.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders for Blockly descriptions, warnings and clipboard controls while preserving existing translations and formatting tokens.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 29 English placeholders for movement announcements and accessible controls while preserving existing translations and formatting tokens.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 27 English placeholders for filters, import warnings, map controls and movement announcements. Preserve existing translations, template expressions and navigation labels.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders while preserving existing translations, rule variables and reminder offsets. Cover parent cards, rule actions and notification categories.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders for archiving, date filters, Leo import and signed-in access. Preserve source tokens, query syntax, URL schemes and existing translations.
Translation and registry suites and 21 human-preference checks pass. Hausa wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Translate the final 26 recovery and history messages and retain 22 reviewed technical labels. Include Igbo in full-file key order, token and missing-prose checks.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Source coverage is not language-quality certification. Wider translation completion remains in progress.
Fill 29 English placeholders while preserving existing translations, source tokens, time estimate units and activity recovery limits.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 26 English placeholders while preserving existing translations, source tokens, cancellation boundaries and delivery failure distinctions.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 27 English placeholders while preserving existing translations, source tokens, retention limits and distinct run outcomes.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders while preserving existing translations, source tokens, local content guarantees and preview limits.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 29 English placeholders while preserving existing translations, source tokens, observation limits and distinct close and cancel semantics.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 32 English placeholders while preserving existing translations, source tokens and distinct sprint states and estimate semantics.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 32 English placeholders and update the README count to 133 languages above 90 percent source coverage. Preserve existing translations and rule permission semantics.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders while preserving existing translations, announcement tokens, composed spacing and keyboard shortcuts.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 34 Blockly shortcut placeholders while preserving existing translations and distinct navigation directions, focus targets and move actions.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 21 Blockly placeholders while preserving existing translations, shortcut tokens and opposite screen reader mode actions.
Translation and registry suites and 21 human-preference checks pass. Igbo wording, especially mathematical descriptions, needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 20 Blockly arithmetic and constant placeholders while preserving existing translations and source tokens. Use descriptive wording for roots and inverse functions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Igbo mathematical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 26 Blockly placeholders while preserving existing translations, shortcut tokens and distinct text operations and comparison bounds.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 Blockly input placeholders while preserving existing translations, source tokens and distinct positions, operands and numeric bounds.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders for bitmap coordinates, editing controls and condition inputs while preserving existing translations and source tokens.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders for Blockly descriptions, warnings and backpack controls while preserving existing translations and formatting tokens.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 29 English placeholders for movement announcements and accessible controls while preserving existing translations and formatting tokens.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 29 English placeholders for filters, import warnings, map controls and movement announcements. Preserve existing translations, template expressions and navigation labels.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders while preserving existing translations, rule variables and reminder offsets. Cover parent cards, rule actions and notification categories.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders for archiving, date filters, Leo import and signed-in access. Preserve source tokens, query syntax, URL schemes and existing translations.
Translation and registry suites and 21 human-preference checks pass. Igbo wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Translate the final 20 recovery and history messages and retain 24 reviewed keyboard legends, OS brands and math symbols. Include Yoruba in full-file key order, token and missing-prose checks.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; older wording still needs a quality audit and browser review was not run. Wider translation completion remains in progress.
Fill 29 English placeholders while preserving existing translations, source tokens, recovery distinctions and imported time estimate rules.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 31 English placeholders while preserving existing translations, source tokens, cancellation limits and delivery failure distinctions.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 36 English placeholders for preview omissions, report retention, run outcomes and mapped estimates. Preserve existing translations and distinguish ignored missing values from explicit null.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 40 English placeholders while preserving existing translations, source tokens, observation limits and one-way Sync behavior.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 41 English placeholders for sprint planning, events, cancellation and reporting. Preserve tokens, existing translations and unknown-estimate semantics.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 31 English placeholders and update the README coverage count to 132 languages above 90 percent. Preserve existing translations and token inventories.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 26 English placeholders while preserving substitution tokens, keyboard shortcuts and spacing in composed announcements.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 32 English placeholders while preserving distinct navigation directions, focus targets and movement actions.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 19 English placeholders while preserving substitution tokens, keyboard names and distinct accessibility states.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 21 English placeholders while preserving function notation, logarithm bases, substitution tokens and distinct operators.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence mathematical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 26 English placeholders while preserving substitution tokens, comparison operators and distinct navigation actions.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders while preserving substitution tokens, coordinate names and distinct arithmetic and list positions.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 23 English placeholders while preserving substitution tokens, bitmap coordinates and distinct opening and closing actions.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 22 English placeholders while preserving substitution tokens, input plurality and distinct clipboard and backpack actions.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 25 English placeholders while preserving substitution tokens and distinct add, remove, collapse and expand controls.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders while preserving movement directions, map placement instructions and substitution tokens.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders while preserving reminder offsets, template variables, private filter behavior and recovery navigation labels.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders while preserving access restrictions, HTML tags, rule variable syntax and always-delivered reminders.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 25 English placeholders while preserving date query syntax, numeric limits and Leo import semantics.
Translation and registry suites and 21 human-preference checks pass. Yoruba wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Translate nine remaining English placeholders and register 28 retained keyboard legends, OS names and mathematical symbols. Check all 3783 Shona keys for source order, placeholder inventory and missing prose.
Translation and registry suites and 21 human-preference checks pass. Source coverage does not establish linguistic quality. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 27 English placeholders while preserving cancellation limits, retained work and mail-server acceptance semantics.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 25 English placeholders while preserving cancellation boundaries, distinct failure reasons and time estimate semantics.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 32 English placeholders while preserving tokens, null semantics, queue controls and delivery limitations.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 27 English placeholders while preserving report limits, retention periods, hidden values and partial run limitations.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders while preserving observation limitations, tokens and guarantees about local content and subcards.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders while preserving tokens, unknown estimate semantics and partial report limitations.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders for sprint actions, backlog and events. Update the README count to 131 locales above its 90 percent source coverage threshold.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders while preserving rule permissions, conflict recovery instructions and distinct planning roles.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders while preserving tokens, keyboard shortcuts and whitespace in composed announcements.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 34 English placeholders while preserving substitution tokens, whitespace semantics and distinct variable conflict messages.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 31 English placeholders while preserving substitution tokens, indexing directions and text lookup behavior.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders for keyboard navigation and text operations, preserving substitution tokens and keyboard names.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 32 English placeholders while preserving substitution tokens and distinct screen reader, movement and focus actions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders for workspace controls, variables, functions and angle instructions while preserving tokens and keyboard names.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 28 English placeholders using descriptive mathematical wording while preserving function notation and substitution tokens.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona mathematical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 30 English placeholders while preserving tokens, distinct statistical measures and inclusive or exclusive random bounds.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona mathematical wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 29 English placeholders, preserving substitution tokens, mathematical constants and inclusive bounds.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 29 English placeholders while preserving substitution tokens, null and comparison semantics.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 36 English placeholders for slicing, removal, insertion and sorting, preserving substitution tokens and copy semantics.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 33 English placeholders while preserving keyboard names, operating system names and substitution tokens.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 50 English placeholders for conditions, list positions, numeric inputs and text operations while preserving substitution tokens.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 45 English placeholders for conditionals, editing and bitmap accessibility. Preserve keyboard names and substitution tokens.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders and preserve keyboard names, substitution tokens and RGB bounds.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders while preserving existing translations. Cover singular and plural block descriptions, deletion restrictions, warnings and color blending ranges.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders while preserving existing translations. Cover field types, add versus remove actions, collapsed comments and exact positional tokens.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 24 English placeholders while preserving existing translations. Cover map placement choices, movement directions and exact positional tokens; keep the Alt key name unchanged.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 25 English placeholders while preserving existing translations. Cover access restrictions, rule variables, link schemes and notification exceptions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 25 English placeholders while preserving existing translations. Cover query syntax, inclusive date boundaries, list age and archive exclusions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Translate 61 remaining prefixed entries and restore S3 and Cron names. Cover report explanations, forecast limits, placeholders and shared labels. Add a whole-file guard against the prefixed English defect.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Ordinary English placeholders and broader translation work still remain.
Translate 39 prefixed English entries and restore the GridFS name. Preserve placeholders, CPU thresholds and time units while distinguishing migration actions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 63 prefixed English entries to replace.
Translate 40 prefixed English entries. Preserve placeholders, service names and region examples while distinguishing pausing from stopping and access keys from secret keys.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 103 prefixed English entries to replace.
Translate 34 prefixed English entries and restore six service names. Preserve placeholders and distinguish backup schedules, connection results and migration actions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 143 prefixed English entries to replace.
Translate 37 prefixed English entries and restore three technical names. Preserve placeholders and distinguish lockout timing, scheduled job states and migration warnings from errors.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 183 prefixed English entries to replace.
Translate 37 prefixed English entries and restore three storage names. Preserve placeholders and distinguish request states, transfer directions and memory measurements.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 223 prefixed English entries to replace.
Translate 37 prefixed English entries and restore three abbreviations. Preserve placeholders, shared report labels and distinct waiting indicators.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 263 prefixed English entries to replace.
Translate 37 prefixed English entries and restore three abbreviations or font names. Preserve placeholders, single-token search terms and syntax examples.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 303 prefixed English entries to replace.
Translate 30 prefixed English entries and restore ten protocol names and syntax tokens. Preserve placeholders, configuration formats and shared labels.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 343 prefixed English entries to replace.
Translate 40 prefixed English entries. Preserve placeholders, weekday bounds and comma-separated examples while keeping checked and unchecked actions distinct.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 383 prefixed English entries to replace.
Translate 38 prefixed English entries and restore two product names. Preserve configuration identifiers, placeholders and the distinction between free and total memory.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 423 prefixed English entries to replace.
Translate 38 prefixed English entries and restore two storage unit abbreviations. Cover invitation tokens, blocked versus unlimited uploads, logo links and protocol names; preserve unrelated values.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 463 prefixed English entries to replace.
Translate 32 prefixed English entries and restore three sort labels. Cover matching preview labels, import tokens, selection controls and exact placeholders; preserve unrelated values.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 503 prefixed English entries to replace.
Translate 42 prefixed English entries and restore three literal date formats. Preserve placeholders and distinct color choices. Color vocabulary reference: https://wisc.pb.unizin.org/lctlresources/chapter/color-and-clothing/
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording, especially shade descriptions, needs fluent-speaker review; browser review was not run. There are still 538 prefixed English entries to replace.
Translate 54 entries and restore 11 numeric, symbolic and named values. Cover voting choices, font sizes, zoom limits and exact placeholders; preserve all unrelated values.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. There are still 583 prefixed English entries to replace, and wider translation work remains in progress.
Fill 44 missing source strings while preserving existing values, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Shona wording needs fluent-speaker review; browser review was not run. Audit found 648 older prefixed English entries still needing replacement. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Lower-confidence Turkmen wording needs fluent-speaker review; browser review was not run. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings in each locale while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings in each locale while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source strings in each locale while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Fill 44 missing source additions while preserving existing translations, placeholders and todo.txt syntax. Cover uncertain delivery, duplicate delivery warnings and legacy email review actions.
Translation and registry suites and 21 human-preference checks pass. Browser and fluent-speaker review were not run. Wider translation completion remains in progress.
Replace mixed-language search instructions, visibility predicates, report titles and login wording while preserving search examples and placeholders.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 16 mixed-language search and view strings with Romansh. Preserve unrelated translations, key order and tokens. Cover exact operator syntax, shared labels and incomplete due-card visibility.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 20 mixed-language notification, visibility and board strings with Romansh. Preserve unrelated translations, key order and tokens. Cover private-only visibility, new versus all cards, irreversible deletion and Butler import limits.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 16 mixed-language import, list and permission strings with Romansh. Preserve unrelated translations, key order and tokens. Cover API paths, JSON fields, archive navigation and irreversible label deletion.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 23 mixed-language board-view, role and import strings with Romansh. Preserve unrelated translations, key order and tokens. Cover permission restrictions, opposite star actions and exact Kanboard/Deck field names.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Correct 18 language and meaning errors. Replace the export description incorrectly used for account anonymization with the actual irreversible account action, disabled login and retained history. Preserve unrelated translations, key order and tokens; cover import/export restrictions and migration scope.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 12 mixed-language cloud, avatar and migration strings with Romansh. Restore the client_email identifier and external navigation labels. Preserve unrelated translations, key order and tokens; cover avatar exclusions and migration field names.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 12 mixed-language storage, checklist and access strings with Romansh. Preserve unrelated translations, key order and tokens. Cover compaction timing, file formats, shared labels and signed-in access.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 15 mixed-language search, checklist, permission and attachment strings with Romansh. Preserve unrelated translations, key order and tokens. Cover search examples, global-admin rights, checklist order and storage names.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 20 mixed-language filter, import, display and settings strings with Romansh. Restore exact advanced-filter examples and escaping. Preserve unrelated translations, key order and tokens; cover shared labels and technical identifiers.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Correct 12 language and meaning errors, including five chart explanations that omitted source conditions. Preserve unrelated translations, key order and tokens. Cover forecast limits, missing data, observation thresholds and technical names.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 20 mixed-language confirmation and status strings with Romansh. Correct cannot-interrupt wording to cannot-undo where required by English. Preserve unrelated translations, key order and tokens; cover deletion limits and removal scope.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 16 mixed-language settings and status strings with Romansh. Bring the stale card-loading description into line with automatic loading and both environment variables. Preserve unrelated translations, key order and tokens; cover privacy wording and technical identifiers.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 20 mixed-language access, import and rule strings with Romansh. Preserve unrelated translations, key order and tokens. Cover public-board editing, member removal, account deletion and technical identifiers.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 20 mixed-language user, editing and permission strings with Romansh. Preserve unrelated translations, key order and tokens. Cover shared labels, Home removal, imported-member permission limits and card-link restrictions.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 20 mixed-language status, maintenance, storage and import strings with Romansh. Preserve unrelated translations, key order and tokens. Cover shared labels, team membership guards, problem counts and technical names.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 24 mixed-language sharing, subtask, sorting and import strings with Romansh. Preserve unrelated translations, key order and tokens. Cover shared labels, avatar defaults, template visibility and JSON/CSV format names.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 31 mixed-language import, upload, template and loading strings with Romansh. Preserve unrelated translations, key order and tokens. Cover the loading data-loss warning, shared template labels and import format names.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 30 Italian or mixed-language card, date, list, swimlane and settings strings with Romansh. Preserve unrelated translations, source key order and replacement tokens. Cover native vocabulary and irreversible deletion warnings.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Replace 31 Italian or mixed-language board, membership, display and deletion strings with Romansh. Preserve unrelated translations, source key order and replacement tokens. Cover native vocabulary, shared labels and permanent-deletion warnings.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Translate 13 remaining recovery, history and Blockly strings. Recognize 21 additional shared words, brands and mathematical symbols. Verify all 3783 source keys, their order and replacement tokens, and zero unreviewed English placeholders.
Translation and registry suites and 21 human-preference checks pass. Low-confidence wording needs fluent-speaker review; browser review was not run. Older Italian and mixed-language Romansh values remain to be corrected, so source coverage does not establish language quality. Wider translation work remains in progress.
Fill 25 email-resolution and legacy-review placeholders, preserving existing translations and replacement tokens. Cover uncertain delivery, access checks, distinct rebind and discard actions, and permanent discard warnings.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Fill 25 activity-recovery and rule-email placeholders while preserving existing translations and replacement tokens. Cover recovery states, shared pause and resume labels, irreversible cancellation and unchanged original activities.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Fill 25 delivery-failure, estimate and activity-recovery placeholders, preserving existing translations and replacement tokens. Cover permanent cancellation, preservation of future messages, SMTP failure types and retry semantics.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Fill 30 preview and report placeholders while preserving existing translations and replacement tokens. Cover report limits, retention, omitted values and partial-run warnings. Refresh the README coverage count to 130 after Romansh crosses its threshold.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. The coverage threshold does not establish full translation quality. Wider translation and wrong-language cleanup remain in progress.
Fill 25 daily-observation and synchronization strings, preserving existing translations and tokens. Cover UTC observations, unknown estimates, export scope, no source writes, and retention of local cards and subcards.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Fill 31 report, state and confirmation placeholders while preserving existing translations and replacement tokens. Cover unknown versus zero estimates, comparable units and policies, partial reports, and unfinished-card handling.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Fill 46 planning, sprint, backlog and event strings while preserving existing translations and replacement tokens. Cover shared view labels, distinct sprint actions and completion policies, and timebox units.
Translation and registry suites and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Fill 27 placeholders and correct three report-navigation labels. Recognize the correctly translated shared word Problems as a Romansh invariant. Preserve existing translations and tokens, and verify report navigation and keyboard direction.
Translation, registry and 21 human-preference checks pass. Low-confidence Romansh wording needs fluent-speaker review; browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Fill 42 English placeholders while preserving existing translations and template variables. Cover trigger ordering, reminder timing and limits, dependency direction, and notification terminology.
Translation and language-registration tests pass, along with 21 human-preference checks. Romansh wording has low confidence and needs fluent-speaker review. Browser review was not run. Wider translation and wrong-language cleanup remain in progress.
Fill 30 date-filter, import and visibility placeholders. Replace 16 Italian or mixed-language archive messages with Romansh, preserving other existing translations and all replacement tokens.
Translation and language-registration tests pass, along with 21 human-preference checks. These Romansh translations have low confidence and need fluent-speaker review. Browser review was not run; broader translation and wrong-language cleanup remain in progress.
Translate 64 remaining strings in each of fy and fy-NL, preserving existing translations and replacement tokens. Cover complete source keys, recovery warnings, selection direction and history actions.
Translation and language-registration tests pass, as do 21 human-preference checks. Frisian wording needs fluent-speaker review; browser review was not run. The wider translation work remains in progress.
Fill 25 English activity recovery and time estimate strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including matching delivery controls and distinct recovery states. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English email queue, delivery failure and time estimate strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct queue actions and SMTP failure types. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 18 English synchronization diagnostic and email queue strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including retention limits and technical literals. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English source-field and synchronization report strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including report limits, consistent source labels and distinct outcomes. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English synchronization conflict and preview strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct local/source choices and preview actions. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English workflow and observation strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including UTC references, observation limits and distinct sprint confirmations. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 30 English sprint report and event strings in each of fy and fy-NL. Update the README coverage count to 129. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including report placeholders and distinct sprint states. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 30 English Scrum planning labels in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including consistent backlog labels and distinct sprint actions. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Translate 22 keyboard and editor labels in each of fy and fy-NL and recognize 17 unchanged brands, symbols and Frisian terms per locale. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including keyboard directions and locale-specific invariants. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 11 English import-report and map-view strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including matching admin navigation and distinct map actions. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 22 English notification, reminder, dependency and saved-filter strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including template variables, reminder bounds and distinct actions. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 20 English rule and notification strings in each of fy and fy-NL. Preserve existing translations, replacement tokens and rule variables. Translation, registry and human-preference checks pass, including distinct trigger, action and recipient labels. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 12 missing or English strings in each of fy and fy-NL for imports, list-age filters and board visibility. Preserve existing translations, replacement tokens, import syntax, URL schemes and HTML tags. Translation, registry and human-preference checks pass. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 18 English date-filter and automatic archive strings in each of fy and fy-NL. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including numerical limits and date range direction. Frisian wording needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Translate the final 17 missing or English recovery, login and history strings. Add Friulian to full source-key order, placeholder and untranslated-prose regression coverage. All 3783 current source keys are present with no flagged prose remaining. Translation, registry and human-preference checks pass. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation and language-quality work remains in progress.
Fill 22 missing or English email resolution and legacy recovery strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct recovery confirmations. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 missing or English notification cancellation and rule email recovery strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct recovery actions and recipient states. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English activity recovery and time estimate strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including matching delivery controls and distinct recovery states. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English email queue, delivery failure and time estimate strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct queue actions and SMTP failure types. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 18 English synchronization diagnostic and email queue strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including retention limits and technical literals. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English source-field and synchronization report strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including report limits, consistent source labels and distinct outcomes. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English synchronization conflict and preview strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct local/source choices and preview actions. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English workflow and observation strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including UTC references, observation limits and distinct sprint confirmations. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 30 English Scrum report and event strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including report counts and distinct sprint states. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 30 English Scrum planning labels and update the README coverage count to 127. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including consistent backlog labels and distinct sprint actions. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Translate 15 keyboard labels and recognize 15 unchanged brands, symbols and Friulian terms. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including keyboard directions and locale-specific invariants. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 11 missing or English import-report and map-view strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including matching admin navigation and distinct map actions. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 22 missing or English notification, reminder, dependency and saved-filter strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including template variables, reminder bounds and distinct actions. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 20 missing or English rule and notification strings. Preserve existing translations and replacement tokens, including rule variables. Translation, registry and human-preference checks pass, including distinct trigger, action and recipient labels. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 12 missing or English strings for imports, list-age filters and board visibility. Preserve existing translations, replacement tokens, import syntax, URL schemes and HTML tags. Translation, registry and human-preference checks pass. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 18 missing or English date-filter and automatic archive strings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including numerical limits and range direction. Consult ARLeF language materials for time vocabulary. Friulian wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Translate the final 13 missing or English recovery, login and history strings. Add Faroese to full source-key order, placeholder and untranslated-prose regression coverage. All 3783 current English keys are present, with no flagged prose remaining. Translation, registry and human-preference checks pass. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation and language-quality work remains in progress.
Fill 22 missing or English strings for rule email resolution and legacy delivery recovery. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct rebind and discard actions. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 20 English placeholders for rule email recovery decisions and confirmations. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct recovery actions and recipient states. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 20 English placeholders for activity notification controls and rule email delivery. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including consistent controls and distinct delivery states. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English placeholders for delivery failures, time estimates and activity recovery. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including technical literals and distinct recovery states. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English placeholders for mapped estimates and email queue recovery. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including technical literals and distinct queue actions. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English placeholders for source-field reports and synchronization diagnostics. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including retention limits, source-label consistency and distinct outcomes. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 25 English placeholders for synchronization conflicts and preview messages. Update the README coverage count to 126 as Faroese crosses its threshold. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 30 English placeholders for Scrum observations, workflow categories and synchronization conflicts. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including UTC, observation limits and action distinctions. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 40 English placeholders for sprint planning, events, states and reports. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including distinct completion states and consistent backlog labels. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 37 English placeholders and recognize 13 Faroese invariants for brands, math notation, Alt and the imperative set. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including navigation direction and invariant coverage. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 19 English placeholders for saved filters, import warnings and map placement. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, with checks for distinct map and card labels and matching admin navigation. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 16 English placeholders while preserving existing translations and template variables. Add regression checks for reminder bounds, dependency direction and URL interpolation. Translation, registry and human-preference checks pass. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 18 missing or English strings while preserving existing translations and replacement tokens. Check literal rule variables and trigger/action labels. Translation, registry and human-preference checks pass. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 12 missing or English strings while preserving existing translations and replacement tokens. Check literal filter and todo.txt syntax. Translation, registry and human-preference checks pass. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Fill 18 missing or English strings while preserving existing translations and replacement tokens. Check date ranges and recency values. Translation, registry and human-preference checks pass. Faroese wording has lower confidence and needs fluent-speaker review; browser review was not run. Wider translation work remains in progress.
Translate the final 22 missing recovery strings and retain 19 technical terms and Scrum loanwords. Include Luxembourgish in full source-order, placeholder and untranslated-prose regression checks. All 3783 source keys are present. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check distinct resend, mark-as-sent and discard confirmations. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check delivery confirmation and attempt states. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 21 missing or English strings while preserving existing translations and replacement tokens. Check distinct delivery actions and missing/changed activity states. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check temporary/permanent failures, time-estimate distinctions and null token. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check distinct delivery actions and busy/failure messages. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 17 missing or English strings while preserving existing translations and replacement tokens. Check retention periods, null token and distinct run states. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check matching source and preview labels and report limits. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check local/source and keep/detach choices. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 15 missing or English strings while preserving existing translations and replacement tokens. Check UTC wording, observation limits and unknown versus zero estimates. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 24 missing or English strings while preserving existing translations and replacement tokens. Check distinct sprint states and close/cancel confirmations. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 27 missing or English strings while preserving existing translations and replacement tokens. Check backlog consistency and distinct sprint actions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 24 missing or English strings while preserving existing translations and replacement tokens. Check saved states and estimate fields. Update README coverage after Luxembourgish crosses 90 percent. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 18 missing or English strings while preserving existing translations and replacement tokens. Check search directions and keyboard names. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 21 missing or English strings while preserving existing translations and replacement tokens. Check trim directions, prompt types and get/set actions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check first/last, start/end and letter-case distinctions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 25 missing or English strings while preserving existing translations and replacement tokens. Check distinct directions, next/previous stacks and top/bottom positions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 21 missing or English strings while preserving existing translations and replacement tokens. Check enabled/disabled, abort/finish and focus-target distinctions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 22 missing or English strings while preserving existing translations and replacement tokens. Check return/no-return and navigation distinctions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 13 missing or English strings while preserving existing translations and replacement tokens. Check inverse-function distinctions and degree versus radian wording. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check logarithm and exponent distinctions and random-number bounds. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check mathematical symbols and distinct median/mode and square-root labels. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 17 missing or English strings while preserving existing translations and replacement tokens. Check comparisons, AND/OR distinctions and conditional-field references. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 20 missing or English strings while preserving existing translations and replacement tokens. Check distinct sorting directions, text conversions and comparison operators. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 15 missing or English strings while preserving existing translations and replacement tokens. Check first/last and start/end distinctions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 19 missing or English strings while preserving existing translations and replacement tokens. Check distinct retrieval and removal descriptions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 30 missing or English strings while preserving existing translations and replacement tokens. Check distinct dividend/divisor, minimum/maximum and start/end labels. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 30 missing or English strings while preserving existing translations and replacement tokens. Check distinct warning actions, conditions and list positions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 25 missing or English strings while preserving existing translations and replacement tokens. Check distinct conditional descriptions, copy actions and keyboard End. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 29 missing or English strings while preserving existing translations and replacement tokens. Check distinct break/continue and singular/plural input labels. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 29 missing or English strings for map placement and block accessibility. Preserve existing translations and replacement tokens. Verify distinct movement, expand/collapse and add/remove announcements. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 25 missing or English strings for reminders, presets, imports and maps. Preserve existing translations, replacement tokens and template variables. Verify recovery navigation labels. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 30 missing or English strings for visibility, rules and notifications. Preserve existing translations, replacement tokens, rule variables and HTML markup. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Check literal filter syntax and todo.txt markers. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Translate the final nine missing recovery strings and retain nine technical key, OS and math labels. Include Maltese in full source-order, placeholder and untranslated-prose regression checks. All 3783 source keys are present. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 35 missing or English strings while preserving existing translations and replacement tokens. Check distinct resolution actions and legacy confirmations. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 35 English placeholders while preserving existing translations and replacement tokens. Check distinct delivery actions and confirmation states. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Check failure categories and original versus remaining time. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 36 English placeholders while preserving existing translations and replacement tokens. Check retention periods, null semantics and distinct delivery actions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 56 English placeholders while preserving existing translations and replacement tokens. Cover source choices, matching labels and UTC observations. Update the README coverage count after Maltese crosses 90 percent. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 66 English placeholders while preserving existing translations and replacement tokens. Check consistent backlog labels and distinct sprint states and confirmations. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 27 English placeholders while preserving existing translations and replacement tokens. Cover search direction and saved-state distinctions. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains in progress.
Fill 24 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 21 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 23 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 26 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 24 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including movement directions and comment controls. Fluent-speaker and browser review were not run; wider translation work remains.
Replace the final 22 prefixed English values with Maltese and prevent this pattern from recurring. Translation, registry and human-preference checks pass, including forecast limits and time-adjustment caveats. Other English placeholders remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 22 prefixed English values remain alongside other placeholders; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 57 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 92 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 127 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 162 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 197 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 232 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 267 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 302 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 337 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass, including search-token formatting. Another 372 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 407 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 442 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 477 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens and configuration identifiers. Translation, registry and human-preference checks pass. Another 512 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 547 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 582 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 617 prefixed English values remain; fluent-speaker and browser review were not run.
Replace 35 prefixed English values with Maltese while preserving replacement tokens. Translation, registry and human-preference checks pass. Another 652 prefixed English values remain; fluent-speaker and browser review were not run.
Translate 35 incorrectly seeded labels for navigation, calendars and voting. Preserve replacement tokens and existing correct-language translations. Translation, registry and human-preference checks pass. Another 687 prefixed English values remain; fluent-speaker and browser review were not run.
Fill 29 English placeholders and replace two prefixed English menu labels with Maltese. Match import report navigation to actual menu labels. Translation, registry and token checks pass. Another 722 prefixed English values remain to translate; fluent-speaker and browser review were not run.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including rule variables and reminder limits. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including import syntax and visibility markup. Fluent-speaker and browser review were not run; wider translation work remains.
Fill the final 21 prose placeholders and retain standard keyboard labels, OS brands and trigonometric symbols. Add Tagalog to full source-key and token coverage checks. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation and language-quality work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including duplicate delivery warnings and uncertain resend outcomes. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including cancellation limits, missing activities and mail-server acceptance. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including cancellation limits, estimate fields and temporary versus permanent failures. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including retention periods, null handling and delivery recall limits. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Update the README count to 123 languages above the 90 percent threshold. Translation, registry and human-preference checks pass, including preserved subcards and local versus source choices. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including unknown estimates, observation limits and sync direction. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including unfinished work, time units and distinct sprint actions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including rule connection requirements and administrator permissions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 33 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including search navigation and empty workspace descriptions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including text indexing, replacement and length guidance. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including append semantics, character positions and case conversion. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including movement versus scrolling, navigation endpoints and detailed announcements. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including return values, function-only restrictions and screen-reader toggles. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including angle units, function outputs and disabled definitions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 21 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including inverse functions, logarithm bases and negation. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including random bounds, modes and rounding directions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including inclusive limits, coordinate ranges and distinct numeric concepts. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 24 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including comparison boundaries, Boolean conditions and matching conditional labels. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including copy semantics, sorting direction and keyboard navigation. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including division operands, coordinates and consistent positions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including warning controls, input positions and repetition counts. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including deletion counts, pixel coordinates and comment controls. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 24 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including deletion warnings, input counts and conditional labels. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including angle units, comment expansion and keyboard labels. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 24 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including map placement, movement order and scroll directions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including reminder offsets, saved-filter replacement and matching recovery menu labels. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including board access, rule variables and notification exceptions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations, replacement tokens and import syntax. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Translate 29 missing recovery, sign-in and history strings. Retain standard physical key names, OS brands and trigonometric symbols as locale invariants. Swahili now has all 3783 source keys and no flagged English prose. Full key order, replacement-token, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation and language-quality work remains.
Fill 20 missing English-source strings while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including duplicate-delivery warnings and uncertain outcomes. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including permanent cancellation and mail-server acceptance. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including non-recreation of activities, retained pending work and missing versus null source values. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including cancellation boundaries, temporary versus permanent failures and hour units. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including null versus missing values, repeated delivery and pause retention. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including report limits, retention and no-resume/no-undo guidance. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including local content retention, unchanged subcards and replacement reuse. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders and refresh the README coverage count after Swahili passes 90 percent. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including UTC observations, missing days and unknown estimates. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including unfinished work, time units and consistent backlog labels. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including permissions, rule conflicts and distinct completion policies. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including keyboard shortcuts, unsaved changes and one-trigger/one-action guidance. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 34 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including text replacement scope, spaces, trim directions and workspace counts. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 39 English placeholders while preserving existing translations and replacement tokens. Translation, registry and human-preference checks pass, including text positions, append behavior and missing-match guidance. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 44 English placeholders for screen-reader modes, navigation and editor shortcuts. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including mode toggles and movement versus scrolling. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 34 English placeholders for procedures, inputs and editor actions. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including output distinctions and disabled-function warnings. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 34 English placeholders while preserving existing translations and replacement tokens. Check inverse functions and degree versus radian guidance. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 39 English placeholders for number properties, statistics, rounding and random values. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including random range boundaries. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 15 English placeholders while preserving existing translations, formulas and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for comparisons, Boolean operations and conditional values. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for list sorting, text splitting and comparisons. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for list repetition, reversal, insertion and value setting. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for list retrieval, removal and sublists. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for keyboard navigation, list creation and retrieval. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for text inputs and keyboard navigation. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for numeric, list and loop inputs. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for list inputs, conditions and warning controls. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for editor fields, pixel labels and comment controls. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for editor actions and deletion prompts. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for loop and conditional guidance. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for color mixing, block controls and loop flow. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 24 English placeholders for block labels, keyboard names and editor warnings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for editor actions and field types. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 19 English placeholders for map placement and editor movement announcements. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for saved filters, import warnings and map controls. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for activity notifications and due date reminders. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for filters, imports, board visibility and rule controls. Preserve existing translations, syntax examples and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for automatic archiving and date filters. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Translate the final 29 flagged Georgian prose entries and retain ten standard key labels, OS names and math symbols. Add Georgian to full key-order, token and untranslated-prose checks. Translation, registry and human-preference checks pass against 3783 English keys. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for rule email recovery decisions and outcomes. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for notification cancellation and rule email reports. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for activity notification recovery and mapped time estimates. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for email failures, queue recovery and time estimates. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders for synchronization diagnostics and email queue controls. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders for source omissions and synchronization reports. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders for synchronization conflict resolution and previews. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders for workflow categories, daily observations and synchronization conflicts. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders for sprint reports, estimates and event outcomes. Update the README coverage count to 121. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders for Scrum roles, estimates, completion policies and sprint controls. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders for workspace search, rule editing and planning view labels. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 24 English placeholders for variables, workspace summaries and search. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders for text searching, replacement, prompts and trimming. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 34 English placeholders for scrolling, text casing, character positions and substrings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders for screen reader modes and editor shortcuts. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 34 English placeholders for functions, variable renaming and editor controls. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 47 English placeholders for rounding, trigonometry, variable creation and editor navigation. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders for numeric bounds, divisibility, remainders and list statistics. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 41 English placeholders for Boolean values, comparisons, conditional expressions, arithmetic and constants. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 37 English placeholders for list indexing, repetition, sorting, insertion, replacement and splitting. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 21 English placeholders for list retrieval, removal and sublist boundaries. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including checks distinguishing reading from removal and indexing from either end. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 21 English placeholders for keyboard navigation and list creation. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including movement confirmation and empty lists. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 32 English placeholders for numeric, loop and text inputs. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including division operands, coordinates and repeat-count consistency. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 27 English placeholders for keyboard help, editor controls and list inputs. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including input layout and repeat counts. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 27 English placeholders for deletion, editing and bitmap fields. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including deletion counts, pixel coordinates and keyboard End semantics. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 22 English placeholders for conditional branches, loops and copy actions. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including while/until conditions and fallback branches. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 27 English placeholders for editor actions, color controls and loop flow. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including color limits and break versus continue. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 34 English placeholders for accessible field types, block structure and variable deletion guidance. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 29 English placeholders for map placement, movement announcements and accessible controls. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders for reminders, saved filters and import reports. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including reminder offsets, template syntax and localized report navigation. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders for board access, rule variables and notification preferences. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including permissions, variable syntax and action order. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders while preserving existing translations and replacement tokens. Cover archive exclusions, date ranges, list age and import instructions. Translation, registry and human-preference checks pass, including query and todo.txt syntax. Fluent-speaker and browser review were not run; wider translation work remains.
Translate 28 missing or English recovery and history strings in each Azerbaijani locale. Retain standard key labels, OS brands, short trig symbols and Sprint as locale-specific invariants. Verify all 3783 source keys and replacement tokens with no flagged English prose, including pending keys. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; older vocabulary review and wider translation work remain.
Fill 30 missing or English strings in each Azerbaijani locale for rule email review and resending. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including mail-server acceptance and duplicate delivery warnings. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders in each Azerbaijani locale for activity notification recovery and rule email reports. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including non-recreation, cancellation and retained pending work. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders in each Azerbaijani locale for email queue recovery and synchronized time estimates. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including cancellation, uncertain delivery and paused retries. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders in each Azerbaijani locale for synchronization reports and diagnostics. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including retention, recovery limits and missing versus null values. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders in each Azerbaijani locale for sync previews, duplicate recovery and source omissions. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 25 English placeholders in each Azerbaijani locale for daily observations, partial reports and synchronization conflicts. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders in each Azerbaijani locale for sprint events, reports and completion states. Preserve existing translations and replacement tokens. Update the README count to 120 locales above the translation coverage threshold. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders in each Azerbaijani locale for Scrum settings, sprint actions and backlog planning. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 31 English placeholders in each Azerbaijani locale for workspace announcements, search and rule editing. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including shortcuts, comment fragments and rule constraints. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 36 English placeholders in each Azerbaijani locale for navigation shortcuts, focus controls and parameter warnings. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including movement versus scrolling and navigation directions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 38 English placeholders in each Azerbaijani locale for trigonometry, screen reader controls and editor actions. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders in each Azerbaijani locale for keyboard navigation, comparisons and mathematics. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 42 English placeholders in each Azerbaijani locale for list, loop, math and text inputs. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including division operands, coordinates and start/end distinctions. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 37 English placeholders in each Azerbaijani locale for editor fields, keyboard labels, backpack actions and comment controls. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including pixel coordinates and keyboard End semantics. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 38 English placeholders in each Azerbaijani locale for accessible fields, block structure and keyboard labels. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 34 English placeholders in each Azerbaijani locale for saved filters, import reports, map placement and screen-reader announcements. Preserve existing translations and replacement tokens. Translation, registry and human-preference checks pass, including movement directions and private filter replacement. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders in each Azerbaijani locale, preserving existing translations and replacement tokens. Cover rule variables, trigger order, notification preferences and reminder offsets. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 30 English placeholders in each of the three Azerbaijani locale files, preserving existing translations and replacement tokens. Cover archive exclusions, list age, board visibility, date query syntax and todo.txt imports. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Translate 18 recovery, SAML and history strings. Retain standard physical key labels, operating system brands and short trigonometric symbols as locale-specific invariants. Verify all 3783 source keys, replacement tokens and absence of flagged English prose, including pending keys. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; older vocabulary review and wider translation work remain.
Fill 35 English placeholders for manual email resolution and legacy email review. Preserve existing translations and replacement tokens. Cover duplicate delivery, uncertain resends and current-card preparation. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders for activity notification recovery and rule email reports. Preserve existing translations and replacement tokens. Cover non-recreation, irreversible cancellation, retained pending work and mail-server acceptance. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 35 English placeholders for email delivery recovery and synchronized time estimates. Preserve existing translations and replacement tokens. Cover irreversible cancellation, uncertain delivery, paused retries and null source values. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 51 English placeholders while preserving existing translations and replacement tokens. Cover conflict recovery, preview omissions, report retention and missing versus null source values. Translation, registry and human-preference checks pass. Fluent-speaker and browser review were not run; wider translation work remains.
Fill 20 English placeholders while preserving existing translations and replacement tokens. Cover observation limits, incomplete imports and local synchronization conflict resolution. Translation, language registry and human-preference checks pass. Fluent-speaker and browser review were not run; other translation work remains.
Fill 30 English placeholders while preserving existing translations. Cover unknown estimates, partial reports, cancellation behavior and distinct sprint states. Fluent-speaker and browser review were not run; other translation work remains.
Fill 40 English placeholders while preserving existing translations. Cover consistent names, planning states, time units and opposite actions. Update the documented coverage count to 117 languages. Fluent-speaker and browser review were not run; other translation work remains.
Fill 25 English placeholders while preserving existing translations. Cover search shortcuts, unsaved changes, administrator permissions and rule cardinality. Fluent-speaker and browser review were not run; other translation work remains.
Fill 30 English placeholders while preserving existing translations. Cover navigation directions, next and previous commands, and composed workspace announcements. Fluent-speaker and browser review were not run; other translation work remains.
Fill 35 English placeholders while preserving existing translations. Cover screen reader state transitions, disabled functions and opposite navigation commands. Fluent-speaker and browser review were not run; other translation work remains.
Fill 34 English placeholders while preserving existing translations. Cover inclusive comparisons, inverse functions, arithmetic roles and opposite actions. Fluent-speaker and browser review were not run; other translation work remains.
Fill 31 English placeholders while preserving existing translations. Cover coordinate labels, text positions, copy and cut feedback and movement confirmation keys. Fluent-speaker and browser review were not run; other translation work remains.
Fill 35 English placeholders while preserving existing translations. Cover opening and closing controls, start and end positions, repeat counts and arithmetic input roles. Fluent-speaker and browser review were not run; other translation work remains.
Fill 37 English placeholders while preserving existing translations. Cover coordinate order, copy and cut, keyboard controls and distinct backpack actions. Fluent-speaker and browser review were not run; other translation work remains.
Fill 39 English placeholders while preserving existing translations. Cover movement directions, add and remove actions, angle values and replacement tokens. Fluent-speaker and browser review were not run; other translation work remains.
Fill 27 English placeholders while preserving existing translations. Cover template variables, saved filter replacement, map placement and import report navigation. Fluent-speaker and browser review were not run; other translation work remains.
Fill 32 English placeholders while preserving existing translations. Cover rule variables, reminder timing, notification exceptions and URL scheme names. Fluent-speaker and browser review were not run; other translation work remains.
Fill 28 missing entries and English placeholders while preserving existing translations. Cover date range boundaries, source syntax, activity age and signed-in access. Fluent-speaker and browser review were not run; other translation work remains.
Translate the remaining 35 prose placeholders and extend the existing Albanian invariant list with 12 unchanged key names, product names and mathematical terms. Check all source keys and tokens plus cancellation scope and retry behavior. No flagged English prose placeholders remain; fluent-speaker and browser review were not run.
Fill 43 English placeholders while preserving existing translations. Cover cancellation scope, retry behavior, temporary and permanent failures, and activity preservation. Fluent-speaker and browser review were not run; other translation work remains.
Fill 40 English placeholders while preserving existing translations. Cover report retention, partial changes, omitted values, explicit null and email retry scope. Fluent-speaker and browser review were not run; other translation work remains.
Fill 48 English placeholders while preserving existing translations. Cover partial reports, observation limits, unknown estimates and synchronization actions that preserve local cards. Fluent-speaker and browser review were not run; other translation work remains.
Fill 60 English placeholders while preserving existing translations. Cover unknown estimates, planning states and opposite actions. Update the documented coverage count to 116 languages. Fluent-speaker and browser review were not run; other translation work remains.
Fill 58 English placeholders while preserving existing translations. Check navigation directions, search key combinations, composed announcements and rule editing restrictions. Fluent-speaker and browser review were not run; other translation work remains.
Fill 55 English placeholders while preserving existing translations. Check mathematical roles, coordinate variables, keyboard actions and opposite screen reader states. Fluent-speaker and browser review were not run; other translation work remains.
Fill 61 English placeholders while preserving existing translations. Cover input roles, comparisons, start and end positions, keyboard navigation and replacement tokens. Fluent-speaker and browser review were not run; other translation work remains.
Fill 67 English placeholders for block fields, comments and keyboard controls while preserving existing translations. Check token order, comment state, opposite operations and keyboard End meaning. Fluent-speaker and browser review were not run; remaining translation work continues.
Fill 54 English placeholders while preserving existing translations. Cover reminder timing, template variables, saved filter replacement and opposite movement directions. Fluent-speaker and browser review were not run; other translation work remains.
Fill 88 missing entries and English placeholders while preserving existing translations. Cover date filters, import syntax, visibility restrictions, rule variables and uncertain email delivery. Albanian now includes all current source keys. Fluent-speaker and browser review were not run; other placeholders remain.
Translate the remaining 55 prose placeholders and classify 13 unchanged key names, product names and mathematical abbreviations. Extend full key-order and token checks to Esperanto and cover cancellation scope, retries and source preservation. No flagged English prose placeholders remain; fluent-speaker and browser review were not run.
Fill 154 English placeholders for Scrum planning and reports, synchronization diagnostics and email queue controls. Preserve variables and distinguish unknown estimates, partial changes, retries and cancellation scope. Existing translations are unchanged. Fluent-speaker and browser review were not run.
Fill 212 English placeholders for accessible block editing, mathematical descriptions and rule editor messages. Preserve replacement tokens and search key combinations; cover state toggles, opposite operations, rule cardinality and access restrictions. Update the documented coverage count to 115. Fluent-speaker and browser review were not run.
Fill 129 English placeholders for card filtering, reminders, visibility, map placement and accessible block editing. Preserve rule variables and filter syntax; cover visibility restrictions, timing, scope and opposite actions. Existing translations are unchanged. Fluent-speaker and browser review were not run.
Fill 44 missing entries and 11 related English placeholders while preserving existing translations. Cover delivery uncertainty, duplicate delivery, access checks and import syntax with regression checks. Fluent-speaker and browser review were not run.
Correct 21 mixed-language rule action messages and preserve opposite actions, member roles and current-list scope. Add token and meaning regression checks. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 19 mixed-language entries, preserving event distinctions, member versus assignee roles, time placeholders and Butler import limitations. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate eight mixed-language descriptions. Replace the incorrect export text in the account anonymization confirmation and preserve defaults, restrictions and technical identifiers. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 21 mixed-language entries while preserving keyboard shortcuts, link syntax, personal settings scope and deletion warnings. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 22 activity entries, preserving argument order, timestamp layout and distinctions between deadline states. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 20 mixed-language entries. Preserve invitation tokens, tenant configuration, Sandstorm access warnings and the distinction between notifications and activity recording. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 16 mixed-language entries and restore the single home-board constraint. Preserve temporary lockout meaning, opposite activation actions and export restrictions. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 15 mixed-language entries and restore literal examples. Clarify AND versus OR, unarchived card inclusion and positive integer limits. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 16 mixed-language instructions and restore literal query examples and backslash escaping. Check example fields, operators and placeholders. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate eight mixed-language import messages, restore the literal Jira API path and align visible menu labels. Preserve todo.txt syntax and placeholders. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 22 mixed-language entries while preserving ZIP limits, job-scoped deletion, saved credential behavior and technical names. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 26 mixed-language entries while preserving archive recovery semantics, role restrictions, placeholders and technical names. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 37 mixed-language entries, preserving removal scope, notifications, placeholders and undo warnings. Correct actions versus rules in the board deletion notice. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 31 mixed-language entries while preserving scheduling failure scope, distinct job states and all recoverable item types. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 30 mixed-language entries while preserving format tokens, import extensions and validation examples. Restore matching-card and username distinctions. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 20 mixed-language entries, correct the swimlane selector, and preserve global administrator rights and organization deletion conditions. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 21 mixed-language recovery entries while preserving field names, duplicate removal conditions, archive scope and the undo warning. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 45 mixed-language migration entries and restore the literal files/attachments path. Verify database URLs, configuration names, commands and distinct migration states. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 53 attachment, account lockout and copy-dialog entries. Preserve placeholders and attachment restoration semantics, and reject the reviewed Polish remnants. Kashubian wording is lower confidence and needs fluent-speaker review.
Translate 99 scheduling, repository and flow-report entries. Preserve technical values and reject the old language suffix throughout the locale. Other mixed-language text still needs review. Kashubian wording is lower confidence, especially statistical report explanations.
Replace 60 mixed-language entries while preserving product names, console menu labels, hostnames and technical identifiers. Kashubian wording is lower confidence and needs fluent-speaker review.
Replace 80 mixed-language entries and verify placeholders, technical formats, product names and opposing control labels. Kashubian wording is lower confidence and needs fluent-speaker review.
Replace 40 Polish-seeded strings, preserving search syntax, signed day offsets, font name, sort directions and completion states. Kashubian wording is lower confidence and needs fluent-speaker review; older mixed-language entries remain.
Replace 45 Polish-seeded strings while retaining weekday schedules, before and after offsets, move directions and unmapped-node warnings. Kashubian wording is lower confidence and needs fluent-speaker review; older mixed-language entries remain.
Replace 50 Polish-seeded strings, preserving product names, environment variables and reactivity modes. Distinguish email subjects from bodies and free from total memory. Kashubian wording is lower confidence and needs fluent-speaker review; older mixed-language values remain.
Replace 55 Polish-seeded strings, preserving assigned-card restrictions, login links, units and upload versus download directions. Kashubian wording is lower confidence and needs fluent-speaker review; more older mixed-language entries remain.
Replace 65 Polish-seeded strings, restore the empty comment placeholder and date codes, and retain enrollment email tokens and assigned-card role scope. Kashubian wording is lower confidence and needs fluent-speaker review; more older mixed-language entries remain.
Correct 65 older strings, preserving format tokens and restoring positive-integer height validation, recipient direction and multi-board scope. Kashubian wording is lower confidence and needs fluent-speaker review. More mixed-language entries remain.
Fill 81 recovery and keyboard strings, retain 11 technical terms and check all locale placeholders. Kashubian wording is lower confidence and needs fluent-speaker review. Older Polish-seeded values remain outstanding despite zero flagged English placeholders.
Fill 81 placeholders and replace four Polish date-filter labels. Verify date commands, rule variables, reminder offsets and access wording. Kashubian wording is lower confidence and needs fluent-speaker review; older mixed-language entries remain outstanding.
Fill 84 English placeholders and add regression checks for unknown estimates, partial reports and daily observation export limits. Kashubian wording is lower confidence and needs fluent-speaker review; older Polish-seeded values remain outstanding.
Fill 63 English placeholders and verify conflict, null-value and recovery semantics. Refresh the measured README completion count. Kashubian wording is lower confidence; older Polish-seeded values and fluent-speaker review remain outstanding.
Fill 56 missing or English strings with token and recovery warning regression coverage. Kashubian wording is lower confidence and needs fluent-speaker review. Older Polish-seeded values remain to be corrected.
Replace unrelated export instructions with permanent account anonymization effects and correct five email-template labels. Add regression coverage for retained history, disabled login and irreversibility. Scottish Gaelic wording is lower confidence and needs fluent-speaker review.
Translate 80 recovery and keyboard strings and record 13 unchanged technical terms. Verify all locale tokens and recovery warning wording. Scottish Gaelic wording is lower confidence and needs fluent-speaker review; older translations still need linguistic review.
Fill 81 English placeholders, preserving date commands, rule variables and reminder offsets. Add regression coverage for access and notification wording. Scottish Gaelic wording is lower confidence and needs fluent-speaker review.
Fill 84 English placeholders, preserving report limitations and tokens. Add regression checks and refresh the measured language completion count. Scottish Gaelic wording is lower confidence and needs fluent-speaker review.
Fill 63 entries for sync conflicts, previews, reports and Jira estimates. Preserve existing translations and source tokens; cover source isolation, retained content, report limitations and explicit null handling.
Scottish Gaelic wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 56 missing or English entries for rule email delivery, resend and discard actions, variable insertion and todo.txt import. Preserve existing translations and executable tokens; cover delivery uncertainty and permission warnings.
Scottish Gaelic wording is lower confidence. Dictionary terminology was checked; fluent-speaker and browser review were not run.
Thanks to xet7 !
Correct seven entries in each Welsh locale: member mapping, email template labels and an anonymization warning that incorrectly called replacement values temporary. Preserve other translations and add positive and negative wording regressions.
Welsh wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 80 entries in each Welsh locale and retain 13 reviewed keyboard labels, product names and mathematical symbols. Add both locales to whole-locale checks and cover uncertain delivery and irreversible cancellation warnings.
Existing translations were preserved. Welsh wording is lower confidence; fluent-speaker and browser review were not run. Older wording still needs review.
Thanks to xet7 !
Fill 81 entries in each Welsh locale without replacing existing translations. Preserve filter syntax and rule variables; cover reminder offsets, signed-in visibility, notification exceptions and list age semantics.
Welsh wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 84 Scrum and navigation entries in each Welsh locale. Preserve existing translations and source tokens; cover unknown estimates, partial reports and daily observations. Update the measured README language count to 112.
Welsh wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 63 entries in each Welsh locale for sync conflicts, previews, reports and Jira estimates. Preserve existing translations and source tokens; cover source isolation, content retention, report limitations and explicit null handling.
Welsh wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 56 missing or English entries in each Welsh locale for rule email delivery, resend and discard actions, variable insertion and todo.txt import. Preserve existing translations and executable tokens; cover delivery uncertainty and permission warnings.
Welsh wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Correct seven entries with changed meaning or mixed Spanish wording. Preserve other translations, including the already correct anonymization warning. Add positive and negative wording regressions.
Fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 79 remaining entries and retain 13 reviewed keyboard labels, product names and mathematical symbols. Add Basque to whole-locale key, placeholder and completeness checks; cover delivery uncertainty and irreversible cancellation warnings.
Existing translations were preserved. Fluent-speaker and browser review were not run; older wording still needs review.
Thanks to xet7 !
Fill 81 entries without replacing existing translations. Preserve filter syntax and rule variables; cover reminder offsets, signed-in visibility, notification exceptions and list age semantics.
Fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 84 Scrum and navigation entries while preserving existing translations and source tokens. Cover unknown estimates, report scope, daily observations and unavailable imports. Update the measured README language count to 110.
Fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 63 entries for sync conflicts, previews, reports and Jira estimates. Preserve existing translations and source tokens; cover source isolation, content retention, report limitations and explicit null handling.
Fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 56 missing or English entries for rule email delivery, resend and discard actions, variable insertion and todo.txt import. Preserve existing translations and executable tokens; cover delivery uncertainty and permission warnings.
Basque wording has not had fluent-speaker or browser review.
Thanks to xet7 !
Correct ten entries: account anonymization consequences, archive restoration location, timeline state and French email-template labels. Add positive and negative meaning regressions while preserving other translations.
Breton wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 83 remaining entries and retain 13 reviewed keyboard labels, product names and mathematical symbols. Add Breton to whole-locale key, placeholder and completeness checks; cover delivery uncertainty and irreversible cancellation warnings.
Existing translations were preserved. Breton wording is lower confidence; fluent-speaker and browser review were not run. Older translations still need linguistic review.
Thanks to xet7 !
Fill 81 entries without replacing existing translations. Preserve filter syntax and rule variables; cover reminder offsets, signed-in visibility, notification exceptions and list age semantics.
Breton wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 80 Scrum entries while preserving existing translations and source tokens. Cover unknown estimates, report scope, daily observations and unavailable imports. Update the measured README language count to 109.
Breton wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 63 entries for sync conflicts, previews, reports and Jira estimates. Preserve existing translations and source tokens; cover source isolation, local content retention, report limitations and explicit null handling.
Breton wording is lower confidence; fluent-speaker and browser review were not run.
Thanks to xet7 !
Fill 56 English or missing entries for rule email delivery, administrator review, resend and discard actions, variable insertion and todo.txt import. Preserve existing translations and executable tokens; add warning and syntax regressions.
Breton wording is lower confidence. Dictionary terminology was checked; fluent-speaker and browser review were not run.
Thanks to xet7 !
Restore account anonymization consequences, board restoration location, cancellation actions and member mapping meaning. Add positive and negative wording regressions; preserve all other translations and source tokens.
Occitan wording is lower confidence and has not had fluent-speaker review.
Thanks to xet7 !
Correct 12 mixed-language or inaccurate messages, including account creation described as activation and the missing Trello print-and-export step. Preserve other translations, placeholders and API syntax. Add regression coverage. Wording remains lower confidence pending fluent-speaker review.
Translate 79 messages and retain reviewed shared words and technical labels. Verify all source keys, placeholders, existing translations and irreversible recovery wording. Occitan wording is lower confidence pending fluent-speaker review; the broader language audit remains open.
TODO Later: even a same-list move to top or bottom writes a position History
row and a legacy UserPositionHistory row from Card.move, and a cross-list or
cross-board move relabels, renumbers, maps custom fields and re-syncs
checklists and attachments - the analysis for whoever makes moves durable.
Thanks to xet7 !
Fill 81 messages while preserving existing translations and executable syntax. Cover visibility restrictions, reminder offsets and list-age filtering. Occitan wording is lower confidence pending fluent-speaker review.
Upcoming gains the checklist actions and the retention-sweep test fix;
TODO Later now counts twenty durable rule action types and lists the
structural ones that remain.
Thanks to xet7 !
checkAll, uncheckAll, checkItem and uncheckItem join the durable rule
actions. The command is captured with performAction's own lookups - the
checklist by title on the card, the items in the order ReactiveCache returns
them, the bulk selection, and checkItem's item query as it is - so a replay
acts on the same items with the same outcome, including the ordinary
action's no-ops.
The ordinary action writes each item in turn, and the checklist item hooks
record on every write, changed or not: uncompleteChecklist before it when
the whole checklist was finished, checkedItem or uncheckedItem after it, and
completeChecklist when the whole checklist is finished - plus a History row
on the item when isFinished changed. The command simulates exactly that on
the captured items and saves one unit per item. The runner goes item by item:
the before-write activity, the conditional write with the item's three hooks
deferred, its History, its after-write activities - each delivered durably -
so a rule triggered by one item runs before the next, as ordinarily.
The History planner now accepts checklist-item rows (entity and card kept
apart), and the recording deferral can name one checklist item. Unit tests
cover the hook sequence for check-all, uncheck-all and hidden-all lists,
no-ops and tampering; a full-app test runs a "check every item" rule through
the stored rule stage - two checkedItem, one completeChecklist, two History
rows, and nothing more on replay. 566 server tests pass.
Thanks to xet7 !
A retention sweep reads one page of 100 receipts, oldest first, and keeps its
place. The full-app test database outlives a run, so receipts left by other
suites and earlier runs came to fill the first page, and the webhook test's
single sweep compacted nothing. Production pages through every receipt over
successive sweeps; the tests now do the same through a shared sweepUntil
helper, and assert that their own plan was compacted.
Thanks to xet7 !
Fill 84 Scrum and navigation messages while preserving existing translations and tokens. Cover unknown estimates, partial reports and daily observation limitations. Refresh the measured language completion count. Occitan wording is lower confidence pending fluent-speaker review.
Fill 63 messages while preserving existing translations and executable tokens. Cover one-way synchronization, report limitations, explicit null and unique estimate fields. Occitan wording is lower confidence pending fluent-speaker review.
Fill 45 placeholders while preserving existing translations and executable syntax. Cover duplicate delivery, permanent discard and lost access wording. Occitan wording is lower confidence pending fluent-speaker review.
Upcoming's List Sync group gains the colour, label, completion, date and
member rule actions; TODO Later narrows the remaining durable rule work to
the structural actions - moves and sorting, checklists, and creating,
copying or linking cards and swimlanes - and says why each needs its own
saved command.
Thanks to xet7 !
addMember and removeMember join the durable card-field actions. The people
are resolved at capture exactly as performAction resolves them - the acting
user for the #2522 option, username tokens through buildRuleVars (#4294),
and the card's assignees for "remove every member" (#2674) - by a new
RulesHelper.resolveMemberTargets beside it, and saved with their usernames,
so a replay acts on the same people. The ordinary action's one $addToSet or
$pull per person becomes one planned change of `members`, with a joinMember
or unjoinMember activity for each person who really joined or left, as the
members hook writes it, and one members History row.
The new helper phrases its conditions so the source guards that locate
performAction's member branch still find it, and the recording-isolation
test now uses assignees as its example of a field that is never deferred.
Unit tests cover resolution being saved, joins and leaves, nobody resolved
and tampered targets; the full-app durable Sync test adds a member to every
created card through a rule - one joinMember activity and one members
History row. 565 server tests pass.
Thanks to xet7 !
Correct 30 mixed-language messages and update board restoration guidance. Preserve other translations and placeholders. Cover role restrictions, linked-card deletion order, card terminology and movement directions. Wording is lower confidence pending fluent-speaker review.
setDate, updateDate, setDateRelative and removeDate join the durable
card-field actions. Their semantics follow the ordinary actions exactly:
setDate only fills a date that is unset, updateDate always sets it,
setDateRelative offsets from "now", removeDate unsets - with "now" fixed at
the command's capture time, so a replay makes the same change.
A full-app probe of the ordinary writes showed what they record: a dates
History row (added, edited or removed) and the timing hook's a-<field>
activity with username, card title, time key and the old value ('' when
there was none) - and no timeValue when the date is removed. The command
plans exactly that. The timing hook in server/models/cards.js gains a
'timing' deferral kind so the saved activity is written instead of the
hook's own, never both; the runner can now unset a field as well as set it.
Unit tests cover all four actions, setDate on a filled date, the timing
activity's shape for set and unset, and a replay-stable relative date; the
full-app durable Sync test gives every created card a due date through a
rule - one a-dueAt activity and one dates History row. 565 server tests pass.
Thanks to xet7 !
Correct 26 mixed-language messages while preserving other translations and placeholders. Restore the single home-board constraint and non-deletion warning. Cover native terminology, archive actions and comment restrictions. Wording is lower confidence pending fluent-speaker review.
Boards with any rule action other than email and archive stayed on direct
Sync. Six more now take the durable path: setColor, addLabel, removeLabel,
removeAllLabels, markCardComplete and markCardIncomplete.
Each rule invocation saves one command (server/lib/syncRuleCardCommand.js)
with its card field's value before and after the action - with the ordinary
action's semantics: $addToSet/$pull for single labels, a missing field stays
missing on removeLabel, 'white' is stored as null - and the effects the
ordinary hooks would write: one History row for the field and, for
addLabel/removeLabel, the addedLabel/removedLabel activity only when the
label really changed (removeAllLabels writes none, as ordinarily). The runner
applies the saved change conditionally with ordinary History deferred for
that one write, then writes the planned History rows and activities, each
activity delivered durably with its own rules and notifications, and returns
the invocation only when all of it is confirmed. Replays use the saved
command, never the rule's current configuration.
The History planner is parameterized by allowed fields (Sync's own, or these
three), and the recording deferral covers labelIds, color and dueComplete.
Unit tests cover every action, the effects, absent-versus-null selectors and
tampering; the full-app durable Sync test labels every created card through
a rule - one History row, one activity, nothing more on replay. The rule
selection test's example of an action without an adapter is now
moveCardToTop. 565 server tests pass.
Thanks to xet7 !
Fill 84 Scrum and navigation messages while preserving existing translations and tokens. Cover unknown estimates, partial reports and daily observation limitations. Update the measured language completion count. Asturian wording is lower confidence pending fluent-speaker review.
Upcoming gains the guard-sharing change that made rule archive a durable Sync
action and the archive command retention; TODO Later now lists only the
rule actions still without a durable adapter as design work.
Thanks to xet7 !
With archive now a durable Sync action, the stored archive runner writes
commands that keep up to 1000 cards' titles and placement, and effects rows
that keep the History and activity content written for them - forever, with
only a timeless final receipt to say they finished.
The runner now writes a completion receipt with its time after it finishes.
SYNC_RECEIPT_METADATA_DAYS (90) later the sweep removes the effects row - read
only while the command is unfinished - and replaces the command in place by
its id, checksum, invocation and plan. A late replay of the rule action
returns the invocation as done when the runner's own final receipt confirms
it; card and final receipts stay.
A MongoDB test covers compaction, the untouched recent and unfinished
commands and the refusals; the full-app durable Sync test compacts the real
archive command from its rule run and replays it. 565 server tests pass.
Thanks to xet7 !
Stored Sync stages nest - a rule archive's delivery inside the rule stage,
inside activity delivery, inside the Sync journal - and every layer's guard
checks before and after by calling the guard below it twice. One innermost
check therefore re-ran every outer guard 2^depth times, and one Sync-created
card archived by a rule took 36 s, which is why archive was held back from
durable Sync (1e40e7e06).
reuseWithinEvaluation (server/lib/syncGuardWindow.js) runs each guard at most
once per evaluation: calls made while another guard's check is running share
the results of guards already checked in it. A call from ordinary work -
between stages, after an adapter or a write - starts a new evaluation and
checks everything afresh, so a policy change or a lost lease between stages
still stops the next one. A first attempt with a 100 ms reuse window missed
exactly that and was dropped. Failures never carry over, a guard reached
again through its own check runs fresh instead of waiting on itself, and the
journal's ownership read is left as it was.
With that, archive and unarchive are registered as durable rule adapters
through the stored archive runner, and boards whose rules archive or
unarchive now take durable Sync. The full-app durable Sync test runs an
"archive every created card" rule end to end: 1.1 s for the whole test. The
rule selection test's example of an action without an adapter is now
moveCardToTop. 565 server tests pass.
Thanks to xet7 !
Correct 22 mixed-language messages, including automation triggers and list-width settings. Restore executable filter examples and preserve all other translations. Add vocabulary and syntax regression checks. Wording is lower confidence pending fluent-speaker review.
Correct 35 mixed-language messages and restore member-removal notification meaning. Preserve placeholders and other translations. Cover permanent deletion, read-only role summaries, positioning and API configuration syntax. Wording is lower confidence pending fluent-speaker review.
Upcoming gains the rule plan compaction entry; TODO Later records that rule
archive commands and effects need no sweep yet, because only the stored
archive runner writes them and it is not registered as a durable adapter.
Thanks to xet7 !
Maintainer decision of 2026-09-30: receipts are compacted after 90 days to
a permanent minimal id and outcome. A stored rule plan keeps the whole
documents of the rules and actions it matched - a send-email action's
recipients and text among them - and was kept forever; its 'rules' receipt
had no time, so it could not be told old.
runStoredSyncRules now writes a completion receipt with its time once every
action is done, and the shared plan retention compacts the plan in place
SYNC_RECEIPT_METADATA_DAYS (90) later. A replay of the compact form returns as
done; every other stage that needs the whole plan - email command capture,
archive capture - refuses the stub, and the legacy email review treats a
command whose plan is compacted as finished, never rebindable. Action
receipts stay, and rule email commands keep their own schedule.
A MongoDB test covers compaction, the untouched unfinished plan, replay and
the legacy review; the full-app rule test compacts a finished plan and
replays it. 565 server tests pass.
Thanks to xet7 !
Correct 40 activity messages while preserving placeholders and other translations. Restore comment-edit semantics and test checklist completion, source and destination lists, and Aragonese activity vocabulary. Wording remains lower confidence pending fluent-speaker review.
Upcoming gains the webhook plan compaction entry and names it in In short
and the Retention group; TODO Later narrows the remaining retention work to
rule plans and rule archive content, whose replay readers re-validate the
full plan and derive other commands' checksums from it.
Thanks to xet7 !
Maintainer decision of 2026-09-30: receipts are compacted after 90 days to
a permanent minimal id and outcome. A stored webhook plan keeps each
target's URL, its X-Wekan-Token credential and the request body; a response
row keeps the receiver's reply and a comment plan the comment text - all
forever, and the plan had no completion evidence of its own.
The notification plan retention becomes a shared kit (syncPlanRetention.js)
and webhooks use it: once every target's delivery is confirmed, a completion
receipt with its time is written. After SYNC_RECEIPT_METADATA_DAYS (90) the
sweep first removes the plan's response and comment-plan rows - read only
while a delivery receipt is missing - then compacts the plan in place. A late
replay of the compact form returns as delivered; delivery and comment
receipts, already minimal, stay.
A MongoDB test covers what is removed and what an undelivered plan keeps,
replay of the compact form and its refusals; the full-app webhook test
compacts a really delivered plan, checks the URL and token are gone, and
replays it without touching the comment. 565 server tests pass.
Thanks to xet7 !
Replace Spanish remnants in 42 interface messages. Preserve placeholders and all other translations. Add regression checks for Aragonese wording, imported-member permission limits, irreversible deletion and saved-secret retention. Wording is lower confidence pending fluent-speaker review.
Upcoming gains activity incarnations, Sync notification plan compaction,
durable manual and scheduled Sync with replay, and the two full-app test race
fixes; its In short names every decision carried out. TODO Later now lists
what remains with measured or structural reasons: durable adapters for rule
actions other than email (archive measured at 36 s per card through nested
guards), retention of the other stored Sync content, which first needs
completion receipts, atomicity bounded by backends without transactions, and
older lists without an incarnation, which need a migration decision.
Thanks to xet7 !
Replace Spanish and mixed-language wording in 30 reviewed messages. Preserve all other locale values, placeholders, API paths, JSON fields and Markdown syntax. Add language and syntax regression checks. Wording remains lower confidence pending fluent-speaker review.
The stored archive runner could be registered as the archive/unarchive
adapter, but a full-app run showed one Sync-created card archived by a rule
taking 36 seconds: each nested delivery stage re-runs every outer guard -
journal scope, intent, lease, rule and archive checks - hundreds of times.
Boards with archive rules stay on direct Sync until that guard composition
is made cheaper.
Thanks to xet7 !
The write-ahead Sync journal and its stored stages - History, activities,
rules, notifications, webhooks - were built and tested but never called:
manual and scheduled Sync wrote cards directly, so a restart mid-run lost
whatever effects had not happened yet. Maintainer decision of 2026-09-30:
finish the Scrum/Sync handoff by calling the stored stages from both.
A run now takes the durable path when it can do everything
(server/lib/listSyncSteps.js): the board enabled Sync effects (and the
instance enabled cron effects for a scheduled run), the list has a versioned
scope, and every rule action on the board - extra actions included - has a
durable adapter. Only sending email has one, so a board with any other rule
action keeps the direct writes, exactly as before; nothing that ran stops
running.
server/lib/listSyncApplication.js turns the reconcile plan into saved steps
from the stored cards, refusing a local edit made since the fetch as the
direct update would, and carrying custom fields only with a mapped estimate.
Effects are planned once with the shared planner and each step is applied
through the hooked card and activity adapters and durable delivery, as the
actor and trigger that started it. Both paths build the same new card.
A new run first finishes the list's unfinished operation, reusing the list
lease it holds instead of taking it twice. The wekan-list-sync-replay cron
job resumes interrupted operations every minute; intents now record their
trigger (version 2) so a replay runs under the same activation switches. An
operation still preparing has written nothing and is discarded - a replay has
no source data to rebuild it from.
Tests: eligibility for every condition, and step building including the
local-edit refusal (node); the direct-path suite now pins that it never calls
the durable one; a full-app run creates, updates and archives through the
journal, crashes at a card write, replays without duplicate cards or
activities, and falls back to the direct path for a rule action without an
adapter. 565 server tests pass.
Thanks to xet7 !
Fill 45 English placeholders while preserving existing translations and executable tokens. Cover duplicate delivery, permanent discard and lost access wording. Aragonese wording is lower confidence pending fluent-speaker review.
Fill 81 remaining messages while preserving existing translations and syntax tokens. Add regression checks for visibility, reminder offsets and filter behavior. Neapolitan wording is lower confidence pending fluent-speaker review.
Maintainer decision of 2026-09-30: receipts are compacted after 90 days to a
permanent minimal id and outcome, notification plans included. A stored Sync
notification plan keeps every recipient's rendered subject and HTML, and had
no completion evidence of its own, so it could never be told finished and
was kept forever.
Delivery now writes a receipt once it is confirmed - planId, activity hash,
plan checksum and completedAt; the first stored receipt wins. After
SYNC_RECEIPT_METADATA_DAYS (90) an hourly sweep replaces the exact plan the
receipt was written for with { _id, activityHash, checksum,
compactReceiptVersion } in place, so its id stays taken. A replay that finds
the compact form returns as delivered and never rebuilds or resends; within
the period a replay still re-checks every recipient, as before. A compact
plan without its matching receipt, or for changed activity content, is
refused.
The other receipt rows named in the decision - Scrum completions and
requests, Sync operation intents and completions - already hold only ids,
hashes and timestamps, so there is nothing in them to compact.
MongoDB tests cover what is compacted and what is not, idempotence, replay
of the compact form and the refusals; the full-app delivery test compacts a
real delivered plan and replays it without sending.
Thanks to xet7 !
The startup recovery scan runs every second and holds an intent's lease
while it looks at it, so the test's pause, resume and cancel calls could meet
sync-busy - a correct answer to an operator, who retries. A full run hit it
once, on the un-pause. The test now retries busy the way an operator would;
the one call that asserts busy on purpose, against the test's own delivery,
is left as it is.
Thanks to xet7 !
Maintainer decision of 2026-09-30: every Scrum record and activity gets a
new random incarnation each time it is created, so a record deleted and
recreated with the same values is not taken for the original. Scrum records
have it; activities now do too. The server insert hook sets it on every
insert - whatever the caller passed - before the notification intent
captures the activity, and recovery already compares the stored activity
with the captured one, so another lifetime is refused as unconfirmed.
Sync-planned activities are deferred past this hook on purpose: their
deterministic id is their identity, and a replay must match the plan.
A full-app test inserts an activity, recreates it under the same id with the
same values and timestamps, and recovery refuses it; the original lifetime is
still accepted, and a caller cannot choose the incarnation. 564 server tests
pass.
Thanks to xet7 !
The activity notification recovery scanner reads 100 pending intents at a
time and keeps its place between calls, and the startup scan shares it. In a
full test run other suites leave more than 100 intents pending, so one call
could stop before this test's intent - by the random order of ids - and the
paused-scan assertion saw no skipped row. It failed once in a full run and
passed on its own.
Production pages through every intent over successive scans, which
tests/integration/activityNotificationRecovery.test.cjs already pins. The
test now uses its own scanner and pages to the end, repeating a pass that
lost a race with the background scan.
Thanks to xet7 !
Upcoming gains the History writers group and its In short names all four
2026-09-30 decisions; TODO Later records the recovery as built and why a
token from an older server still needs the offline tool.
Thanks to xet7 !
A legacy History writer holds a token on its board's gate until its write is
confirmed, with no expiry, so an uncertain write can never become invisible to
a chain migration. A writer that died kept its token forever, and the only way
to remove it was the offline tool with every server stopped. Maintainer
decision of 2026-09-30: build online recovery - liveness heartbeats and fencing
on History inserts - and keep the offline tool as the fallback.
Each writer on an upgraded server now writes a lease beside its token
(historyWriterLeases) and renews it while it works. Before every legacy
History insert it claims the row id in the lease; only an active lease can
claim. Recovery takes a lease only after it has been expired for another full
lease period (HISTORY_WRITER_LEASE_MS, default 30 s). It sets the lease to
fenced, so the old writer can neither renew nor claim. A claimed row id is
then fenced by a tombstone under that id, so the unique _id makes a late
insert fail, and a row that had already landed stays. Only then is the token
removed. Safety rests on the fence; clocks only decide when a lease may be
taken, and a slow writer that is taken loses its write with an error.
Tombstones use the reserved board id #history-writer-fence and are invisible
to every board reader. A Sync append refuses a tombstone as its planned row.
A chain migration takes abandoned writers by itself while the board drains,
a full gate does the same once, and releases/recover-history-writer.cjs
--recover-expired does it on demand, online. Tokens without a lease (older
servers) are listed and still need the offline procedure.
MongoDB tests with two clients: a writer that died after claiming a row is
taken and its late insert fails; a fenced slow writer cannot claim; a row
that landed before the fence is kept; live writers and lease-less tokens are
not taken; a migration drains a dead writer and completes on the real chain
head. A full-app test covers the production binding. 563 server tests pass.
Thanks to xet7 !
Upcoming opens with the ReplyBleed security fix and adds Scrum incarnations,
rule email compaction and Sync History linking at write, grouped by topic.
TODO Later drops the resolved reservation blocker and lists what remains:
activity incarnations (not needed, with the reason), compaction of plan
receipts without a completion timestamp, online writer-token recovery, the
rest of the Scrum/Sync handoff and one intermittent full-app test.
Thanks to alex131125 and xet7 !
The reply-by-email token signed only the card, so every recipient of a
card's notification held the same Reply-To address, and the webhook took the
comment author from the reply's From address - a field the sender controls.
Anyone holding one card's reply address could comment on it as any user
(GHSA-mc7c-cv99-64h7, CWE-346).
Validation also found that server/routes/inboundEmail.js was never imported
by server/imports.js, so released versions did not register
/api/inbound-email: the documented feature did not work and the flaw could
not be reached. The endpoint is registered now, together with the fix.
The reply address is now reply+<cardId>.<userId>.<expiry>.<mac>: an
HMAC-SHA256 over the card, the recipient and the last valid day
(INBOUND_EMAIL_REPLY_DAYS, default 30). The author is that recipient; the
From address must be one of the recipient's own addresses; the recipient must
be enabled and still allowed to comment on the card, including the
assigned-only restriction. INBOUND_EMAIL_WEBHOOK_SECRET, when set, must be
presented by the provider (X-WeKan-Inbound-Secret header or ?secret=) and is
compared in constant time. The old card-only token is refused.
A forged token, a spoofed sender and a wrong provider secret are recorded as
ReplyBleed in Admin Panel -> Problems; expired and pre-fix tokens and lost
access are refused quietly, since real users replying to old mail reach them.
Tests: token round trip, per-recipient addresses, tampering, expiry and
legacy tokens; the author decision reproducing the advisory's attack; a
tree-wide negative that no code picks an author from a From address; and a
full-app HTTP test against /api/inbound-email.
Thanks to alex131125 and xet7 !
Fill 45 placeholders while preserving existing translations and syntax tokens. Sardinian terminology has not had fluent-speaker review; wording is lower confidence.
Sync and the stored rule archive runner fixed each History row's
previousHash and integrityHash when they PLANNED a batch, so the board's
History chain had to stay unchanged from planning until the last row was
written - which needed a multi-row History reservation that would block
or queue ordinary edits, and made a coordinated board refuse Sync.
Maintainer decision of 2026-09-30: plans are content only, and each row
is linked when it is appended, like an ordinary edit.
On a board still on the legacy path, the batch holds one writer token
(so a chain migration drains it) and each row is linked after the
chain's tip - the newest hashed row, then its successors, since the rows
of one batch share a planned createdAt and "newest" alone is a tie that
forks the chain. On a coordinated board each row goes through the chain
head. Appends are idempotent by row _id, and a replay accepts only the
planned content. The archive runner needs no reservation any more; it
captures the actor's redo candidates when it plans. Plans in the old
chained format are refused. An index on (boardId, previousHash) serves
the tip walk.
Unit and MongoDB tests: plans carry no link; rows chain correctly in
both modes and replay once; ordinary History written between planning
and writing is neither blocked nor forked; a coordinated board appends
through the head; old-format plans and adapters without admission or
linking are refused. The full-app archive case now runs without a
reservation and verifies the resulting chain. 561 server tests pass;
the one other failure in that run, the activity notification recovery
scan, is intermittent and passes on its own.
Thanks to xet7 !
Maintainer decision of 2026-09-30: receipts are compacted after 90 days
to a permanent minimal id and outcome, as the notification outbox does.
A rule email command keeps the whole mail - body and attachments up to
12 MB - and its outcome row keeps recipient addresses, forever. Now,
once an attempt is sent or dropped, is older than
SYNC_RECEIPT_METADATA_DAYS (90), and its rule invocation has a receipt
- after which no replay reads the mail - the command is replaced in
place by its _id, checksum, invocation, plan and effect ids, and the
outcome row is removed. The attempt row, already minimal, stays as the
receipt a late retry finds. A command whose checksum no longer matches
its attempt, or that has no receipt yet, is left for a later pass.
The recovery detail returns an empty recipient list for a compacted
command (only finished attempts are compacted, and none can be resolved
again), and the legacy review no longer counts compacted rows. The
sweep runs hourly (SYNC_RECEIPT_SWEEP_INTERVAL_MS).
A MongoDB test compacts only the old finished command with a receipt,
leaves the one without a receipt, the recent one and the unconfirmed one,
repeats without change, reads the compact form in both readers, and
refuses a command that no longer matches its attempt.
Thanks to xet7 !
Scrum History judged "the record has the expected values and revision"
as proof of which record it was. One deleted and recreated under the
same _id - a restore by someone else - passed that test, so a retried
undo could delete a record from another lifetime. Maintainer decision
of 2026-09-30: every Scrum sprint, release and event now carries a
random incarnation per lifetime, set where it is created (the Scrum
methods, a board transfer import's plan, and a Scrum History restore).
A new checkpoint records the incarnation each target has and, for a
record it creates, the one it will have, chosen before writing so a
replay recognizes its own insert; the write-state check refuses any
other lifetime. History content never carries an incarnation, so a
restore always starts a new lifetime, and an update never unsets it.
Checkpoints and records from before incarnations keep working as they
did, and their completion identity is unchanged.
Unit tests cover every state with and without incarnations and the
planned lifetimes; a full-app test creates, undoes and redoes a sprint,
then replaces it with an identical record of another lifetime mid-undo:
the retry refuses and leaves it, where the previous code deleted it.
562 server tests pass.
Thanks to xet7 !
TODO Later now answers the questions the Scrum/Sync handoff waited on
(maintainer decisions of 2026-09-30): Sync and rule archive History rows
are linked to the chain when appended, not when planned, so ordinary
History is never blocked; Scrum records and activities carry
incarnation ids to tell a recreated record from the original; receipts
and request IDs are compacted to a permanent minimal id after 90 days;
and History writer tokens get online recovery with liveness heartbeats
and fencing tokens, the offline tool remaining as a fallback.
v12.12 was published before its section gained the Irish translation
entry (7f9fc419c), which changelogEntriesBelongToTheirRelease reports.
It moves unchanged to a new Upcoming section, and v12.12 is again what
was published.
Thanks to xet7 !
The Rules page had a separate right sidebar holding only Back and its
views (List view, Workflow view, Blocks, History, Import / Export
rules). It now uses the board's own sidebar, like the board: the same
views are in its Board Settings under Rules, and choosing one there
while on the Rules page closes the sidebar so the view is not left
under it. The separate sidebar and its template are removed.
Its Back button went with it, so picking a view in the Board View menu
while on the Rules page now goes back to the board, asking first before
unsaved Blocks work would be lost.
Source tests pin the Rules route to the board sidebar, the absence of
the old controls, the way back and the Board Settings entries' existing
translated labels. The Rules browser specs open views through Board
Settings; a member who is not a board admin is not offered them there,
and the Blocks view still refuses such a member. The Rules and Board
Settings specs pass in Chromium and WebKit.
Thanks to xet7 !
Board Settings now lists the Rules page's own views directly under
Rules - List view, Workflow view, Blocks, History and Import / Export
rules - so the one wanted is a click away instead of Rules first and
then its page sidebar. Each opens the Rules page in that view, asking
first before unsaved Blocks work would be lost, as the Rules page
sidebar does; Import / Export rules opens its popup from the menu. They
share Rules' board-admin guard. A divider now separates that group from
Change Color.
The Import / Export entry reads "Import / Export rules" in Board
Settings and in the Rules page sidebar, using the popup's own title key,
which is already translated in 234 locales, so no new text is added.
The menu order test pins the new first group, its divider, the guards,
the handlers and the label keys; a Playwright case checks the order and
that Workflow, History and Import / Export rules open what they say, in
Chromium and WebKit. The Board Settings and Rules docs follow.
Thanks to xet7 !
Semi-open boards (f122a8287) made server/filterPresets.js require
models/lib/boardPermission, but the MongoDB integration test runs that
file in a sandbox without require, so it failed with "require is not
defined" before testing anything. The sandbox now hands it the real
module and refuses any other; the atomic-save and scoped-deletion
assertions are unchanged and pass.
Thanks to xet7 !
The release telemetry audit, which the release bump job also runs,
reports every new URL origin in shipped source as a risk indicator. A
comment in models/lib/todoTxtFormat.js linked the format specification
on github.com, so the next release would have stopped there. The comment
now names the specification without a URL; the CHANGELOG and the format
coverage document keep the link.
Thanks to xet7 !
changeHistoryQuery pins that History rows change in place only in their
undo flags; the claim that undo and redo now take sets them through
named objects, so the guard reads those and still refuses any other
field. historyRestoreAppliesWhatIsShown pins that undo and redo apply
their own direction; they now reach applyRow through applyClaimed with
that direction. The rule email report UI test's sandbox gains a stub
for the legacy review template that follows in the same file. The rules
each guard pins are unchanged.
Thanks to xet7 !
The Monte Carlo export link carries the options of the chart that is
shown, which come from the server's answer, not from the form. The test
waited for a table row containing "7", which the previous answer can
satisfy by coincidence, then opened the export menu before the new
options arrived: it failed two runs in three. It now waits for the
answer carrying targetCount 7 before opening the menu, and passed four
runs in four against a production bundle.
Thanks to xet7 !
Adds both entries to the External imports group and todo.txt to the
In short paragraph, and updates the TODO Later import/export note: new
formats now cost one English instruction string pending Transifex, not
a translated string in every locale, and todo.txt is the template.
Thanks to xet7 !
todo.txt (github.com/todotxt/todo.txt) is the one-task-per-line format
the todo.txt CLI and many compatible apps share, and WeKan read and
wrote neither. It now imports from the import picker and exports from
the board's export menu. A completed task goes to Done; priority becomes
a priority:A label; creation and completion dates become the card's
creation and end dates; +project and @context become labels; due: and
t: become due and start dates; WeKan's list: extension keeps list names
across a round trip. Other key:value pairs stay in the title, a
malformed date is reported instead of guessed, and a completion date is
written only with a creation date, as the format requires. todo.txt has
no descriptions, comments or members, so an export leaves them out.
tests/todoTxt.test.cjs runs the specification's examples, the round trip
and the negatives; the format guards learn the new format and a
Playwright case imports a todo.txt through the page. The new
instruction text is in English, pending translation.
Thanks to xet7 !
The GitLab export formatter emitted title, description, state, labels
and due date, while the GitLab importer now reads every assignee, the
author, creation and close dates and embedded notes (#2698). A board
exported to GitLab and imported back kept none of those. The formatter
now writes them; it writes no iid, because a re-import would then add a
second "Source:" line to each description. The round-trip test pins the
people, dates, comments and an unchanged description.
Thanks to xet7 !
v12.11 was published before its section gained twenty translation
entries describing commits it does not contain, which
changelogEntriesBelongToTheirRelease reports once the release tags are
present. Those entries move, unchanged, to a new Upcoming section, and
v12.11 is again exactly what was published. Upcoming records the rule
email recovery and legacy review, GitLab import fidelity, watchers and
followers, the raced undo/redo fix, the GitLab fixture fix, the Sync
activation switches, Sync History admission and Scrum worker
serialization.
TODO Later records what the 2026-09-30 pass built and tested, that the
intermittent import spec did not reproduce against a production bundle,
and what remains with its reason - chiefly the multi-row History
reservation, which needs a maintainer decision between blocking ordinary
History during a Sync batch and leaving Sync rows unhashed until append.
Thanks to xet7 !
The board lock in server/scrum.js is a per-process queue, so two server
processes could resume one Scrum History operation at the same time and
interleave its writes; the checkpoint only proved the plan was the same,
not who was running it. Each resume now claims the checkpoint with its
own worker id after adopting the operation id. The newest claim wins,
and every guard between writes checks the claim, so a displaced worker
stops at its next guard instead of writing on. The worker id is kept out
of the plan selector, so completion hashes are unchanged, and a claim
never attaches to a replaced plan or a checkpoint without an operation id.
Unit tests cover newest-wins, displacement, a lost claim reply and a
replaced plan; a MongoDB test with two clients shows the older process
refused after the newer claim. 561 server tests pass.
Thanks to xet7 !
Sync writes History rows it hashed when it planned them, straight into
the collection. On a board whose History chain had been migrated to the
coordinated head, that would fork the chain, and a migration could start
while a Sync batch was still writing. persistSyncFieldHistory now writes
only as a writer the History gate admits: production ChangeHistory holds
a legacy writer token for the whole batch, so a migration drains it
first, and a coordinated board refuses Sync History with
sync-history-coordination-required instead of appending around the head.
There is no default: an adapter without admission is refused. The rule
archive path shares it through persistSyncEffects.
A MongoDB integration test uses the real gate: the batch holds a token
while it writes and releases it after; a migration waits for a batch in
flight, a new batch is refused while the board drains, the in-flight one
completes; a coordinated board refuses and writes nothing. Test adapters
gained the pass-through admission; 561 server tests pass.
Thanks to xet7 !
Two undos that picked the same History row - two tabs, or a repeating
key - both applied it, both flipped its flag unconditionally afterwards
and both recorded a reversal, so History gained duplicate reversal rows.
Undo and redo now claim an ordinary row first with a conditional flag
write (a live row for undo, an undone and not superseded row for redo);
only the caller whose claim matched applies it, and the others are told
they lost the race. An apply that does nothing or throws gives the claim
back exactly as taken, so a row is still never left marked undone when
nothing was undone. Scrum rows keep their own conditional finalizer.
A new full-app test fires four concurrent undos and four redos at a real
card edit: one wins each time, the title changes once and one reversal
is recorded per direction; it fails against the previous code ("exactly
one undo wins"). It also pins that a claim is released after an apply
that does nothing and after one that throws. Source guards follow.
Thanks to xet7 !
The shared GitLab fixture says it has one comment (user_notes_count 1)
but carried none, and the GitLab adapter now truthfully reports such a
comment as not imported. The browser audit used this fixture as its
example of an import with no losses, so it failed every run. Embedding
the comment, as GitLab's notes endpoint returns it, makes the import
complete again and exercises the new comment mapping in the browser.
Thanks to xet7 !
Maintainer decision of 2026-09-30: running rules, notifications and
webhooks for changes that Sync makes is off by default. A board opts in
with syncEffectsEnabled, and scheduled runs additionally need the new
instance flag enableSyncCronEffects, so an upgrade changes nothing on
its own. Every stored-effect stage (rules, rule email and archive,
notifications, webhooks) now requires its caller to name the trigger,
manual or scheduled, and checks both switches inside its guard with the
board it re-read, so whatever later wires Sync to these stages cannot
skip them. The switches have no interface yet: the stages are not
called by Sync until the History coordination they wait on is done, and
a toggle that does nothing would mislead.
tests/syncActivation.test.cjs pins the decision, the defaults and that
each stage checks it; the full-app suite refuses a missing trigger, a
board that has not opted in and a scheduled run without the instance
switch, writing nothing, then succeeds once both are on. 560 server
tests pass.
Thanks to xet7 !
Both were reported as losses because an import never grants access. A
watcher needs none: a source watcher now becomes a card watcher when the
members mapping maps them to a member of the new board. Anyone else -
unmapped, or mapped to a user who is not a board member and would
otherwise receive a private board's notifications - is counted in the
import's loss report instead. Nextcloud Deck sharing rules stay a
reported loss by design.
The planner tests pin board-member-only watching and the count of the
rest; the Asana and OpenProject tests pin the mapping from their
fixtures. A source guard for GitLab's Requested By follows the
gitlabUser helper introduced for the GitLab adapter.
Thanks to xet7 !
The GitLab adapter kept only title, description, state, due date, the
first assignee, author and labels, and silently dropped the rest of an
Issues API v4 issue. It now carries every assignee, the author, creation
and close dates, time spent, labels with details, milestone, iteration,
issue type and confidentiality as tags, weight, time estimate and task
completion as custom fields, embedded notes as comments (system notes
skipped), embedded links as dependencies, and the group/project#iid
reference and web URL at the end of the description, as GitHub cards
already have. A confidential issue is warned about; group-level epics and
comments not embedded in the export are reported as losses.
List Sync reads the same parser, so on the first Sync after upgrading an
existing GitLab card's description gains the source link once, the way
GitHub cards already carry it; Sync still writes only title, description
and time fields on existing cards, and its coverage report names where
each GitLab field goes.
tests/gitlabImport.test.cjs pins every mapping, the loss report and the
planner's handling of members, dates, comments, links and custom fields.
Thanks to gotjoshua and xet7 !
A rule email prepared before commands recorded their source (no source
binding), or one with card Details whose binding predates the related-
source chain, is refused by dispatch and was waiting for a recovery that
did not exist. Admin Panel -> Problems -> Recovery now lists them with
their board, card, recipients, subject and reason. Nothing runs them
automatically. An administrator re-binds one - its content is prepared
again from the card as it is now, after re-checking that the rule's
author is enabled with write access and that the rule and action are
unchanged - or discards it, which records a dropped attempt so the rule
stage completes without mail. Each decision writes one immutable audit
record, is serialized per command, and is final once an attempt exists.
Unit, MongoDB integration and source tests cover listing, re-bind,
discard and every refusal, and pin that only the admin methods call
them; a Playwright case covers review and discard. New interface text is
in English, pending translation.
Thanks to xet7 !
When SMTP accepts a rule email for only some recipients, or its answer is
lost, the attempt stays unconfirmed and WeKan never sends it again by
itself. Admin Panel -> Problems -> Recovery now shows each recipient as
accepted or not confirmed, and an administrator can resend to the
unconfirmed recipients only (same headers, narrowed SMTP envelope), mark
the attempt sent, or drop it. A drop completes the rule stage without
mail. Decisions wait ten minutes after the attempt started, longer than
any SMTP send can last, are one immutable record per attempt, and are
serialized per command across server processes. A resend records its
intent before SMTP, so an interrupted one is shown as in progress rather
than repeated silently. The offline acceptance tool and the online
decision exclude each other.
Unit, MongoDB integration and source tests cover resend, partial resend,
mark sent, drop, the quarantine and the refusals; a Playwright case
covers review and drop in the Admin Panel. New interface text is in
English, pending translation.
Thanks to xet7 !
TODO Later now answers the questions the Scrum/Sync and #2713 work was
waiting on: partial or unknown SMTP acceptance is resolved by an
administrator in Problems → Recovery and never resent automatically,
legacy unbound Sync commands are reviewed one by one in Admin Panel,
and Sync is off by default with per-board opt-in and a separate cron
switch. #4912, #2460 and #2509 stay open.
Thanks to xet7 !
Add missing source fallback keys in English order while preserving every
existing locale value. Share the placeholder parser so percent-encoded
URLs are not mistaken for format tokens.
Retain newer localized keyboard labels and the Bulgarian JSON example in
the correction ledger. Refresh the audit summary and report the actual
59 essentially complete languages in the README.
Locale inventories and focused structural tests pass. Translation filling
is not complete: 105652 actionable values remain across 176 locales.
Thanks to xet7 !
Keep trusted card snapshots intact after checking caller field permissions.
Restore custom-field activities, History and Sync updates; fix popup layouts,
textarea rule descriptions, WebKit date resets and mobile Rules mail rows.
Align Meteor versions, make image fixtures portable, configure local SMTP
for both test server modes, and bridge Docker Desktop database and SMTP
connections. Stabilize paginated, lazy-loading and security test fixtures.
Focused Node and Chromium, WebKit and Docker Firefox regressions pass.
The full Node rerun still reports translation completeness failures.
Thanks to xet7 !
The v12.10 release stopped in its bump job before a tag or GitHub
Release existed, so nothing was published under that number: the
section returns to "Upcoming WeKan ® release" for the next run to take
the same number, with the OpenAPI generator fix and CodeQL alert 546.
Thanks to GitHub CodeQL and xet7 !
CodeQL alert 546 (js/incomplete-sanitization): tests/mochaFullApp.test.cjs
escaped only `.` in a file name before building a RegExp, so a backslash
or another metacharacter would be read as pattern syntax. A complete
escapeRegExp now treats the name as data. Test-only code with no runtime
path, so there is no Hall of Fame row or Problems key.
Thanks to GitHub CodeQL and xet7 !
The release-all bump job stopped in openapi/generate_openapi.py with
"AttributeError: 'NoneType' object has no attribute 'type'": a REST
handler declares `let x;`, and the req.body walker read the declarator's
absent init. The walker now returns early on an absent node, and walks
the branches of a `?:` so fields read there (the board copy title,
among others) are documented again. downlevel_js also rewrites `||=`
and `&&=`, which esprima cannot parse and which made the generator skip
models/lib/importedAttachmentsByCard.js.
tests/openapiReqBodyWalker.test.cjs pins the guard, the ternary walk and
the rewrite, drives the walker on the shapes that broke it, and runs the
whole generator asserting no file is skipped.
Thanks to xet7 !
The trigger and action editors have an "Insert variable" list with the
card's people, card, list and board names and one {customField:Name} per
custom field of the board. Picking one inserts it at the caret of the text
field last focused. Admin-only fields are not offered, since rules do not
read them, and neither is a name the token syntax cannot spell.
Thanks to chessboards and xet7 !
Fixes#3195.
historyRecoveryNotice is a status line with Retry and Forget for one pending
keyboard undo/redo, not a second history view. The guard now pins that shape
(role="status", no table, no history rows) instead of failing on its name.
Thanks to xet7 !
A full-app Mocha run with METEOR_LOCAL_DIR=.meteor/local-test writes
public/build-assets-local-test-app-test/, which the exclude list did not
name, so the scan reported the bundled dependencies' URLs as new origins.
Every public/ and private/ build-assets* and build-chunks* directory is now
skipped, matching .gitignore; ordinary public/ source is still scanned.
Thanks to xet7 !
The Leo literate editor's XML outline is a new import source and export
format: top-level nodes are lists, their children are cards with the node
body as the description and a marked node as done, and deeper nodes are
checklists. Clones keep their headline and children. The raw XML is parsed
server-side with htmlparser2 in XML mode (no DTDs or external entities), a
self-clone cannot loop and a clone bomb is refused at 100000 nodes, and the
parsed tasks then pass the shared import sanitizer.
Thanks to xet7 !
The Kanboard, Nextcloud Deck, OpenProject, Asana, Zenkit and Jira exports
carried a card's title, description, due date and labels, while WeKan's own
importers for those tools read far more - so a board sent out and brought
back lost its comments, checklists, parent links, people, start/completion/
creation dates and custom fields. Each formatter now emits them, when
selected, in the shape its importer reads, and a suite runs every export
back through that importer. The formatters are a pure module; Kanboard's
separate route and builder are gone, so it has the export selection and
the admin-only custom field check every other format has.
Thanks to xet7 !
Since 6d99ad73c (#1598) deleting a card keeps its timestamped activities for
board reports, where #3252 had deleted them, so the test's expectation of a
single remaining activity was stale. It now checks what #3252 still
guarantees: the children go without one activity each, the card's earlier
activities stay, and exactly one deleteCard activity is added.
Thanks to xet7 !
"A must be done before B and C can start, so A is a subtask of both." A card
now keeps every parent in parentIds, with parentId still the primary one
(parentIds[0]), so everything that reads one parent keeps working and a card
written before this has its one parent. Picking an existing card as a
subtask adds this card as a further parent instead of replacing the one it
had; Card -> More lists the other parents and removes one, and choosing the
primary parent keeps the others. Children are found under any parent - the
subtask list and counter, the parent filter, rule email Details and the
exports - and the loop guard and the board publication walk every parent.
Archiving or deleting a card takes only the subtasks it is the one parent
of; a shared subtask just loses it. Every parent written must be on a board
the writer can see (GHSA-jvv9-498p-hxrg), and board copy and WeKan import
remap all of them. The old single-parent walker is removed.
Thanks to ikomhoog and xet7 !
Fixes#3626,
A board admin uploads an image - a floor plan, a site map, a drawing - and
the board's cards become markers on it, coloured by their first label and
numbered like the card. Members who can edit place a card by dragging it
from the "Not on the map" list, or by choosing it and clicking where it
belongs, and move a marker by dragging it; a click opens the card. A place
is stored as percentages of the image (card.mapX, card.mapY, 0-100), so a
marker stays on its spot at any size; the image is a board attachment
served under the board's own access rules. Read-only members see the map
but cannot change it. New strings are English, pending Transifex.
Fixes#3256,
Thanks to javen9881 and xet7 !
A Ctrl+Z whose reply was lost is kept and retried by the next keystroke with
the same request ID, but nothing on screen said it was still waiting. The
board now shows a notice while it has such a request, also after a reload:
Try again resends that same request, so it can never take back a second
change, and Forget it drops it. New strings are English, pending Transifex.
Thanks to xet7 !
Jira imports keep each issue's type in card.scrum.issueType, but it showed
only as a Scrum text field, and only on boards with Scrum turned on. The
minicard now shows it first in its badge strip on every board, with an icon
for the usual types - bug, story, task, sub-task, epic, improvement,
feature, spike, test - matched loosely, and a neutral one for a site's own
types; Scrum's text copy on the minicard is gone. Kanboard categories stay
labels.
Thanks to xet7 !
26 tests in server/lib/tests - notification delivery, stored history, Sync
hooks, rule plans and webhooks - skip unless Meteor.isAppTest, which only
--full-app sets, and every runner used plain `meteor test`, so they had
never run. build.sh and npm test now pass --full-app, WRITABLE_PATH (the app
will not start without it) and the one-second email receipt sweep one test
waits for: 559 pass, none pending. The one that had been failing unseen,
syncHookedCards, gave a card a value for a custom field defined on no board,
which the admin-only-field guard rightly refuses; it now creates the field.
A node guard keeps the runners in this mode and every such file loaded.
Thanks to xet7 !
Board Settings gains Nextcloud Deck's auto-archive: a board admin sets a
number of days from 1 to 3650, and an hourly job archives the board's cards
whose last activity is older than that, with their subtasks and an
archivedCard activity attributed to the board creator, as scheduled rules
are. Empty means off, templates are never archived, and each run takes at
most 200 cards per board. The same date drives the card-aging fade, so a
card fades before it goes. New strings are English, pending Transifex.
Thanks to xet7 !
Notification Settings chose where a notification goes, the bell and email,
but not what about. Member Settings → Notifications now lists the kinds of
card activity - labels, members, assignees, comments, moves, dates,
checklists, attachments, custom field values, archiving and creation - and
an unticked kind reaches neither the bell nor email. It is checked where
each recipient's channels are decided, when the plan is frozen and again
before delivery. Due-date reminders and @mentions always arrive. A suite
requires every named activity type to be in a group or deliberately
unmutable. New strings are English, pending Transifex.
Fixes#572,
Thanks to bhagyas and xet7 !
Checks every item of the umbrella issue against the current code and the
linked issues, marks what is already done, and gives the six remaining
requests as ready-to-run gh commands. Filing and closing issues is a
publishing step left to the maintainer.
Thanks to xet7 !
The panel's content has grown to about 1100px, so in the test's 900px window
it filled the height and dragging down was correctly clamped, which failed
the "moved down" check. The test now uses a 1300px window and also checks
that a long drag keeps the whole panel inside the window.
Thanks to xet7 !
A board's visibility can now be "Signed-in users" (permission 'instance'):
every signed-in user of this WeKan reads it, nobody signed out does, and
only its members edit it, as on a public board. One pure rule,
models/lib/boardPermission.js, feeds isVisibleBy and the shared selector
builder, so the board and card publications, All Boards, search, exports,
attachments, backgrounds, watching, filter presets and rule destinations
follow it. Anonymous paths - link previews, public avatars, the no-token
export branch and archived lists - stay public-only, and a source suite
lists every remaining public-only check so a new one is reviewed. The
Public page and GET /api/boards list instance boards for a signed-in caller
only; "private boards only" refuses them like public ones; a WeKan export
keeps the setting on import. New strings are English, pending Transifex.
Fixes#3249,
Thanks to relikd and xet7 !
The viewer now passes urlSchemes beside stripLinks, and the URL-scheme
field's hint sits between its label and the external-link keys.
Thanks to xet7 !
The uiFonts guard requires pixel sizes to follow the font-scale preset. The
grip's box is a fixed 24px square and its line-height only centred the
glyph, so flexbox now centres it at any scale.
Thanks to xet7 !
A trigger's list, swimlane, title, checklist and user fields may now use the
variables actions already resolve - {creator}, {assignees}, {members},
{customField:Name} and the rest - read from the card when the rule is
checked: "when a card is moved to the list named in its {customField:Stage}".
Such a value never equals what an activity carries, so the rule engine now
also fetches this board's triggers that hold a token and resolves them; every
other field keeps the ordinary exact-or-wildcard rule. People tokens match any
one of the named users, "by {assignees}" is kept as typed and compared by
username, and an unknown token stays literal. The editor explains this in a
new hint, English pending Transifex.
Thanks to xet7 !
The card's attachment gallery and the minicard cover downloaded every image
at full size. A raster image now has a thumbnail route: a WebP of at most
512 pixels, made with sharp under the GIF converter's input limits and kept
in a bounded in-memory cache keyed by file version. It is the attachment
route itself, after board access, the download block and limit, and the
storage read flag, so a thumbnail is readable exactly when the original is;
it is cached privately only. SVG and non-images have none, and a damaged
image redirects to its original. The gallery and cover use it.
Fixes#3275,
Thanks to blaggacao and xet7 !
Review and correct 139 older strings. Restore poker values 8 and 100, valid privacy markup, board and assignment wording, and complete time-adjustment help. Preserve unrelated values and add regression checks.
Thanks to xet7.
The test gave the import boundary a one-pass replace(/<[^>]*>/g, '') stub,
which CodeQL flagged as incomplete multi-character sanitization:
"<<script>script>" leaves "<script". The test now uses the server's own
sanitizer (server/lib/inputSanitizer.js) and checks that spliced markup does
not survive. A tree-wide suite requires every tag-stripping replace to loop
to a fixed point or to write non-HTML output (the PDF text writer), and
proves it catches the flagged stub. No shipped code had the one-pass form.
Thanks to GitHub CodeQL and xet7 !
Correct 231 Indonesian strings, including reversed rule directions, search case sensitivity and keyboard shortcut states. Translate newly added URL scheme help in 52 locales and test exact scheme names. Broader Indonesian vocabulary review remains open.
Thanks to xet7.
Admin Panel → Features → URL already stored "automaticLinkedUrlSchemes", but
nothing read it: eight schemes were hardcoded, recognised by linkify, then
refused by markdown-it's validateLink and both sanitizer passes, so none was
ever clickable. The listed schemes now pass all three filters and nothing
else does; the setting is empty by default. javascript, data, vbscript and
similar schemes are never linked, in the app's parser and again inside the
markdown package. A listed scheme opens its application in place rather
than through an empty tab.
Fixes#3218,
Thanks to chrisi51 and xet7 !
Fill 272 missing or English values and correct nine Malay-seeded basic labels. Verify source keys, interpolation tokens and recovery meaning. Older Indonesian prose still needs a Malay vocabulary audit.
Thanks to xet7.
A fenced block with a known language is coloured by highlight.js; a block
with no or an unknown language stays plain, as before. Both sanitizer passes
keep only hljs-* and language-* classes, only on code and span inside pre,
through a hook added for that call; every other class is still removed.
Thanks to xet7 !
Translate 530 missing or English values across both Romanian locales and preserve existing translations. Verify source keys, interpolation tokens, native terminology and recovery meaning.
Thanks to xet7.
The #3136 invite notification never ran: its params object was declared
inside the email block, so Notifications.notify() threw a ReferenceError
that its own catch logged. With it in scope, the tray would still have
refused an entry with no activity, and the email service would have queued
a second email beside the invitation. The invite now delivers the
membership's addBoardMember activity, which the bell already renders, to an
existing invitee's tray only, honouring their tray setting. A re-activated
membership records that activity; re-inviting an active member adds nobody.
Refs #3136.
Thanks to CondensedTea and xet7 !
"CUSTOM-IP:PORT" listed addresses with `ip address`, which macOS does not
have, and the CURRENT-IP options tried only en0 and en1 on macOS, so a Mac
whose network is on another interface got localhost. current_ip_address now
reads the default-route interface's address on macOS and the kernel's source
address on Linux; list_ip_addresses uses ip or ifconfig, whichever exists.
The custom option also offers the detected address as its default.
Thanks to xet7 !
Translate 527 missing or English values across both Dutch locales while preserving existing translations. Verify source keys, interpolation tokens and recovery meaning.
Thanks to xet7.
Translate 276 missing or English values, preserve existing translations, and verify source keys, interpolation tokens and recovery meaning.
Thanks to xet7.
Translate 268 missing or English values and preserve existing translations. Verify source keys, interpolation tokens and recovery meaning.
Thanks to xet7.
The markdown package rendered "- [ ] Task" as a disabled checkbox, but its
"only a task checkbox" rule sat in a HOOKS config key that DOMPurify does
not have, so it never ran. The card viewer then sanitized the output again
with a config forbidding every input, so no checkbox reached the page, and
forms.css hides bare checkboxes app-wide anyway. Both sanitizer passes now
install a real afterSanitizeAttributes hook for the call that keeps only a
disabled checkbox and removes every other input, and the viewer shows it.
The dead HOOKS keys are removed from all three sanitizer configs.
Refs #2419.
Thanks to rodrigocipriani and xet7 !
Translate 270 missing or English values, preserve existing translations and verify source keys, interpolation tokens and recovery meaning.
Thanks to xet7.
Fill 1,047 missing or English values across four German locales, using Swiss spelling for new Swiss German text. Preserve existing translations, source key order and interpolation tokens. Retain reviewed shared labels and mathematical notation.
Fourteen translation suites and 21 human-preference checks pass against current English keys.
Thanks to xet7.
Fill 1,287 missing or English values across five French locales. Preserve existing translations, source key order and interpolation tokens. Review unchanged shared French terms, keyboard labels and mathematical notation per key.
Thirteen translation suites and 21 human-preference checks pass against current English keys.
Thanks to xet7.
A board copy inserted each card with the source board's custom-field ids
and re-keyed them after cloning the definitions. Since the admin-only field
guard refuses a card whose field belongs to another board, duplicating any
board with custom fields selected failed with "Protected custom field".
The definitions are now cloned first and card copy inserts the new ids
directly; a value whose field was not cloned is dropped. The selective
duplication browser case that reproduced it passes again.
Thanks to xet7 !
copyBoard assigned every caller-supplied property onto the source board
object, including _id. board.copy() loads the source lists, cards and
children by this._id, so an admin of any board could pass a private
board's _id and receive a full copy of a board they cannot see - the
board-level form of CopyIdentityBleed. Only title, sort, type,
withoutCards and copyOptions are accepted now, each type-checked, and they
are applied to a separate object, so the cached source board is never
changed. Any other field is refused before anything is read and recorded
as CopyIdentityBleed in Admin Panel -> Problems; a wrong type on a
supported field is an ordinary input error without an attack record. The
REST copy route takes only a string title, on a separate object.
A unit test covers the contract, and a source scan fails if caller input is
assigned onto the source board or the source board object is copied again.
The browser spec reproduces the disclosure and checks the refusal, the
Problems record, that malformed input blocks nobody, and that supported
copies still work.
Thanks to xet7 !
Fill 1,685 missing or English values in six Simplified Chinese locale files, including reviewed Blockly and rule editor text for Singapore. Preserve existing translations and source interpolation tokens; retain reviewed keyboard legends and product names.
Twelve translation suites and 21 human-preference checks pass against current English keys.
Thanks to xet7.
A SAML login that would take over an existing non-SAML account with the
same username (SAML_MERGE_EXISTING_USERS off) tripped the
saml.account-conflict canary, and the REST comment-edit refusal tripped
comment.foreign-edit, but neither id was in the canary catalog. Both
attempts were recorded in Admin Panel -> Problems as an "unknown" canary,
without their category, severity or description. They now have entries:
saml.account-conflict under a new authn.saml-link category
(SamlAccountMergeBleed, the SAML counterpart of CasAccountMergeBleed), and
comment.foreign-edit under the existing comment authorization category.
The coverage test only checked that catalogued canaries were tripped
somewhere. It now also scans every canary call in the app and its local
packages and fails on any id that is not catalogued; with either entry
removed it names the id and the file.
Thanks to xet7 !
Fill 796 missing or English values across Traditional Chinese, Taiwan and Hong Kong locales. Preserve existing translations, source key order and interpolation tokens; retain reviewed keyboard legends and product names.
Eleven translation suites and 21 human-preference checks pass against current English keys.
Thanks to xet7.
Upcoming gains several-trigger rules (completing #4294), the interface
fixes found by source guards, the move-selection picker, and the test and
tooling repairs. TODO Later lists what is paused and not in this release:
the viewer sanitizer hook fix and code highlighting (parked as patches),
suites still failing, two issues found but not fixed, and the decisions not
yet built.
Thanks to xet7 !
Fill 800 missing or English values across three Japanese locales, including current rule and positioning text. Preserve existing translations and source tokens; retain reviewed technical labels and translate boolean display names.
Ten translation suites and 21 human-preference checks pass against current English keys.
Thanks to xet7.
The mixed-selection move picker showed its before/after choices as arrow
characters inside <option> elements, where no icon can be drawn; they are
now the words Before and After (English, pending translation). The Move
selection popup opened with an empty header, so it had no title or close
button text; both places that open it now pass the translated title. The
menu inventory records the due-reminder Save as a form submission. Tests
cover each, including negative checks that the arrows and the untitled
popup do not return.
Thanks to xet7 !
The archive-chart tooltip and the menu-column popup used physical left/right
offsets, which do not mirror in right-to-left languages; they now use
inset-inline-start/end. A board-settings column label used a fixed 12px
font size that ignored the interface font scale. The Time view's
adjustments table printed card titles directly instead of through the
sanitized Markdown viewer the rest of the view uses; a negative test now
requires every rendered title to be inside the viewer. The card-copy
override refusal logs in the catch shape the security guard requires.
The REST login and impersonation guards now pin insertActiveLoginToken,
which replaced Accounts._insertLoginToken so a disabled account cannot get
a token, with new negative checks that no REST path mints a token without
that check.
Thanks to xet7 !
Eighteen suites had been failing because the code they read had changed on
purpose: notification delivery moved to stored plans, exports gained the
admin-only field guard and Scrum transfer, board copy became selectable,
card copy re-keys filtered custom fields, the Scrum settings menu entry was
added, new operator-only scripts, directory-login canaries, a rate-limit rule
that made the first mention of undoLast no longer the method, rule parts and
Sync-scoped recording in hooks evaluated by test harnesses, per-board card
dragging, and a Dependabot file-type bump. Each guard now pins the new
behaviour with a comment naming the change, keeps its original assertions,
and several gained positive and negative cases (export field guard, copy
defaults, Sync-scoped recording, card dragging, method lookup).
Two list-sync browser specs logged a second context in with the same
session token; they now get their own tokens, and the token guard counts
distinct pages instead of logins.
Thanks to xet7 !
Fill 575 locale values across both Korean files and the new before/after labels in all 23 completed locales. Preserve existing translations and source placeholders; retain reviewed keyboard legends and mathematical symbols.
Nine translation suites and 21 human-preference checks pass against the current English source.
Thanks to xet7.
58d166029 gave the popup an onCreated hook for its due-reminder toggles,
and this suite's Blaze stub had no onCreated, so it failed before any
assertion. The stub now records it, and the suite checks that the
reminder section is board-only, that a valid entry is sent, that an
out-of-range day is refused before any server call, that "no reminders"
saves an empty list and that an empty field restores the server default.
Thanks to xet7 !
Fill 528 missing or English values across both Galician locales, including the new rule editor keys. Preserve existing translations, source key order and interpolation tokens; keep reviewed shared technical labels unchanged.
Eight translation suites and 21 human-preference checks pass against the current English source.
Thanks to xet7.
Document the completed locales, passing tests and remaining 215-locale backlog. Newly added source keys remain part of the translation goal.
Thanks to xet7.
Fill 1,341 locale values: finish four Portuguese locales with regional terminology and reviewed Blockly text, and translate four newly added rule editor keys in all 19 completed locale files. Preserve existing translations and interpolation tokens.
Seven translation suites and 21 human-preference checks pass against the current English source. No remote writes.
Thanks to xet7.
Maintainer decision for #4294 and #2953: a rule keeps its own trigger and
action and may add up to ten more of each. It fires when any of its
triggers fires - once per activity, even when several match - and runs its
actions in order, its own first. Rule details lists the extra parts with
Add another trigger / Add another action and Remove; the new
rules.addPart/removePart methods keep the board-admin and cross-board
destination checks of rules.createRule and refuse a button as an extra
trigger.
Ordinary, scheduled and button rules, the durable Sync rule plan, rule
History (snapshot, part edits and restore), deletion, board copy, WeKan
export/import, the rules JSON export/import, the REST representation and
the admin rules report all handle every part. Two rule-engine test
harnesses now receive the new helpers they evaluate.
New strings are in en.i18n.json, pending translation.
Thanks to xet7 !
Fixes#2953,
Translate 260 missing or English values, preserving existing translations and exact interpolation tokens. Review unchanged product names, keyboard legends, mathematical symbols and the Sprint term per locale.
Italian completeness, source order and token checks pass. Recheck Finnish, Arabic, Turkish and Spanish against current English keys; all six translation suites and 21 human-preference checks pass.
Thanks to xet7.
Record tested completion of ten more locale files and the 220-locale backlog. Newly added English keys remain in scope for every target language.
Thanks to xet7.
Fill 2,815 missing or English values in Turkish and nine Spanish locales. Reuse reviewed Spanish Blockly accessibility wording for the Colombian locale, while preserving every existing translation.
Check all source keys, interpolation token inventories and remaining placeholders. Keep only reviewed shared vocabulary, keyboard legends, product names and math notation unchanged. Turkish and Spanish regression suites, all-locale URL/token validation and human-preference checks pass.
Thanks to xet7.
Document the four completed locales, passing regression checks and remaining translation backlog. Wrap two existing Upcoming topic descriptions to satisfy the changelog format check.
Thanks to xet7.
Fill 1,020 missing or English source values across Finnish and the Arabic, Algerian Arabic and Egyptian Arabic locales. Preserve existing translations and review unchanged Blockly product names and mathematical symbols per locale.
Verify key order, source placeholder inventories, localized prose and zero remaining placeholders in these four locales. Translation URL/token suites and human-preference verification pass.
Thanks to xet7.
Another agent translates the locale files, so writing English copies into
all of them raced its edits (and one commit swept its uncommitted
translations in, since rebuilt). add-pending-keys.mjs now writes
en.i18n.json and pending-transifex.json only; the interface falls back to
English for a key a locale lacks. --all-locales keeps the old behaviour.
Thanks to xet7 !
#4278 asks for rule reminders to assigned people when a card is due,
overdue or starting. Scheduled due triggers existed; the trigger form now
chooses Due Date or Start, so a rule can fire when a card starts within N
days or should have started N days ago. With {assignees} in the email
recipient the reminder goes to the card's assignees. The date filter is a
pure module with unit tests; triggers saved before keep the due date.
Thanks to xet7 !
Fixes#4278,
a91855993 routed member usernames through ruleUsernames() and the email
recipient through recipientVars(). ruleEmailFailure runs RulesHelper in a
VM with each free name injected, so the missing recipientVars made the
email path throw and the test wait forever; it now injects an identity
recipientVars. Two guards pinned the literal
getUser({ username: action.username }) line; a literal username is still
looked up unchanged, so they now pin the per-username lookup and keep
their defensive-resolution and crash-pattern assertions.
Thanks to xet7 !
Maintainer decision: rule actions reuse the email {token} syntax.
{creator}, {assignees} and {members} name the triggering card's people -
usernames in text, email addresses in the send-email recipient, where
entries without an address are dropped - and {customField:Name} reads the
displayed value of a custom field, excluding admin-only fields. Member
actions accept a token that names several people and act on each; list and
swimlane names of move actions are resolved too. Literal values behave
exactly as before and unknown tokens stay literal.
The email action shows the new tokens in a second hint, English pending
Transifex.
Thanks to xet7 !
SAML 2.0 lets an identity provider sign the Response, the Assertion or
both. WeKan always required a signed Response (node-saml's default, never
overridden), so an identity provider that signs only the Assertion was
rejected with "Invalid document signature" before the Assertion was
checked. Admin Panel -> People -> SAML now has SAML_IDP_PROFILE: standard,
which keeps the previous behaviour, or signed-assertion-redirect, which
requires the Assertion signature instead and logs in by full-page redirect.
SAML_WANT_RESPONSE_SIGNED, SAML_WANT_ASSERTIONS_SIGNED and SAML_LOGIN_FLOW
can each be set on their own; left at Default they follow the profile. A
configuration requiring neither signature is refused, and validation
errors now reach the administrator instead of "Internal server error".
In redirect mode the ACS stores the validated profile under the random
credential token and redirects to /sign-in?samlToken=; the page exchanges
it and opens /. The assertion never enters the address. The token is
remembered in the tab that started the login, and a samlToken that tab did
not start is not exchanged, so a link cannot sign a browser into another
account. Errors return as samlError and are shown on the sign-in page.
A test signs responses with a throwaway key and runs them through the
shipped node-saml; browser tests log in through the fixture identity
provider in both flows. The new message is English, pending Transifex.
Thanks to xet7 !
Board Settings -> Notifications now has due-date reminders: a board's own
offsets override NOTIFY_DUE_DAYS_BEFORE_AND_AFTER (an empty list turns
them off, an empty field restores the server default), and a switch sends
the board's reminders to all of its enabled outgoing webhooks, whatever
activities those subscribe to. The reminder scan now runs when either the
environment variable or a board sets offsets, and writes a reminder only
for the offsets that board uses. A validated method limited to board
admins stores the setting.
The new strings are English, pending Transifex. add-pending-keys.mjs now
writes symlinked locales once and checks every file before writing any.
Thanks to xet7 !
Fixes#5323,
Maintainer decision: features that need new strings add them in English
to every locale rather than waiting for translation. add-pending-keys.mjs
inserts the keys into en.i18n.json and every locale in order and lists
them in pending-transifex.json; fill-translations.mjs --missing counts
them separately, and a Transifex translation replaces the English text
through the ordinary pull merge.
TODO Later now records the other decisions: rule variables, any-trigger
rules, the Scrum/Sync journal approach, the thumbnail route, semi-open
boards, the parents array, highlighting, the URL-scheme allowlist, Jira
issue types, the Map view, splitting #4790, #5323's parts, and what is
not planned now.
Thanks to xet7 !
The catalogue had one date range and no Timestamp or top-level ts field,
the OpLog tail's shape, so a backend that answered those wrong or never
pushed them down passed. The seed now carries Timestamps with different
increments, a plain number, a string and a missing value under ts, and
integers of both MySQL JSON types under big. The new cases found that
FerretDB's MySQL backend dropped 1577934245000 from a range (typed
UNSIGNED INTEGER); fixed in wekan/FerretDB, after which all 110 cases agree
on SQLite, PostgreSQL 18, MySQL 9.7 and MariaDB 12.3.
Thanks to xet7 !
The CLI test expected the fresh-clone path, but the mirror tool looked for
an existing checkout under the real .tools directory, so the test failed
on any machine that had .tools/wekan-gitlab. WEKAN_MIRROR_TOOLS_DIR now
relocates the checkouts and the test points it at its own temporary
directory.
Thanks to xet7 !
Summarizes what this pass completed and groups what remains by the reason
it is blocked: new interface text that would need translating into every
locale, maintainer decisions, infrastructure or affected data (Docker
Desktop here waits for an administrator to approve its privileged helper),
and architectural Scrum/Sync work. Records Node suite health against the
pass's starting commit.
Thanks to xet7 !
Ctrl+Z and Ctrl+Y called changeHistory.undoLast/redoLast without a
request ID, so a reply lost to a disconnect or reload left the user
pressing again and undoing a second Scrum change. Each keystroke now
carries an ID kept in sessionStorage until the server answers; the next
keystroke on that board retries the unanswered ID first, and in the same
direction that retry is the keystroke. Unanswered IDs expire after ten
minutes. A non-Scrum row, which keyed requests do not cover, is refused
without change and the keystroke falls back to the unkeyed call, which
the Scrum design now documents as a deliberate exception.
Thanks to xet7 !
Adds the Upcoming entries for the import commits, updates In short, and
rewrites the TODO Later import/export entry to what remains: showing the
loss report on the import page needs a translated string, file contents
need live API connectors, Deck/OpenProject/Asana sharing and watchers have
no safe mapping, Zenkit's native export is unverified and the matching
exporters were not revisited.
The Requested By guard now reads the planner, where KanboardCreator's
requested_by mapping moved, and still pins both of its assertions.
Thanks to xet7 !
Every external parser and the shared planner returned unsupported and
warning entries, but nothing stored them, so an import that dropped a
board's sharing rules or a parent outside the file looked like a complete
one. The Kanboard-shape and Jira creators now record one bounded
import-completed-with-warnings Recovery row with the board, the importing
user and the first paths and reasons; a complete import records none.
The import page does not show the report yet: that needs a new
translated string.
Thanks to xet7 !
JiraCreator dropped comments, sub-tasks, the parent hierarchy, priority,
components, fix versions and ordinary custom fields, and sorted every card
at -1. It now imports comments with ADF bodies as text, shown by the
author's display name and posted by the mapped user when the account id
was mapped; links sub-tasks and child issues imported together to their
parent and lists the others as a Sub-tasks checklist; makes priority,
components and fix versions board labels; imports customfield_ values
named from the search response's names map, except the field already
mapped as the estimate; and keeps the search order.
The checklist and comment inserts are shared with the other external
importers. jiraPageWarnings recognizes an incomplete enhanced or classic
search page for the import report.
Thanks to xet7 !
Zenkit documents no single-file JSON export schema, and the parser read
only an adapter shape with titles, stages, due dates and tags. It now also
reads the Zenkit API's entries and elements, with value keys as the zenkit
client deserializes them: stages from the stage categories element, labels
from other category elements, description, due and start dates, persons as
members, the hierarchy element as parent links, dependencies, text, number
and date fields as custom fields, checklists, creation dates and sort
order. Deleted entries are skipped. Files, formulas, cross-list references,
comment counts and element kinds with no documented value format are
reported rather than guessed.
The adapter shape gains ids and parents, several assignees, item-level
custom fields, comments and checklists, and reports keys it does not know.
Thanks to xet7 !
The Asana parser kept name, notes, section, due date, assignee and tags.
It now keeps start, completion and creation dates, custom-field values of
every Asana kind (number, text, enum, multi-enum, date, people and the
display value of formula fields), comment stories with their authors,
dependencies as blocking links, and the task permalink. Subtasks fetched
as tasks of their own become child cards; subtasks only embedded on their
parent become a checklist. Followers and attachment metadata are listed in
the unsupported report.
Thanks to xet7 !
The OpenProject parser kept subject, description, due date, status,
assignee and type. It now also keeps start and creation dates, spent time
from ISO 8601 durations, priority, category, version, responsible and
author, embedded comments, the parent hierarchy and embedded relations.
Custom fields take their names from embedded schemas; estimated time and
progress become number custom fields instead of being lost. Watchers and
attachment metadata are listed in the unsupported report.
External imports can now carry custom-field values, parent links and
dependencies. The creator makes one board custom field per source field,
typed number, checkbox or text from the values present, and links cards
once every card exists. Links resolve only within the same import;
unknown targets, duplicate ids and parent cycles are reported, never
guessed.
Thanks to xet7 !
The Deck parser kept title, description, due date, one assignee and label
names. It now keeps every assignee as a card member when mapped, comments
embedded from the OCS comments API, archive state, done and creation
times, and stack and card order. Trashed stacks and cards are skipped and
counted in warnings. Sharing rules never become board members, because a
file naming a user must not grant access; they and Nextcloud attachments
are listed in the unsupported report.
External imports now keep the source order of cards instead of giving
them all sort 0, and create checklists and comments through the writer
that keeps their dates.
Thanks to xet7 !
Every WeKan schema's createdAt autoValue replaces the value on insert, and
collection2 cleans the inserted document in place. The WeKan, Trello, Jira
and Kanboard importers passed each source creation date to such an insert,
so the #1992 date restoration and Trello's board, card, list and comment
dates were all reset to the import time. A WeKan export of a card created
in 2020 re-imported as created today.
writeImportedEntity now reads createdAt before the insert and writes it
back through the raw collection, and the creators' board, card and comment
inserts go through it. A source audit requires every dated creator insert
to use it, and a browser test re-imports a WeKan export and compares the
card's creation date.
Thanks to xet7 !
The assembled Kanboard JSON-RPC document reached KanboardCreator unparsed,
so only fields Kanboard spells like the shared task shape survived, and a
task identified only by column_id landed in the first list. parseKanboard
now resolves column, swimlane and category ids, turns subtasks into a
checklist and comments into card comments by their author, archives closed
tasks, maps Kanboard color ids onto the WeKan palette, and keeps start,
completion and creation dates, time spent, priority and the task URL.
Estimates, files and task links are listed in the unsupported report.
Thanks to xet7 !
Every external import goes through KanboardCreator, which wrote a card and
nothing else, so a parser had nowhere to put a subtask, a comment, a start
or end date, a closed state or a card color. The normalized task now
carries them, models/lib/importedTaskPlan.js decides what each becomes, and
the creator inserts checklists, comments with their addComment activities,
and the card fields the plan returns.
Dates that do not parse, and Kanboard's "0" for no date, are no longer
stored as Invalid Date or 1970. A comment by an unmapped author is posted
by the importing user with the author's name leading the text.
Thanks to xet7 !
The import boundary returns a new value with prototype keys and active
markup removed; it does not clean its argument. The Deck, OpenProject,
GitHub, GitLab, Gitea/Forgejo, Asana, Zenkit and Markdown branches of
importBoard parsed the raw upload instead, so cards kept the markup that
Admin Panel -> Problems recorded as sanitized. The Trello API import named
its workspace and job result from the raw board the same way.
The new source test flags any import-direction boundary whose raw input
is used afterwards, anywhere under models/ or server/, and proves its
detector finds the code as it was before this change.
Thanks to xet7 !
Require existing destination containers on the requested board before sort
reads, number allocation, field remapping and child copies. Reject soft-deleted
containers and return a clear REST input error without security-account blocks.
Preserve board-wide lists and archived-container board-copy behavior.
Cover invalid and valid placement with unit, DDP, REST and browser tests,
including existing ordinary, rule, Scrum, list and swimlane copy paths.
Record the remaining source-access and concurrent-write review.
Thanks to xet7 !
Prevent caller overrides from replacing the authorized card identity before
private children are loaded. Apply title and description strings to a
separate transformed card; refuse all other fields before copying.
Summarize extra-field attempts as CopyIdentityBleed while treating malformed
text as an ordinary input error. Add unit, source-scan and live browser
regressions, and document the reproduced disclosure and remaining review.
Update the shared-form regression for the existing multiline title editor.
Thanks to xet7 !
Add a UTC annual archive grid with count shading, colored label tooltips,
year selection and keyboard/touch details. Scope counts and labels to the
board and each assigned-only viewer, and exclude restored or undated cards.
Verify calendar aggregation with unit tests and real MongoDB/DDP/Chromium
coverage for the UI, all assigned-only roles, invalid input and isolation.
Document current-archive semantics and the remaining issue-audit work.
Thanks to xet7 !
Fixes#5444,
Document the tested optional request-ID API and current recovery guarantees. Keep keyboard persistence, non-Scrum replay, retention, shared writer coordination and the rest of TODO Later explicitly open.
Four changelog and release-preflight suites pass. Thanks to xet7 !
Accept optional caller-persisted request IDs in the History methods. Store immutable selections before mutation and reconcile completed requests without advancing the stack again. Keep empty and unsupported results stable; require fresh access and retain conflicting or missing recovery evidence.
Verify selection races, lost cleanup, concurrent redo, browser reload and private storage with Node, MongoDB, full-app and Chromium coverage. Keyboard request persistence, non-Scrum replay and shared writer coordination remain unfinished.
Thanks to xet7 !
Add the tested completion increment to Upcoming and keep public request identities, recovery controls, retention and writer coordination explicitly unfinished. Move the previous cleanup entry out of v12.09 because its implementation postdates that release, preserving the entry text. Repair the existing editor-entry line wrapping.
Four changelog and release-preflight suites pass. Thanks to xet7 !
Bind immutable private receipts to the original operation, actor, source hash and checkpoint plan. Reconcile uncertain cleanup from saved completion without rewriting History or deleting successor work. Reject damaged evidence and changed plans.
Cover false and lost acknowledgements, restart replay, finalization corruption and browser privacy with Node, MongoDB, full-app and Chromium tests. Public request identities and cross-document coordination remain unfinished.
Thanks to xet7 !
Record confirmed checkpoint cleanup and its unit, MongoDB, full-app and Chromium coverage. Retain shared coordination, durable completion and uncertain-read recovery as unfinished work.
Thanks to xet7.
Read the board checkpoint after conditional deletion. Reject false acknowledgements, unreadable results and successor checkpoints; reconcile a lost reply only after confirmed absence.
Cover cleanup failures and retries in unit and full-app tests, preserving the undo timestamp and one restoration event. Document remaining cross-document and restart coordination limits.
Thanks to xet7.
Record validated fixes, regression coverage and compatibility limits in Upcoming. Keep the broader TODO Later implementation queue paused.
Thanks to Hama1cco and xet7 !
Fix AdminFieldBleed (GHSA-m8gh-2h78-f57x): redact protected values from publications, API responses, method results, History and analytics; constrain value searches and refresh live permissions.
Guard inserts, full-array modifiers, copying and shared definitions at the request write boundary. Require authorized exports and summarize explicit attacks in Admin Panel Problems. Add positive, negative and Chromium regression coverage.
Thanks to Hama1cco and xet7 !
Read the validated node-saml inResponseTo property and retain single-use response checks. Forward ACS errors to the sign-in form without exchanging missing credentials.
Verify real signed-response login, replay rejection, popup errors and safe UI text rendering. Document deployment-specific identity mapping.
Thanks to xet7.
Retain an immutable operator decision before repairing the exact attempt receipt. Preserve dispatch guards and never send or reset mail. Cover interruption, conflicting decisions and real MongoDB CLI recovery. Other operator recovery remains unfinished.
Thanks to xet7.
Discover comment owners before reading prose, enforce card and assigned-only access, and retain each included source chain for stored dispatch checks. Reuse public comment rendering without exporting target checklists or files; reject oversized captures instead of truncating.
Thanks to kabi178, bbyszio and xet7 !
Include local archive state and lifecycle timestamps under the linked-card local section. Keep card number and color on the source selected by existing display getters.
Thanks to kabi178, bbyszio and xet7 !
Reuse ordinary Details rendering for explicitly local fields, including custom-field visibility and definition fingerprints, public names, notes, locations and authorized relationships. Persist related-source evidence in the existing email binding.
Thanks to kabi178, bbyszio and xet7 !
Read archive, due completion and overtime from the target board. Resolve only public names for unique active members, matching linked-board Members and Assignees getters, and recheck target access.
Thanks to kabi178, bbyszio and xet7 !
Classify all 71 top-level schema fields with an AST-backed inventory guard. Record remaining linked-board display and per-card discussion gaps explicitly.
Thanks to kabi178, bbyszio and xet7 !
Remove the duplicate spent-time setter that shadowed linked-board support. Await time credit before clearing sessions, and isolate automatic Pomodoro transitions from Tracker with one pending transition per view.
Thanks to xet7 !
Separate wrapper placement, persisted timers and visible Scrum metadata from authorized source content. Recheck local access, link identity and Scrum visibility before returning the snapshot.
Thanks to kabi178, bbyszio and xet7 !
Exercise authenticated REST moves with filesystem and GridFS attachments, all content choices, and negative authorization and destination matching cases.
Thanks to kabi178, bbyszio and xet7 !
Group readable targets while retaining distinct relationship types and source-access checks. Cover malformed legacy arrays and real SMTP rendering.
Thanks to kabi178, bbyszio and xet7 !
Describe captured filenames, actual sizes, MIME types, cover status, upload dates and public uploader names. Recheck card access and cover state after metadata reads and reject deleted cards or foreign attachment rows.
Thanks to kabi178, bbyszio and xet7 !
Share related-source resolution between Details and checklist discussion. Retain reference evidence for stored messages even when Details is disabled, and omit inaccessible targets.
Thanks to kabi178, bbyszio and xet7 !
Resolve related link titles through current readable sources and retain their chains in immutable command evidence. Recheck related access, admin rights and custom-field definitions before stored dispatch. Refuse legacy Details snapshots without related evidence.
Thanks to kabi178, bbyszio and xet7 !
Decode valid reaction codepoints and count distinct reactors without exporting account identifiers. Keep reactions scoped to included live comments and record remaining related-content authorization work.
Thanks to kabi178, bbyszio and xet7 !
Include public creators and comment authors, stickers, lifecycle metadata, checklist deadlines and reset schedules. Record remaining full-content gaps against the models and card export layout.
Thanks to kabi178, bbyszio and xet7 !
Resolve sprint and release names within the source board and recheck references after preparation. Persist Scrum disclosure flags with source evidence for stored sends.
Thanks to kabi178, bbyszio and xet7 !
Use live target results with both boards' visibility settings. Persist target disclosure policy in version-three source bindings and recheck it before dispatch.
Thanks to kabi178, bbyszio and xet7 !
Keep private voter names and unfinished poker choices out of Details. Persist voting visibility with source bindings and reject changed disclosure settings before stored dispatch.
Thanks to kabi178, bbyszio and xet7 !
Export persisted Flowtime and Pomodoro fields and public session owner names alongside recurrence settings. Cover zero values, malformed fields, access loss and SMTP delivery through both attachment backends.
Thanks to kabi178, bbyszio and xet7 !
Capture versioned source evidence with immutable mail and verify its topology and current access before dispatch. Keep legacy commands readable without authorizing an unbound retry.
Thanks to kabi178, bbyszio and xet7 !
Use live source card data across email text, details, discussion and files. Resolve linked-board display fields without including unrelated source-board cards. Recheck each link and its permissions after preparation, and reject cycles, missing sources or retargeted links.
Refuse linked stored Sync commands until durable source-chain bindings exist, including during dispatch of previously captured commands. Fifty-seven Node suites pass without skips, with real MongoDB and native MailComposer; local SMTP browser scenarios pass.
Thanks to kabi178, bbyszio and xet7 !
Render dates, placement, labels, people, custom-field display values, notes and readable relationships behind an opt-in Details checkbox. Recheck source and related access, admin-only field policy and field definitions; do not serialize account secrets or stale linked-source snapshots.
Thirty-six Node suites and two local SMTP browser scenarios pass. Linked-source details and specialized fields remain open under #2713.
Thanks to kabi178, bbyszio and xet7 !
Offer an independent opt-in choice, render checklist completion and chronological comments, exclude deleted or foreign content and private webhook state, and recheck access after child reads. Bound output without silent truncation.
Verify 35 Node suites and two local SMTP browser scenarios, including disabling the option. Other card metadata remains open under #2713.
Thanks to kabi178, bbyszio and xet7 !
Add an opt-in native checkbox, read live triggering-card files through the storage strategy, recheck source access and attachment metadata after reads, and preserve byte snapshots through email localization and transport.
Verify filesystem and GridFS MIME delivery with local SMTP capture, soft deletion, missing-file failures, keyboard selection, permission changes and durable attachment integrity. Full card-content rendering remains open in #2713.
Thanks to kabi178, bbyszio and xet7 !
Add bounded backend-neutral stream snapshots with canonical base64 payloads, sanitized names and cleanup on read errors. Accept validated byte snapshots in stored Sync mail commands while retaining text-size and integrity checks.
Rule UI, authorized storage reads and transport integration remain pending for #2713. Thirteen focused Node suites pass.
Thanks to kabi178, bbyszio and xet7 !
Reuse ordinary card copying, preserve the source, validate destination placement and current write access, and stop recursive copy chains while allowing independent events.
Verify automatic and manual copies, checklists and comments, revoked access, forged destinations and same-board recursion in Chromium. All 29 rule Node suites pass.
Thanks to javen9881, phaseshift3r, gugmt15 and xet7 !
Fixes#4160,
Fixes#3235,
Move the completed #935 item out of TODO Later and record source authorization, reactive dates, browser coverage and remaining broader work.
Thanks to neooleg and xet7 !
Use one-hop source dates for ordering and badges, observe source data and permissions, and recheck them after each scan. Handle partial linked-board member metadata without breaking row rendering.
Cover source edits, access revocation/restoration, assigned-only readers, source moves, all four date fields and ordinary lazy pagination.
Thanks to ghost, neooleg, Creanimo, GusevVictor, avatar1024 and xet7 !
Fixes#935,
Search and sort projected board metadata before fetching a 25-card page. React to board/card/list/swimlane edits, retract revoked pages, and nest table cells beneath their rows. Verify pagination beyond 5000 cards and document remaining linked-source date ordering parity.
Thanks to neooleg and xet7 !
Support keyboard selection, indeterminate excluded labels and explicit due-date reset. Preserve saved combinations and document the separate Table-view lazy-loading gap found during regression checks.
Thanks to neooleg and xet7 !
Pair label matches with their owning board, keep missing-label queries empty and preserve numeric card alternatives. Compare overdue deadlines with the current instant instead of UTC midnight. Verify combined searches across authorized and private boards in Chromium.
Thanks to GusevVictor, neooleg and xet7 !
Connect bundled filter templates, selectors, scoped resets and validated preset restoration through one registry. Use the recency controls as a built-in provider and verify extension lifecycle in Chromium.
Thanks to neooleg and xet7 !
Capture and restore all current filter controls with validated versioned state, refresh relative dates when applied, and reject invalid expressions before changing live filters. Persist same-name replacements atomically, restrict writes and subscriptions to their owner and board access, and clean preferences on board or account deletion.
Thanks to Creanimo, avatar1024, neooleg and xet7 !
Use local calendar boundaries and the user's week start, refresh active shortcuts with the shared clock, and release the subscription on switching or clearing. Cover month-end and year rollover plus actual sidebar boundaries.
Thanks to neooleg, Creanimo and xet7 !
Match any recorded list, swimlane or incoming board move within inclusive local-day bounds, regardless of later moves. Reuse the authorized reactive ID publication for text and history, preserve lazy-loading coverage and validate history joins and live UI behavior.
Thanks to neooleg and xet7 !
Match literal text across the authorized board, including cards outside lazy windows. Publish only matching IDs, refresh on content and access changes, and release observers when filters change. Verify the batched join, cancellation and permission races, and actual Chromium behavior.
Thanks to neooleg and xet7 !
Provide independent recent-day, week, thirty-day and older presets with a shared reactive clock, strict date values and composable card selectors. Verify timer cleanup and the real sidebar in Chromium.
Thanks to neooleg and xet7 !
Document verified sidebar date-range controls and retain the unfinished historical movement, combined text, relative preset and saved-filter requirements in TODO Later.
Thanks to neooleg and xet7 !
Choose inclusive local calendar bounds for seven card dates without an advanced expression. Preserve valid filters after invalid input and combine ranges with existing filters. Cover selectors and the actual sidebar in Chromium.
Thanks to neooleg and xet7 !
Add explicit built-in date names, missing-date comparisons and typed date guards while preserving quoted custom-field names and the shared date parser. Cover sidebar selection and actual rule execution.
Thanks to TiibCD and xet7 !
Fixes#1915,
Stream recorded movement histories during the versioned startup repair. Fill only missing dates with placement-bound conditional writes, preserve unknown or contradictory histories, and keep unresolved concurrent changes retryable.
Thanks to grasshide and xet7 !
Fixes#1499,
Keep column age stable across unrelated edits and record moves in the same card write. Add a rolling, board-scoped hide filter while retaining legacy cards with unknown entry dates. Historical activity backfill remains pending.
Thanks to grasshide and xet7 !
Refs #1499.
Resolve public card, board, list and swimlane titles reactively in detail and minicard views. Preserve existing regex formatting and constrain linked source placement metadata to its authorized board.
Thanks to solunix and xet7 !
Fixes#3815,
Remove rendered payloads immediately after matching intent completion and in bounded background sweeps. Require valid checksums and dispatch identity, retain pending or damaged plans, and use atomic in-place replacement with readback for uncertain acknowledgements.
Validated with 20 Node/MongoDB cases, three additional Node suites, four Meteor cases and thirteen Chromium delivery and cleanup scenarios. Operator handling and unresolved-payload retention remain open.
Thanks to xet7.
Scan bounded pending-intent batches after startup, advance past failed and busy rows, and coordinate immediate delivery and recovery through renewable per-intent reservations. Resume saved plans or prepare the first plan from an unchanged persisted activity; retain unresolved orphans without recreating activities.
Validated with 14 Node/MongoDB cases, three additional Node suites, four full-app Meteor cases and twelve Chromium scenarios including scheduled local SMTP recovery. Operator controls and payload retention remain open.
Thanks to xet7.
Save a checksummed plan before delivery, bind the original activity and dispatch actor, and reuse tray receipts and email job identities on retry. Recheck current recipient scope and channel preferences without rebuilding saved content.
Validated with ten MongoDB cases, three Node suites, three Meteor cases and ten Chromium notification scenarios. Background replay, ownership and plan retention remain open.
Thanks to xet7.
Await private intent persistence before the activity insert, preserve scoped Sync deferral, and retain pending evidence until all recipient subscribers acknowledge local writes. Compact successful intents while preserving dispatch identity; keep webhook delivery independent.
Validated with six MongoDB cases, three Node suites, three full-app Meteor cases and ten Chromium email delivery and private-storage scenarios. Automatic pending-intent recovery remains unfinished.
Thanks to xet7.
Confirm immutable intent persistence before inserting an activity, reconcile uncertain acknowledgements, and require exact activity snapshots for recovery. Bind insertion authority to the creating call so orphan retries cannot resurrect deleted activities.
Four real MongoDB fault and concurrency tests pass. This internal primitive is not activated in activity hooks; scheduling, persisted delivery plans and operator lifecycle remain open.
Thanks to xet7.
Retain detailed metadata for 30 days by default, then replace terminal jobs and commands atomically with minimal permanent identities. Preserve unfinished work, holds, generation fencing and actor-bound command replays. Sweep bounded batches independently and reconcile uncertain writes.
Validated with six Node suites, real MongoDB faults and replay tests, the Meteor startup scheduler, and Chromium Recovery replay and authorization checks.
Thanks to xet7.
Map selected aggregate seconds to the board's imported numeric time fields in hours. Preserve zero, missing and explicit null; validate mapping identities and retain unrelated fields through conditional array writes.
Twenty-five focused Node suites and twenty-two Chromium scenarios pass with local MongoDB and Jira fixtures. Live Jira, FerretDB and other browsers remain unverified.
Thanks to xet7 !
Apply configurable finite DNS, connection, greeting and idle limits to standard SMTP URLs, TLS overrides and admin providers. Keep Meteor's native standard transport and reapply limits after Sandstorm mail updates.
Eleven focused Node suites and fifteen Chromium scenarios pass. Real SMTP tests prove greeting and idle timeouts close connections, preserve queued payloads and allow later delivery. Absolute delivery deadlines remain pending.
Thanks to xet7 !
Select due recipients rather than a message-limited prefix. Share overlapping scans and process four recipients concurrently while preserving recipient leases. Verify pool limits, query recovery, large-backlog fairness and actual slow SMTP delivery.
Nine focused Node suites and fourteen Chromium scenarios pass; the strengthened scheduling-order browser case also passes. Startup-only harness was not rerun.
Thanks to xet7 !
Persist attempt reservations before sending, classify permanent SMTP failures, add bounded jitter and retain failed messages for administrator retry or cancellation.
Verify fourteen focused Node suites and fourteen Chromium scenarios with local MongoDB and SMTP. The separate stopped-app startup harness was not rerun.
Thanks to xet7 !
Index SAML, LDAP, OAuth login providers and Passwordless login separately from general Login settings. Update the implementation inventory to match the current routes.
Thanks to xet7 !
Summarize recipients without mail payloads, paginate with bounded memory, and coordinate pause/resume/cancel with sender reservations. Persist holds, cancellation cutoffs and idempotent request receipts so replay cannot undo newer actions or cancel later messages.
Verify current administrator access, interrupted commands, lost write acknowledgements, stable cancellation and private collection boundaries with MongoDB and Chromium tests. Keep translations limited to English source labels.
Thanks to xet7 !
Replace process-only digest timers with a private event/recipient outbox, renewable recipient reservations, persisted retry backoff and verified SMTP acceptance. Migrate legacy buffers without discarding unqueued text, retain small replay receipts, and recheck membership before delivery.
Verify MongoDB concurrency and interrupted writes, actual application startup recovery, and ten Chromium scenarios using local SMTP. Document at-least-once delivery and remaining recovery controls.
Thanks to xet7 !
Await buffer persistence, serialize per-user digest timers, and remove only delivered lines. Retain failed sends and newer reply targets. Cover SMTP rejection recovery through Chromium and local SMTP capture, plus concurrent sends and database read failures.
Thanks to xet7 !
Require caller-persisted intent IDs and private immutable completion records before cleanup. Resume known completions without rebuilding or reapplying cards, reject conflicting scopes and pending intents, and preserve newer operations.
Thanks to xet7.
Read back plan and marker deletion, retain incomplete cleanup, resolve lost delete acknowledgements only from confirmed absence, and protect successor operations and uncertain outcomes.
Thanks to xet7.
Compare before/after field unions without replacing unrelated metadata. Share literal snapshot predicates with production Sync, isolate result verification from driver mutation, and reject array shapes that cannot survive BSON storage.
Thanks to xet7.
Validate typed custom-field arrays, canonical estimate mapping identities and source baselines. Permit only the mapped field to change, retain local fields exactly, and verify persisted update/clear/create recovery and conflict retention against MongoDB.
Thanks to xet7.
Parse complete finite decimal tokens instead of integer prefixes. Keep text and dropdown matching, reject malformed numeric ranges, and verify database and browser behavior.
Thanks to xet7.
Use recursive grouping and Mongo-compatible whole-selector negation. Preserve left-to-right logical ordering, reject unfinished expressions before queries, and cover MongoDB and Chromium behavior.
Thanks to xet7.
Use elemMatch for field identity and value together in sidebar and rule queries. Reject unknown field names before querying, and verify cross-field false positives with MongoDB and Chromium regressions.
Thanks to xet7.
Compare saved custom-field values by identity for array, dotted and assignment updates. Suppress no-op and failed writes, preserve zero and false, and evaluate advanced-filter rules using boardIds-scoped definitions.
Thanks to xet7.
Select an existing numeric Jira estimate field in the Sync popup, retain field/unit identity, request the selected Cloud field, and merge numeric estimates with source baselines. Preserve zero, explicit null clearing, unrelated custom fields and their order; reject invalid or changed mappings and stale card writes. Cover source selection, conflicts, remapping, invalid values and full popup flow in Node and Chromium.
Thanks to xet7 !
Validate only present target entries while rejecting unknown keys, invalid values and broken placeholders. Keep source validation strict, log omitted key counts, and apply the same rules in offline dry runs. Preserve existing locale bytes and human-owned upload exclusions.
Thanks to xet7 !
Reject oversized rollover, compound History and recovery payloads before the first lifecycle write using BSON byte counts. Cover multibyte data, complete 10000-card plans and large-metadata rejection/retry in Chromium.
Thanks to xet7 !
Fetch projected inputs with an overflow sentinel before start or close, reject oversized card/list sets without partial snapshots, and index done-list membership once. Cover exact limits and unchanged lifecycle/History on rejection with unit and Chromium tests.
Thanks to xet7 !
Check source integrity, conditional write results and persisted undo state. Preserve undo timestamps across retries and delete only the matching operation checkpoint. Test failed writes, damaged final state, replacement checkpoints and recovery in Node and Chromium.
Thanks to xet7 !
Use deterministic event IDs and verify immutable contents and integrity before clearing recovery evidence. Accept exact legacy rows and lost insert acknowledgements; retain checkpoints on failed or conflicting records. Cover replay, concurrency, failed writes and damaged recovery rows in Node and Chromium.
Thanks to xet7 !
Reuse the actual merge and write plan, scope full-list previews to unrestricted writers, bound reports and keep omitted values private. Exclude status-only diffs from update counts. Preserve empty and whitespace source text through schema cleaning so repeated runs do not keep changing comparison baselines. Cover preview parity, saved settings, access changes, no-write errors, storage behavior and browser flows.
Thanks to xet7 !
Fence cleanup with opaque revisions and delete only bounded immutable ID snapshots. Settings saves repair invalid or exhausted counters atomically, while active tokens and newer staging remain intact. Cover reset races, overlapping delayed sweeps, bounded backlogs, lost acknowledgements and popup saves.
Thanks to xet7 !
Generate a fresh credential incarnation for each list insertion, including imports and copies. Match it when selecting credentials and activating settings. Sweep orphan versions by exact identity without reading secrets or deleting a recreated list's new token. Cover recreation races, delayed saves, retryable cleanup and browser flows.
Thanks to xet7 !
Persist conditional replacement decisions before card insertion. Revalidate source previews, restrict replacement to full-list writers and preserve the previous card. Cover stale decisions, database races and lost acknowledgements, moved and detached UI flows, and private collection writes.
Thanks to xet7 !
Offer a scoped archive-conflict preview and a parent-only mapping removal. Refetch source absence and revalidate the parent and archive blocker before conditional resolution. Preserve all subtask content and hide subtask identifiers from previews; assigned-only callers retain their write scope.
Ten Sync Node suites and thirteen Chromium scenarios pass, including both permission scopes, returning source items, unchanged subtasks and existing Sync behavior. Local release audit passes with advisory warnings. Add English source labels only. Creation collisions after moved or detached cards remain pending.
Thanks to xet7 !
Show extra mappings alongside the consistently retained card. Fingerprint the visible group, reject changed previews, and remove only an extra card’s Sync identity and baseline. Protect the same first mapping independently of database cursor order; preserve card content and assignment checks.
Ten Sync Node suites and eleven Chromium scenarios pass, including stale groups, retained-mapping protection, content preservation and successful reconciliation after repair. The local release audit passes with advisory warnings. Add English source labels only.
Thanks to xet7 !
Assigned-only writers review and resolve their own existing cards without triggering whole-list reconciliation or changing shared Sync status. Keep assignment in the conditional resolution write and recheck access after fetching. New shared conflict summaries omit card identifiers and values.
Ten Sync Node suites and ten Chromium scenarios pass, including hidden-card preservation, assignment revocation, scoped writes, source switching and credential recovery. The local release audit passes with advisory warnings. English source label only; no translation fill.
Thanks to xet7 !
Show local and source values for title, description and spent-time conflicts. Refetch and fingerprint the comparison before accepting a field choice, preserve unrelated fields and conditionally reject changed cards. Keeping local values advances their source baseline; no upstream writes occur.
Require unrestricted board write access and recheck it after fetching. Ten Sync Node suites, field-specific regression checks, catalog ordering and nine Chromium scenarios pass. Duplicate identity, archive/move conflicts, assigned-only resolution and transactional reconciliation remain pending. Add English source labels only.
Thanks to xet7 !
An hourly streaming sweep advances a server-owned generation before deleting unselected older versions. Settings activation checks its captured generation, so a delayed save cannot select a retired credential. Preserve active and newly staged versions, isolate per-list failures and keep secrets out of cleanup logs.
Eleven Node suites pass, including real MongoDB concurrent saves and sweeps, late inserts, uncertain writes and restart retries. Eight Chromium Sync scenarios pass across the initial run and focused corrected-test rerun. Upgrade all writers together; missing-list and malformed-generation records remain for review.
Thanks to xet7 !
Normal imports enter the recoverable cleaning phase before deleting private plan rows. Propagate cleanup failures and keep the destination guarded until deletion is acknowledged, so offline recovery can finish partial cleanup.
Three Node suites and two Chromium native-import/board-copy scenarios pass. MongoDB tests inject interruptions at every cleanup boundary and resume through the maintenance implementation.
Thanks to xet7 !
Validate complete plans before undoing stored writes in reverse order. Restore only original Scrum metadata and conditionally remove exact imported planning records. Persist reverse progress and cleanup, retain recovery claims on failure, and safely discard incomplete preparation without destination writes.
MongoDB integration tests cover every planned collection, write gaps, changed targets, cleanup interruption, corrupted cursors and the CLI. FerretDB remains unverified.
Thanks to xet7 !
Preserve localized help URLs during upload validation and pull merging. Repair translated string-template tokens in 20 locales and retain exact-value corrections. Cover all catalogs, genuine damaged tokens and the actual pull merge.
Thanks to xet7 !
Validate every target before writing, preserve stable IDs and conditional updates, and retain non-expiring recovery claims on failure. Resume complete staging and interrupted cleanup; reject partial plans and changed targets. Online recovery and other board-import stages remain unfinished.
Stage private per-document recovery steps before destination writes and retain board-level pending state through marker cleanup. Preserve stable IDs and exact before/after values for interrupted writes. Public coordinated replay and rollback remain unfinished.
Reject Scrum mutations and History changes until imported data is complete. Remove editing capabilities, show the pending warning and return a clear 409 for report downloads. Verify exclusion and normal access after the importer clears its marker.
Thanks to xet7 !
Introduce version 2 while accepting version 1 imports. Remap snapshot references, retain measured epochs and partial-data notices, bound transfer sizes, and prevent collectors from observing unfinished imports.
Thanks to xet7 !
Share the bounded history reader with report downloads. Preserve exact UTC capture times, estimate policies, unknown counts and partial or truncated result notices. Verify real workbooks and extracted PDF text through the application.
Thanks to xet7 !
Render timestamped scope, remaining and completed observations with count and estimate metrics. Preserve gaps and unknown estimates, label partial and truncated results, and ignore stale requests after sprint changes.
Thanks to xet7 !
Keep the first actual observation of each UTC day, preserve unknown estimates, and bound history reads. Verify retry and concurrent capture behavior with MongoDB and reader permissions with Chromium.
Thanks to xet7 !
Stage a private credential version and select it with settings using a conditional revision update. Preserve the previous or new pair across interrupted saves and lost acknowledgements. Prevent stale saves and delayed cleanup from replacing newer credentials, including after a clear.
Thanks to xet7.
Share a private database reservation across scheduled Sync, manual Sync and configuration methods. Renew during fetches, verify ownership before writes, reclaim expired reservations and preserve replacement owners during cleanup. Document the remaining in-flight write fencing and multi-document recovery requirements.
Thanks to xet7.
The maintainer is evaluating Transifex translation features to obtain most translations there. Keep earlier translation checkpoints as reference material only.
Thanks to xet7.
Use a stable database ID for new list/source/issue mappings. Stop on creation conflicts without replacing moved cards or claiming successful reconciliation. Cover concurrent database writers, browser clients, moved-card recovery and unrelated database failures.
Thanks to xet7.
Preserve previous-source cards across project switches, require matching credentials before fetching, and bind identifiable legacy mappings before replacing or clearing settings. Reject unknown legacy sources and stale fetched configurations.
Verify with nine focused Node suites, nine Chromium scenarios and the local release audit. Concurrent jobs and configuration writes remain nontransactional.
Thanks to xet7 !
Limit the exceptions to three exact URLs in package-lock.json. Cover accepted funding metadata and rejected URL variants, application URLs and suspicious keywords.
Thanks to xet7 !
Record current Scrum, Rules/History, translation, issue/UI/security audit and verification status under TODO Later. Refresh translation counts, stop active-work wording and wrap Upcoming entries. Completed implementation remains in its feature/fix-specific commits.
Thanks to xet7 !
Fill 60 English placeholders for navigation, screen-reader hints, workspace summaries and search. Preserve tokens and existing translations. Extend vocabulary and direction checks; accessibility terminology remains pending native-speaker review.
Thanks to xet7 !
Fill 38 English placeholders while preserving existing translations and tokens. Extend checks for variable operations and functions with and without return values. Specialist terminology remains pending native-speaker review.
Thanks to xet7 !
Fill 86 English placeholders for arithmetic, statistics, rounding and trigonometry. Keep mathematical notation and placeholders intact and extend catalog regressions. Specialist Gujarati mathematical terminology has low confidence and needs native review.
Thanks to xet7 !
Fill 75 English placeholders while retaining tokens and existing translations. Extend regression coverage for list actions and ascending/descending order. Specialist terminology remains pending native-speaker review.
Thanks to xet7 !
Fill 66 English placeholders while preserving existing translations and tokens. Expand Gujarati vocabulary and placeholder checks. Specialist mathematical and accessibility terminology remains pending native-speaker review.
Thanks to xet7 !
Fill 80 more English placeholders, including short control-flow aliases and bitmap states. Extend catalog checks and verify Gujarati dragging, field editing, save and context menus in Chromium. Specialist terminology remains pending native-speaker review.
Thanks to xet7 !
Fill 87 English placeholders for navigation, accessibility, colors and basic blocks. Preserve placeholders and existing translations, and update the active translation status. Gujarati terminology remains pending native-speaker review.
Thanks to xet7 !
Add field ID and unit controls to Jira import, validate before writes, and preserve mapping through Jira/native exports and board copies. Keep new fields hidden and reuse existing labels and permissions.
Thanks to xet7 !
Reject contradictory sprint states and snapshot timestamps before native import writes. Preserve valid active and cancelled history and cover import and board-copy round trips.
Thanks to xet7 !
Prevent inherited object properties from becoming list IDs or dependency targets. Reproduce the old import failure and verify literal names, Scrum categories, rendered cards and valid dependency links.
Thanks to xet7 !
Map explicit Jira category keys to existing hidden list metadata, validate before import writes, and honor Scrum selection in both directions. Verify Jira and native round trips and unchanged time tracking.
Thanks to xet7 !
Use stable card IDs to break equal effective-rank ties across fetches and clients. Preserve explicit zero ranks and normal list-position fallback. Cover metadata edits and refreshes in Chromium.
Thanks to xet7 !
Document verified History integration, shared-component behavior, localized editing and responsive layouts. Keep translation status explicitly paused and retain concurrency and backend limits.
Thanks to xet7 !
Clear stale manual-button metadata for automatic triggers and populate it for REST button rules. Verify actual menu visibility and History undo/redo for editor and REST changes.
Thanks to xet7 !
Copy changed shared triggers and actions while preserving unshared IDs and REST patch semantics. History restores component references and cleans unreferenced private copies.
Thanks to xet7 !
Reuse one cleanup path for editor, REST and History deletion. Restore shared components only when their recorded content is unchanged, without writing over another rule.
Thanks to xet7 !
Use the authenticated API user and shared rule snapshot wrapper for creation, edits and deletion. Verify one entry per operation, no-op suppression, permission denial and undo/redo.
Thanks to xet7 !
Reject mixed units, estimate sources, custom fields and completion policies before importing a board instead of combining unlike historical totals.
Thanks to xet7 !
Allow clearing rank metadata, retain zero as an explicit rank and avoid negative board positions blocking unrelated Scrum edits. Verify History and native transfer preserve cleared values.
Thanks to xet7 !
Display the current release and provide a board-local selector using existing card validation and History undo/redo. Keep card options independent from the planning-page release selector.
Thanks to xet7 !
Snapshot the configured workweek and use the start calendar for inclusive planned date counts. Preserve legacy unknowns, validate transferred calendars and include counts in reports and exports.
Thanks to xet7 !
Reuse active board membership validation, administrator-only configuration and History undo/redo. Add source labels with English catalog fallbacks and verify settings do not grant permissions.
Thanks to xet7 !
Carry metadata into newly created destination lists during list and swimlane moves, while retaining existing destination categories and revisions.
Thanks to xet7 !
Preserve descriptive metadata, reset copied revisions and remove foreign planning references and rank from cross-board card and subtask copies. Full-board copies keep the existing transfer remapper.
Thanks to xet7 !
Reuse numeric Jira time conversion and preserve local time edits with source baselines and conditional writes. Cover settings, zero values, excluded fields and invalid time.
Thanks to xet7 !
Add Title and Description switches with existing behavior as the default. Validate selections through the existing write-authorized method and exclude unselected fields from merge and baseline updates. Verify persistence, unsupported values and new-card defaults.
Thanks to xet7 !
Archive only planned source-matched cards with conditional text and location checks. Reject runs with active subtasks outside the archive plan rather than cascading through locally created work.
Thanks to xet7 !
Strip source IDs and text baselines from card/subtask copies. Reject existing duplicate local mappings before reconciliation rather than selecting an arbitrary card. Verify source preservation, actual copy bodies and board duplication.
Thanks to xet7 !
Track accepted source text, merge upstream-only edits, retain local-only edits and stop conflicting runs before card writes. Guard updates against concurrent text or location changes and display structured failures correctly in Sync now.
Thanks to xet7 !
Validate normalized tasks before reading or mutating cards. Missing IDs, duplicate IDs and malformed text cannot silently disappear from the source set or overwrite another task. Retain ordinary empty-source and unchanged-card behavior.
Thanks to xet7 !
Route standard Atlassian Cloud tenants to v3 search with explicit parser fields and complete token pagination. Reject missing or repeating tokens while retaining offset pagination for self-hosted Jira.
Thanks to xet7 !
Follow Jira offsets and provider next-page headers, validate advertised totals, and abort incomplete, looping or failed fetches instead of returning partial issue sets. Bound page/item counts and keep pagination on the configured origin without redirects.
Thanks to xet7 !
Reuse import source validation and handle parser failures through the existing error path. Invalid responses cannot masquerade as an empty source and archive cards. Verify actual sync function behavior and error display.
Thanks to xet7 !
Remove nested and flat fallback estimates when Custom Fields is excluded and spent time when Dates is excluded. Verify both actual import-page selections and preserve the existing full-transfer round-trip.
Thanks to xet7 !
Exercise native export/import and whole-board duplication with new field IDs, renamed estimate markers, remapped Scrum settings and subsequent Jira exports. Document the verified preservation boundaries.
Thanks to xet7 !
Keep valid numeric integration settings when the editor rebuilds visible settings. Exercise the production save handler and rename an imported field through its popup before exporting and re-importing Jira time data.
Thanks to xet7 !
Mark imported numeric estimate fields independently of their names, convert hours back to seconds, and honor Dates and Custom Fields export selections. Verify renaming, omissions, invalid and ambiguous values, and actual Jira re-import.
Thanks to xet7 !
Import numeric seconds as spent hours and hidden numeric original/remaining estimate fields. Preserve zero values, reject invalid durations before creating the board, and allow the estimate field to be selected for Scrum. Verify native export retention and existing Jira import compatibility.
Thanks to xet7 !
Index completed card IDs once instead of scanning the closing snapshot for every committed card. Verify 10,000-card snapshots with a deterministic work bound and unchanged commitment, completion and scope totals.
Thanks to xet7 !
Render accessible responsive count and estimate bars from existing filtered snapshots. Separate incompatible units, estimate sources, custom fields and completion policies, retain unknown counts and partial warnings, and preserve Excel/PDF exports.
Thanks to xet7 !
Resume using the stored revision and rollover destination, and reject retries while a History recovery checkpoint is pending. Add browser coverage for successful recovery, read-only denial and pending History exclusion.
Thanks to xet7 !
Add Scrum to board copy choices and reuse transfer remapping for planning records, metadata and snapshots. Collect copied list and swimlane IDs, omit deselected Scrum data and label reduced snapshots in reports and exports.
Thanks to xet7 !
Validate before import side effects, allocate destination planning IDs and remap cards, lists, fields, users and snapshots. Preserve source provenance and expose bounded loss reports to board administrators without granting permissions.
Thanks to xet7 !
Add a selectable Scrum section, scope planning records and snapshots, report omitted references, preserve anonymization and exclude operational recovery state. Native import and History transfer remain documented follow-up work.
Thanks to xet7 !
Share authorization across all nine exporters and retain access only for already scoped Scrum reports. Return HTTP 403 for denied document exports. Document ExportScopeBleed and cover policy, source-wide guards and HTTP behavior.
Thanks to xet7 !
Rewrite known username mentions in Scrum settings, metadata and transfer records. Run the same pass for streaming board and swimlane exports without changing references or estimates.
Thanks to xet7 !
Prepare shared native transfer, duplication and adapter validation with explicit partial-data losses. Reject invalid snapshots, foreign planning references and ID collisions. Runtime integration remains tracked in the Scrum design.
Thanks to xet7 !
Expose existing release and event update methods in Sprints, including release dates and status, event timestamps and follow-up cards. Preserve unchanged timestamps and use existing admin permissions and History.
Thanks to xet7 !
Record compound sprint transitions in existing board History, preserve unrelated card data and enforce current per-item permissions. Keep recovery checkpoints until timeline and undo status are finalized.
Thanks to xet7 !
Filter historical rows before search, counts and contributor aggregation using current board visibility and assigned-card scope. Gate all reversal paths with the same scope and current card/linked-card edit permission so revoked access or an old board cannot authorize hidden-card writes.
Verified 10 focused Node runner entries and eight Chromium scenarios across History and rule undo/redo, including allowed restores and negative cases. Document detection limits to avoid treating legitimate stale permissions as attributable attacks.
Thanks to xet7 !
Reuse board permissions, card estimates and chart exports for Product Backlog, Sprints, Sprint Report and Velocity views. Add revision-checked sprint lifecycle, recoverable rollover, release/event creation and independently hidden metadata controls. Explicitly label assigned-only reports as partial.
Verified 12 focused Node checks, a real Meteor/Mongo lifecycle regression and three Chromium scenarios including settings visibility, denial paths and actual Excel/PDF files. History integration, full transfer mappings and further report/editor work remain tracked in the Scrum design.
Thanks to xet7 !
Keep remaining language coverage and terminology review explicit, with reproducible non-completion counts. Continue other development while translations are paused.
Thanks to xet7 !
Translate remaining Blockly and Blocks editor prose in Hindi and hi-IN, Tamil, Kannada, Bengali, Nepali and Urdu. Preserve placeholders and correct-language translations; repair incorrectly seeded Urdu shared labels. Specialist accessibility and mathematical terminology merits native review.
All three catalog validation suites pass. Pause translation work here at the maintainer's request; Gujarati, Marathi, Malayalam, Punjabi, Sinhala and Telugu remain pending.
Thanks to xet7 !
Translate remaining Blockly prose and editor controls, preserve placeholders and correct-language existing strings, and replace wrongly seeded shared labels. Ladin dialect and Aromanian technical neologisms are low-confidence; Friulian specialist terms also merit native review.
All catalog order and token checks pass. No external translation service was used.
Thanks to xet7 !
Map sprint planning, reports, hidden metadata and data transfer gaps to existing board menus, permissions and features. Clearly distinguish the design from shipped functionality.
Thanks to xet7 !
Lazy-load Blockly for board administrators, preserve existing rule documents and execution, and provide direct List, Workflow, Blocks and History choices. Route workflow writes through server methods. Include local media, complete message-key coverage, translation merging and placeholder checks.
Translation coverage remains incomplete in additional locales. Low-confidence specialist terminology needs native review in Aragonese, Asturian, Corsican, Neapolitan, Occitan, Sardinian, Sicilian, Kashubian, Upper Sorbian, Silesian, Cornish, Manx, Gaelic and Breton; Faroese and Frisian also merit review.
Verified five Blockly/catalog Node tests, existing rules suites and four Blockly Chromium scenarios, including admin permissions, localization, drag editing and persisted execution.
Thanks to xet7 !
Wrap long text and buttons, remove overlapping absolute positioning, and keep form controls readable on mobile, RTL and dark themes. Four Chromium scenarios cover desktop and mobile layout boundaries.
Thanks to xet7 !
Snapshot rules with their trigger and action, combine compound edits, and restore through existing History with board-admin checks and stale-state protection. Cover lifecycle undo/redo, member rejection and conflicting edits in Chromium.
Thanks to xet7 !
Keep replacement trigger and action documents free of MongoDB update operators while preserving creation timestamps. Cover replacement and modifier updates independently.
Thanks to xet7 !
Credit kta1kri for the LDAP report and record the sibling audit, positive and negative tests, local protocol fixtures and Chromium coverage. Do not classify replay-cache capacity failures as attempted replay.
Thanks to kta1kri and xet7 !
Audit authentication boundaries following GHSA-m87f-f43w-hwmc. Guard both LDAP user-bind paths and the DDP/REST entry point, preserve service-account group lookup, compare CAS groups literally, and require single-use SAML request correlation. Add positive, negative, protocol and browser regressions with Problems summaries.
Thanks to kta1kri and xet7 !
Honor the commented Codeberg registry entry even with stale saved settings. Create missing WeKan source and destination clones under the checkout-relative .tools directory, keeping fetched mirror refs out of the primary checkout. Cover fresh clones, paths with spaces, saved settings, failure handling and existing checkout protections with injected commands.
Thanks to xet7 !
Replace the separate without-cards action with one selection popup, defaulting to all parts and providing Select All, Select None and Cancel. Validate choices on the server and preserve parent requirements, card relationships, custom fields and existing board-admin permissions.
Repair attachment copying to use explicit imports and the installed Meteor-Files promise API, wait for flushed file bytes, and propagate copy errors. Verify selection combinations, source preservation, attachments, multiselection, cancellation and denied calls with Node and Chromium tests.
Thanks to xet7 !
Reuse the existing visible comment data and escaped title attribute. Keep the localized count for multiple or empty comments. Cover count fallback and plain-text handling with Node tests, and verify admin/read-only rendering and private-board comment isolation in Chromium.
Thanks to lonix1 and xet7 !
Fixes#1933,
Existing Boards.copy, Swimlanes.copy and Cards.copy do not copy Activities records. Cover source board/card history exclusion, preservation, confirmation cancellation and permission rejection through the real multiselection duplicate actions, with and without cards. Document the existing behavior and refresh the open-issue audit.
Thanks to sfahrenholz and xet7 !
Fixes#2321,
Reuse the existing parent relationship and filter engine from card actions and the filter sidebar. Keep parent choices board-local, allow read-only filtering, and clear the selection when changing boards. Verify filtering, publication boundaries and unchanged card data with Node and Chromium tests.
Thanks to robinvd and xet7 !
Fixes#1871,
Place Draggable before Minicard and Card in a three-column layout that stacks when the popup is narrow. Expand long option menus and pickers into balanced columns while preserving form layouts, search focus, popup stack navigation and permission-filtered options.
Thanks to xet7.
Honor Submit editors with Enter across textarea editors, retain composition and mention handling, and route shortcuts through existing scoped save actions. Preserve manual resize dimensions after typing and retain single-line structured fields.
Keep the typography reset from overriding themed button foregrounds, including Board Settings column actions. Verify blue, green, dark and Clean Light buttons and chart actions.
Thanks to xet7.
Show Flag Language (Flag Country) on one line, retaining directional isolation and the RTL marker beside the language name. Allow long rows to scroll on narrow screens instead of wrapping or hiding their text. Verify all language rows at desktop and phone widths in LTR and RTL.
Thanks to xet7 !
Release ancestry places ZIP previews (2016d15f2) in v12.01, absent from v12.00, and multi-list swimlane drops with named previews (f772663e5) in v12.03, absent from v12.02. Move both unchanged entries out of Status into those sections.
Extend the changelog guard to reject completed release summaries anywhere under Status, including before TODO Later. Verify the regression fails before relocation and all four changelog suites pass afterward.
Thanks to xet7 !
Remove the board-tile Actions hamburger and its popup. Add Duplicate Board — Without cards immediately below Duplicate Board in the All Boards multiselection sidebar, using the shared confirmation, selected-board guard and existing copyBoard permissions.
Verify both copy modes for two selected boards, cancellation, preserved source cards and denied non-admin duplication and archiving. Update the All Boards documentation.
Thanks to xet7 !
Mark the Monte Carlo and size versus cycle time option submit buttons as primary so existing named and custom theme rules supply their background and readable text colors. Verify both pages in default blue, red, dark and custom themes, including hover and keyboard focus.
Thanks to xet7 !
Keep desktop card windows unpainted until initial anchoring completes. Prefer the right side in LTR and the left side in RTL whenever it fits; otherwise use the other side and keep the window inside the viewport.
Verify first-visible-frame geometry for preferred and fallback sides in both directions, and retain passing date-save position regressions.
Thanks to xet7 !
Align each radio button to the left of its label. Verify rendered control geometry, insertion above and below, and read-only restrictions in Chromium.
Thanks to xet7 !
Date saves and removals refresh the board publication. Preserve the mounted view while that same board refreshes, instead of rebuilding its windows at the default dock and anchoring them again. Reset readiness on board navigation as before.
Verify save/delete for Received, Start, Due and End in Chromium, including persisted dates, badge updates and frame-by-frame window position. Cover initial loading, same-board refresh and navigation in a focused readiness test.
Thanks to xet7 !
Record an individual finding and remaining verification for every open issue. Browser tests confirm swimlane names on minicards and board-specific view visibility already work; retain the remaining 133 requests with their outstanding scope.
Thanks to sebastianha and xet7 !
Fixes#1748, fixes#2107,
Reuse owner-scoped unsaved edits, restore titles only before local typing, and clear saved titles after successful insertion. Verify restoration, list isolation, successful saves and denied cross-user writes.
Thanks to TutloTutlo and xet7 !
Fixes#810,
Offer label combination mode under Filter, retaining exclusions and no-label alternatives. Restore active title, list and advanced text when the panel is recreated. Verify actual calendar grid/list results, unmatched and cleared filters, and label combinations in Chromium; pure tests cover reset and invalid modes.
Thanks to Trunkslike, TheBigBloodyB, Roemer and xet7 !
Fixes#2044,
Fixes#3361,
Use the existing Add Swimlane popup for above/below insertion. Restore reachable board-admin archive and duplicate controls in the All Boards tile Actions menu. Preserve existing server permission guards and confirmation behavior. Unit and Chromium tests cover ordering, successful operations, cancellation and rejected writes.
Thanks to bentiss and xet7 !
Fixes#2131,
Fixes#2644,
Record the 149-open-issue follow-up audit with implementation evidence,
original regression commits where available, and explicit remaining scope.
OIDC logout, case-insensitive mentions, remote mobile card closure,
alphabetical member selection, bulk card colors, existing-card subtasks
and minicard field visibility already work. Extend executable logout-hook
coverage and add browser checks rather than relying on method names alone.
All 31 focused Node suites and 27 distinct Chromium scenarios pass across
this audit and its two separately committed fixes. No live external IdP
or FerretDB backend matrix was tested.
Thanks to RowhamD, Usernameisalreadytaken99, rptl, czinkos, bhueck,
Vida444, xefladrero and xet7 !
Fixes#5061,
Fixes#1938,
Fixes#3114,
Fixes#1686,
Fixes#2160,
Fixes#3213,
Fixes#3198,
History restored a list but left deletion markers on its cards. Replay the
existing soft-delete/restore modifiers across the same board/list and batch.
Keep independently deleted cards untouched and refuse lifecycle rows for
lists moved to another board. Preserve the deletion batch in lifecycle
content and record the actual lifecycle state in history snapshots.
The browser regression failed against the previous implementation. It now
passes undo, redo and undo again with persisted card checks. Separate
read-only rejection, pure negative cases, existing history tests and
flow/time-history browser regressions pass.
Thanks to rawwerks and xet7 !
Fixes#1023,
The card-link picker excluded every real card on its own board, and the
server rejected same-board sources. Include the current board and exclude
only sources already mirrored. Preserve coordinate and write validation,
reject link chains, and prevent whole-board self-links through both buttons.
Labels already use the source card. Verify live source-to-mirror updates
and edits through the mirror, same-board authorization failures, whole-board
self-link rejection, and existing cross-board behavior in Chromium.
Focused linked-card Node regression suites pass.
Thanks to RinTheCatato, mohamed-aziz-chamakh and xet7 !
Fixes#5683,
Audit the 159-open-issue snapshot against existing source and regression
coverage. Record the implementations and their original regression commits
in docs/DeveloperDocs/Open-Issue-Completion-Audit-2026-09.md.
Visibility descriptions, field ordering, archived subtasks, linked covers,
checklist-to-subtask conversion, drafts, subtask badges, archive details,
rule toggles/editing and standalone SAML already have implementations.
All 28 selected Node regression suites pass; live browser/IdP tests were
not rerun for this documentation-only audit.
Remove completed TODO entries, correct two unrelated issue references,
and retain incomplete feature requests and deployment investigations.
Thanks to Meeques, Somantiq, javen9881, matrixes, bastos77,
therampagerado, sfahrenholz, Gobliins and xet7 !
Fixes#4421,
Fixes#3877,
Fixes#3409,
Fixes#2498,
Fixes#2422,
Fixes#3243,
Fixes#3207,
Fixes#1504,
Fixes#2322,
Fixes#708,
Describe the new reports, reused data, retained history, issue audit and validation limits. Keep released sections unchanged.
Thanks to sojournerc, ladistrupl, zombah and xet7 !
Keep card and checklist activity after deletion, retain card snapshots for
flow reports, and show recorded hour adjustments by their actual author in
Time and its PDF/Excel exports. Record whole positions centrally, including
optional move reasons and direct REST edits. Append undo/redo checkpoints
so blocker intervals and time corrections follow reversals through time.
Audit the open time-related requests against existing features. Exact inline
activity timestamps requested in #1673 already exist in dateDisplay and the
activities template; add a browser regression for that existing behavior.
Reuse established locale terminology for the new compact labels. Labels
outside English and Finnish are low-confidence glossary combinations for
human review; existing translations remain untouched.
Validation: 43 focused Node suites pass. Fifteen Chromium scenarios pass
across flow analytics, time history and existing REST move coverage, including
real chart and Time PDF/Excel exports, undo/redo, retention and access denial.
Thanks to ladistrupl, zombah and xet7 !
Fixes#1598,
Fixes#1673,
Reuse board dates, activities, estimates, dependencies and universal History for Aging WIP, Blocker Analysis, Monte Carlo forecasts, XmR and size versus cycle time. Keep Chart.js lazy loaded and share parameterized PDF/Excel exports. Record dependency edits in History with undo/redo and board-boundary validation.
Validation: 21 focused Node suites and 8 Chromium tests pass, including every chart download and dependency undo/redo. Report labels reuse native locale vocabulary and formula notation; compact explanatory wording outside English/Finnish has low confidence and welcomes review.
Thanks to sojournerc and xet7 !
Fixes#1476,
Fix InactiveBleed across REST, attachment and export authentication, header login, LDAP token issuance and People account controls. Apply administrator-created account flags before insertion and guard token creation atomically.
Thanks to esbrito81 and xet7 !
Recognize empty swimlane bodies as mixed-selection drop destinations while keeping swimlane drag starts on headers. Replace the count-only indicator with a tilted stack of selected object names and icons rendered as text. Verify two-list moves, card parents, ordering, collapsed containers, both handle modes and persistence after reload.
Thanks to xet7 !
Reuse one control for draggable, card visibility and swimlane/list settings columns. Apply explicit booleans atomically through board-admin authorization, deriving column fields from existing row definitions and retaining personal overrides. Cover every column, repeated clicks, persistence, isolation and denied updates.
Thanks to xet7 !
Default swimlanes, lists, cards, checklists, checklist items and subtasks to draggable. Share settings UI and policy, update ordinary sortables reactively and enforce the policy for mixed-selection drags while retaining explicit menu moves. Cover persistence, authorization, independent defaults and both checklist handle modes.
Thanks to xet7 !
Select lists, swimlanes, cards and checklist content together. Preserve parent hierarchies, validate move permissions, and create a destination card for checklist content dropped onto a list.
Thanks to xet7 !
Use the same progress display and sortable lifecycle on minicards and opened cards. Restore sorting after unfolding, honor both drag-handle modes, and distinguish a checklist drop from board-list card conversion. Verify persisted reordering and transfers across lists and swimlanes.
Thanks to xet7 !
Read destination rows in document order instead of relying on direct siblings, which can be inline-form wrappers. Exclude the drag helper and placeholder. Reproduce moving 45 from last to first and then into the third placeholder, with handles enabled and disabled, and verify persisted order after reload.
Thanks to AmigaAbattoir and xet7 !
Fixes#6723.
Capture the destination checklist and neighboring items before cancelling the sortable DOM move, so Blaze persists the intended order rather than the original position. Cover boundary and cross-checklist moves, card conversion, and actual dragging with reload persistence.
Thanks to AmigaAbattoir and xet7 !
Fixes#6723.
Document SAML administration and its verification limits, restore the People navigation entry, and order the release entries by feature, interface, fixes and release tooling.
Thanks to xet7 !
Provide shared catalog-driven settings controls with environment fallback, live admin overrides, platform configuration and regression coverage. Document the remaining live identity-provider validation requirement.
Thanks to xet7 !
Give SAML and standalone provider actions the same blue background and white text and icons as the password Login button. Verify normal and hover colors in Chromium.
Thanks to xet7 !
Keep the redirect guard after failed login attempts, require writable session storage, and recheck login state after the asynchronous settings request. Cover callback reloads, failures, storage restrictions and delayed responses.
Thanks to Alishara and xet7 !
Fixes#6722,
Resolve the site administrator role from the authenticated download token so the Files Report can open private-board attachments without requiring board membership. Keep ordinary private-board readers subject to membership checks.
Wrap the shared attachment controls inside narrow table cells so neighboring cells cannot intercept the download button.
Verify DOCX, image and PDF preview/download flows in Firefox and Chromium, plus negative authorization cases and eight focused Node suites.
Thanks to xet7 !
Follow moved documentation and component-owned templates, preserve reviewed translations, update date-format guards and account for AuthTraceBleed coverage. Recognize the backup containment guard and select the release npm version locally.
Exercise invitations with a separate member and an open sidebar, match the multiline swimlane composer, and wait for an interactable sign-in field before keyboard navigation.
Verified all 1216 Node suites, targeted checks in three browsers and nine real backup/restore integration checks.
Thanks to xet7 !
Replace the deleted local meteor-node-stubs fork with npm 1.2.30, update Rsdoctor to 1.6.4 for CVE-2026-61782, and remove obsolete build-tool exclusions from Dependabot. Add guards for tracked local manifests and patched lockfile versions.
Refresh APT indexes in the failed managed LXD base after a package download 404 and retry native and variant builds once. Keep unrelated failures fatal and preserve container state.
The app answered successfully but the EXIT trap waited forever after signaling only its launcher. Run the app in an isolated process group, terminate all descendants with bounded escalation, bound HTTP probes, and retain smoke logs. Cover successful startup, failed startup, stalled HTTP and TERM-resistant children on macOS.
Thanks to xet7 !
Keep libatomic1 in Docker runtime dependencies, package architecture-matched libraries and licensing in Linux ZIPs inherited by AppImage and Flatpak, and include the library in Sandstorm. Install it explicitly in native and emulated builders; remove inherited Linux runtime files from Windows and macOS repacks.
Thanks to Nissulya, PetitManchot and xet7 !
List Mac app archives with unzip, handle shallow telemetry scanner paths, ship the Docker deny-hash policy, and discard inherited foreign ELF addons that shadow target prebuilds. Add offline regressions and a native macOS archive check.
Thanks to xet7 !
Fix the legitimate invitation flow and unnecessary anonymous metadata subscription reported in #6692 comment 5811473169. Keep protected profile writes forbidden. Existing account blocks require administrator review; the separate history checksum report still needs the affected stored rows.
Thanks to Nissulya and xet7 !
Review 46 changed runtime files and refresh the telemetry inventory without weakening the release gates. Check the inventory locally before release mutations and in regression tests.
Fix CodeQL alerts 543 and 544 in the loopback login fixture with bounded anchored OAuth header parsing and generic HTTP failures. Remove the same stack response from the shipped Sandstorm adapter. Preserve successful signatures and proxy authentication; add malformed-input, negative source and live browser coverage.
Validated macOS ARM64 build, 1577-file source inventory, 42926-file bundle scan, ten OAuth/Sandstorm browser checks and focused security/release tests. Hosted CodeQL has not been rerun.
Thanks to GitHub CodeQL and xet7.
Defer password-form replay to the next task so Firefox finishes dispatching the original native submit event. Keep duplicate submissions blocked and retain useraccounts validation and two-factor handling.
Reproduced registration success followed by silent login failure in Firefox 156.0.1. Verified wrong-password rejection, click and Enter login, and cookie resume in the rebuilt app. Add native-event browser and negative unit coverage; Chromium password/TOTP and emailed-code checks also pass.
Thanks to xet7.
Fill two English placeholders in 232 non-English locales, retaining existing item names and exact interpolation tokens. Add locale-wide interpolation and negative checks plus Finnish, French, Japanese and Hebrew browser coverage.
Low-confidence drafts needing fluent review: Cherokee, Inuktitut, Greenlandic, Tigre, Wolaytta, Fulah, Venda, Tamazight, Nahuatl, Veps, Aromanian, Ladin, Kashmiri, Konkani, Dzongkha and Guarani. Record vocabulary sources and verification limits.
Thanks to xet7.
Verify due dates, labels and restored Unicode descriptions, record the complete Chromium baseline and unresolved failures, and document the import/menu fixes in Upcoming.
Thanks to xet7.
Wire organization bulk controls, preserve Jira ADF descriptions and activity-independent attachments, skip empty CSV rows, and reject invalid source shapes. Apply the import-disable switch and shared transfer validation to Trello HTTP imports.
Add documented dummy format fixtures, positive and negative browser checks, menu inventory guards, and an implementation audit with explicit coverage and format limitations. Thirty new Chromium checks and the corrected cross-board move regression pass on macOS ARM64 with FerretDB; focused Node suites pass.
Thanks to xet7.
Show radio choices beneath every creation title, even when empty. Reuse translated item names and register new label fallbacks without overwriting existing translations. Create boards sequentially with shared creation settings and prevent overlapping submissions.
Verify one/many flows in five Chromium tests on a macOS ARM64 build, plus title, board-batch, locale, shared-form and documentation checks.
Thanks to xet7 !
Start the shared scheduler and use valid daily expressions. Preserve all
application collections, BSON types, indexes and selected file versions in
portable archives, validate content before restore, and report failures.
Document every Admin Panel implementation and its validation limits. Add
real-database round trips, failure cases and browser coverage for scheduling,
restore, file downloads, every pane route and permission boundaries.
Thanks to markusst1982 and xet7 !
Fixes#6683,
https://forums.meteor.com/t/meteor-3-6-beta-x-rspack-2-x-native-typescript-node-js-26-pnpm-monorepos-instrumentation-pwas-and-a-friendlier-cli/64773/7
"Anyone testing Meteor 3.6-beta.1 with an AI coding assistant: I’ve published Meteor Agent Skills v1.1.0-beta.1.
It adds a meteor-typescript skill for meteor types, declaration providers, JavaScript IntelliSense, and type checking. The existing skills have also been updated for the beta.1 changes to Rspack, Node.js, Accounts, instrumentation, and DDP.
npx skills@latest add 'meteor/agent-skills#v1.1.0-beta.1'
Native declarations are opt-in, so the TypeScript guidance should help you keep a working @types/meteor or zodern:types setup until you choose to migrate. If an assistant gives you advice that doesn’t fit your app, please share the prompt, Meteor version, and relevant configuration. That feedback will help us improve the skills before the stable release."
Add September feature instructions, preserve guide forwarding links, and document backup coverage with descriptive screenshots from #6683.
Thanks to markusst1982 and xet7 !
Offer one multiline item or separate items for nonempty lines. Preserve order, prevent overlapping submissions and retain multiline list and swimlane names when editing and displaying them.
Thanks to AmigaAbattoir and xet7 !
Fixes#6714,
Remove inherited form-field minimum height, padding and margins from the native checkbox and center its flex row. Verify checkbox dimensions and label alignment in both Disabled and Enabled states.
Thanks to xet7.
Place the bold Member Settings: Date Format heading directly after board status. Show a live Enabled or Disabled checkbox label above the format selector, with rules before week-start and calendar-system controls. Reuse existing translations.
Thanks to xet7.
Place the existing immediate preference below Save and a horizontal rule. Verify the member Date popup displays date-format controls directly above week-start and calendar settings.
Thanks to xet7.
Move week start and calendar system below Date Format and remove their controls and writes from Change Settings. Capture all selections before saving, await week-start persistence, and retain existing translations.
Thanks to xet7.
Apply enabled member overrides before board and global defaults. Show inherited
status and saved formats using existing translations, and keep the card section
named Date. Cover rendering, exports, permissions, date-only formats and settings
isolation.
Thanks to xet7.
Capture sign-in submission before asynchronous provider selection, keep the form mounted, and complete only the chosen login path. Preserve password validation and show provider errors without redirecting.
Thanks to Nissulya and xet7 !
Fixes#6692,
Preserve local instrumentation and security logging, opt out of Meteor package statistics, and audit release bundles and repackaged binaries before publication.
Thanks to xet7.
Add the Occitan, Breton, Wolof, Walloon, Volapuk and Standard Moroccan Tamazight corrections to the upcoming release notes, including their verification and review status.
Correct the Tamazight SMTP port label and refresh audited review records for the translated checklist examples. Preserve their comma-separated input syntax while retaining target-language item names.
Show when a partial file audit stopped before inspecting storage, allow larger bounded inventories, and avoid orphan findings from truncated metadata. Restore available locations in Offices and People, wrap long report cells, and theme report action buttons.
Publish only the public boards a logged-in user has starred and include them in Favorites queries. Preserve private-board access rules when a previously public board changes visibility. Cover the public-board flow in Chromium and the selector boundary in Node tests.
Thanks to nalilord and xet7 !
Fixes#6713,
Read stored per-user list state reactively instead of caching an early expanded fallback in Session. Cover profile hydration and a Chromium reload with the persisted swimlane-scoped value.
Thanks to Nich01asFox and xet7 !
Fixes#6712,
Route internal card links within the current WeKan page and wait for the destination card subscription before treating it as unavailable. Cover same-tab navigation in Chromium and link classification in Node tests.
Thanks to carl-unique and xet7 !
Fixes#6711,
Share native file and bundled libmagic detection across uploads, extension correction and administrator file checks. Include the portable WASM database in runtime dependencies, stage native file for Snap and Linux setup, and verify release assets before smoke boot.
Cover portable detection with native file absent, missing dependencies/assets, isolated release layout and HTML diagnostics in Chromium, Firefox and WebKit.
Thanks to xet7 !
Add shared controls above Files, Filesystem integrity and Recovery for content/type checks, storage inventory and history correlation, with progress, cancellation and a downloadable bounded summary.
Report incomplete upload metadata separately from missing binaries. Compare filesystem and GridFS references, sizes, MIME/extension aliases, hashes and baseline evidence; correlate undo/activity history with candidate renamed or moved files. Preserve files and records, enforce administrator access and scan limits, and explicitly report unknown or unverified coverage.
Verify five focused Node suites and two browser scenarios in Chromium, Firefox and WebKit. Rerun header-first content detection in all three browsers. Document storage and verification limits.
Thanks to xet7.
Anchor runtime resolution at programs/server/npm so Office search indexing finds fflate, PDF indexing finds its worker and assets, and image conversion finds sharp. Parent resolution keeps source development working. Retain the file utility as a Docker runtime dependency for MIME detection.
Verify six focused regression suites and XLSX extraction plus PDF asset resolution in an isolated release container. Include negative archive and missing-dependency coverage.
Thanks to xet7.
Cache Meteor package transforms, make development DEBUG logging opt-in, and replace legacy CAS URL parsing while preserving callback encoding and custom HTTPS ports. Clean attachment metadata during browser fixture teardown, including failure paths.
Verify six focused Node suites, nine sequential browser checks across Chromium, Firefox and WebKit, and live HTTP requests with deprecation tracing. Document the package cache and verification limits.
Thanks to xet7.
Repair 841 wrong-language or previously audited bad values from the pull while preserving 77 Persian and Portuguese updates. Record exact rejected locale/key/value fingerprints, validate fallback tokens and content, and reapply existing audited corrections without replacing newer human wording.
Separate adjacent interpolation tokens and clarify pre-merge diagnostics. Actual merge replay, 11 focused Node suites, 21 human-preference checks and Veps/Tamazight UI cases in Chromium, Firefox and WebKit pass. The Quechua width label uses attested kinraynin; full grammar remains low confidence pending native review.
Thanks to xet7 !
Record sequential EVERYTHING coverage across Chromium, Firefox and WebKit, development database lifecycle checks, application fixes and the companion FerretDB test-runner fixes. State unsupported browser-specific coverage and the unconfirmed intermittent Firefox failure.
Thanks to xet7 !
Add Dev server nobuild to both build menus with shared URL and port options. Load checkout JavaScript and templates in memory using the installed Meteor toolchain; start its bundled MongoDB on the next port and wait for owned database shutdown.
Run server, Node, import, E2E and all supported Chromium, Firefox and WebKit regressions sequentially, stopping on failure. Preserve CommonJS semantics, native fetch, dynamic import namespaces and client source access boundaries. Document verified results and the remaining Windows and intermittent Firefox limits.
Thanks to xet7 !
Read the current user once and only offer description recovery when its published profile explicitly enables it. Session initialization can publish the user before the profile, which previously threw from the details-pane close handler.
Exercise missing users and profiles, recovery opt-in, unchanged drafts, save and discard from the actual handler. A browser regression removes the profile and closes a card without errors; it passes in Chromium, Firefox and WebKit.
Thanks to xet7 !
Run portable drag-sort tests in Chromium, Firefox and WebKit and include global feature toggles in serial runs. Keep drag gestures inside the viewport and focus comment inputs with a real click before testing mention hit targets. Require the swimlane creation popup and its input instead of accepting an unrelated visible lane.
Affected tests passed in all three browsers. Raw CDP touch remains Chromium-only.
Thanks to xet7 !
Use document-ready navigation for report permissions, board comment publication and phone board layout. Explicitly await the non-admin session before testing version lookup authorization. Positive and negative comment visibility and phone geometry assertions remain intact and pass in Chromium.
Thanks to xet7 !
The non-Latin keyboard bridge dispatches synthetic events on document. The shortcut filter previously called closest on that non-element target, throwing during Tab navigation on signed-out pages. Fall back to the focused element and safely reject missing focus.
Unit tests cover native and document-targeted events, form and interactive controls, editable content, selections, disabled shortcuts and Escape. The signed-out layout browser regression passes in Chromium.
Thanks to xet7 !
Replace idle-network navigation waits with document readiness followed by the existing user, control, sorting, search and permission assertions. Template copying, board sort/search and server pagination regressions pass in Chromium on the source server.
Thanks to xet7 !
Preserve Requested By and Assigned By accessibility checks and description persistence assertions without waiting for persistent Meteor connections to become idle. The card members and description browser group passes in Chromium.
Thanks to xet7 !
Keep voted-state, database membership and visible count assertions after document-ready reloads. Positive, negative and vote-count regressions pass against the source development server in Chromium.
Thanks to xet7 !
Avoid idle-network reload waits with persistent Meteor connections. Require the seeded label to be selected and displayed instead of accepting an open popup as success.
The focused label regression and the remaining date-editor Chromium tests pass against the source development server.
Thanks to xet7 !
Replace idle-network navigation waits with document readiness and the existing layout, session and persistence assertions. Meteor SockJS requests can stay active after the page is ready. Verify immediate comment persistence, list order, view changes and repeated sign-in page loads on the source development server.
Remember the requested path before the sign-in guard redirects an initial page load. After authentication, restore that local URL instead of always opening All Boards, using the router redirect API to preserve deployment prefixes. Keep external targets rejected and sign-in enforcement unchanged. Verify direct My Cards navigation, assigned-card inclusion, unassigned-card exclusion and cookie/Sandstorm regressions.
Count unread profile records for the bell without requiring activity details that are subscribed only by the open drawer. Cover read and missing records, and verify a watching member receives a live mention alert in a separate browser session.
Align the sign-up browser regression with the adopted A05 accessibility behavior. Tab reaches both visibility buttons; Space and Enter reveal and hide passwords without submitting a valid form. Enter in the confirmation field still creates the account, and test accounts are cleaned up on failure.
Replace idle-network reload waits in membership and search regressions with document readiness and existing visible-control assertions. Verify member removal instead of silently bypassing missing controls. Bound HTTP and HTTPS readiness checks and reject failed, unhealthy or stalled servers rather than skipping all browser tests successfully.
Avoid a stale empty swimlane or list cache in the destination subscription ready callback. Cross-board moves now select a valid destination and enable Done without waiting for another user change. Cover delayed cache refresh alongside stale requests and rejected operations.
Keep mobile sidebar controls below headers that appear after login or a route change. Cover delayed rendering, resize updates and destroyed templates, and wait for document readiness in the mobile Search browser regression.
The muted-assignment fix in 4be77e7189 and scoped-subscription/settings fix in 03e8a6bf66 are already present. Add browser regressions that capture SMTP email for board, list and card watching, comments, title changes and new cards; verify muted assignments and member email opt-outs stay silent. Correct the channel settings guide and document delivery prerequisites and test setup.
Ten SMTP scenarios and six existing tray/settings scenarios passed across Chromium and Firefox against the current Meteor bundle with MongoDB. Focused recipient and notification Node suites passed. External inbox delivery and FerretDB runtime were not verified; the latest reported delivery failure was not reproduced.
Thanks to Nissulya and xet7 !
Fixes#6658,
Add board controls for minicard collapse, labels above the title, checklist due dates and the outer checklist heading. Keep previously collapsed content accessible when its control is hidden, and preserve stored content and existing defaults.
Add personal and admin date-only formats across calendars and exports. Extend the existing smaller font presets with compact checklist and settings spacing. Cover persistence, permissions, layout and date behavior with unit and browser tests.
Thanks to xet7 !
Use the Markdown textarea for every editor. Remove the retired environment and Snap options from server settings, all platform launchers, build scripts and current documentation. Keep released and archived changelog entries intact.
Keep notification HTML escaping independent of editor configuration, preserve attachment metadata and mention behavior, and cover Markdown and emoji round trips in browser tests. Also import the translator used by attachment error handlers.
Thanks to xet7 !
Repair relative ordering and destination dialog lifecycles, retain failed operations, and make checklist, rule and attachment controls keyboard accessible. Improve scaled layouts and reuse canonical translated labels.
Add source-execution and Chromium/Firefox regression coverage, including awaiting background persistence before reload. Record implementation decisions and validation limits.
Thanks to xet7 !
Initialize autocomplete once per editor, place its styled menu above card windows, and fall back to the textarea when Summernote is unavailable. Verify pointer and keyboard selection for password and LDAP account fixtures.
Thanks to bbyszio and xet7 !
Fixes#6704.
Bind sortable initialization and reactive work to the replaceable list body, and clean it up on destruction. Cover repeated collapse, expansion, reload and drag permissions.
Thanks to Nich01asFox and xet7 !
Fixes#6705.
Recover the project repository used for submission, retry recipe creation while Launchpad indexes main, and reject incomplete snaps before remote cleanup. Run the compatibility wrapper inside Snapcraft's own environment.
Verified all 1173 Node suites sequentially, workflow lint and script syntax. Regression coverage exercises delayed refs, unrelated errors, recovery paths, invalid downloads and launcher argument handling without remote writes.
Thanks to xet7 !
The v11.85 logs show ARMHF failing during Launchpad polling after TLS EOF,
PPC64EL and S390X reaching the six-hour job cap while queued, and AMD64
successfully publishing to the Snap Store before its GitHub attachment hung.
Recover existing recipes before considering a new submission, and share one
five-hour waiter budget across attempts on every architecture. Prepare an
isolated complete source snapshot with Git metadata and logs outside the
project hash, and distinguish architectures so matrix recipes stay separate.
Move downloaded artifacts out before retrying recovery.
Bound GitHub snap attachments and verify remote names and sizes. Exercise
these paths with fixture repositories and mocked network commands, covering
TLS failures, absent recipes, denied credentials, queue timeouts, invalid
artifacts and failed uploads. Workflow parsing, shell syntax and Actionlint
pass. The full 1,172 Node suite sweep is verified with corrected guards rerun.
Live remote builds and publishing have not been run.
Thanks to xet7 !
Place the global date format checkbox and selector between Board member
list and Wait Spinner in Visibility settings. While enabled, use the
selected format for members and guests, call the card section Date and
hide its personal format selector. Preserve personal preferences so they
return when the override is disabled.
Apply the same policy to date displays, filter parsing and browser export
requests. Validate supported formats and retain admin-only settings writes.
Add positive, negative and browser regression coverage and document the
setting. Translate the new label without changing existing translations.
Lower-confidence new translations include Acehnese, Aymara, Cherokee,
Inuktitut, Kashmiri, Nahuatl, Veps, Volapuk and Wolaytta; native-speaker
review would improve these labels.
Thanks to Mathia2 and xet7 !
Fixes#6703,
Bind card and board identifiers explicitly on links rendered under lexical each loops. Open the subscribed card popup from the link rather than the unrelated outer Blaze data context, while preserving modified-click navigation.
Regression tests cover outer-context mismatch, nested click targets, missing identifiers, modified clicks, and sequential card popup selection on both pages. The fresh bundle passes both page regressions in Chromium, Firefox and WebKit using polling reactivity.
Thanks to Mathia2 and xet7 !
Fixes#6702,
Preserve the newer Traditional Chinese wording in historical repair records. Locate the card read-only branch by its matching conditional rather than fixed indentation after optional field grouping.
Verified all 1169 Node suites pass, translation pull/fill protection passes, and missing requester or assigner avatars still fail the repaired guard.
Thanks to xet7 !
Record notification, empty-card-section, WIP-group and duplicate-checklist fixes with their regression coverage. Link the tested FerretDB oplog changes and retain #6701 in TODO Later until the affected snap's database is confirmed. Verify the exact resumed user in the session regression.
Thanks to Nissulya, fabiosalles, Mahgozar, rmb82 and xet7 !
Resume an LDAP account with 1600 directory users, verify reloads before and after a bulk deletion, keep the directory out of client publications, and reject invalid cookies. The test passes against MongoDB and patched FerretDB; retain #6701 pending confirmation of the reported deployment's backend.
Thanks to Nissulya and xet7 !
Refs #6701,
Route group edits through a validated server method with board/site-admin authorization. Compare the saved group array before updating so concurrent edits are not overwritten and database backends need no positional array update. Unit and Chromium tests cover editing, invalid lists and limits, and unauthorized users.
Thanks to Mahgozar and xet7 !
Fixes#6699,
Check the persisted profile notification array and exercise comments delivered to explicit list subscribers on otherwise muted boards.
Thanks to Nissulya and xet7 !
Fixes#6658,
Keep the existing client insert path, wait for its server callback, and retain the editor on failure. Verified click, Enter and Ctrl+Enter creation and persistence in Chromium, plus duplicate-submit and retry unit tests.
Thanks to rmb82 and xet7 !
Fixes#6700,
Stop handled Enter events before the enclosing checklist submits them again. Await the insert, scope the textarea to its form, and reject duplicate submits while the write is pending. Keep failed edits available for retry.
Thanks to rmb82 and xet7 !
Fixes#6700,
Restore native checkbox visibility and keyboard interaction inside the WIP group editor, and resolve list rows through the template helper when editing saved groups.
Thanks to Mahgozar and xet7 !
Fixes#6699,
Render Dates, Members and Sort groups only when their fields can display content. Keep a group when any independently enabled field remains visible.
Thanks to fabiosalles and xet7 !
Fixes#6696,
Keep muted board assignments silent while honoring explicit list and card subscriptions. Resolve notification controls against their owning settings scope and show the selected option.
Thanks to Nissulya and xet7 !
Fixes#6658,
Activate the card link instead of its date badge and open the Board View popup before choosing timeline, chart and Gantt views. The complete 24-case Chromium title suite passes sequentially.
Update the regression check to require one value-guarded ReactiveVar setter shared by the viewer and startup autorun. The focused external-link suite now passes all 24 tests.
Expose the renderer through a Meteor main module, synchronize security settings without repeated reactive invalidation, and route canonical avatar downloads around the automatic Meteor-Files handler. Update browser tests for the visible Board View and list menus, correct fixture IDs and authentication headers, and wait for the settings publication where direct Mongo updates are used.
Use the explicit Tigre Member and Members pair across seven board, filter, selection and rule controls. Reuse reviewed Tigre control terms and add former-seed coverage.\n\nThanks to xet7 !
Replace Tigrinya-specific Speed and Storage spellings with forms attested in the Tigre corpus and native Tigre Wikidata interface. Add exact and former-seed coverage.\n\nThanks to xet7 !
Use corpus-grounded Tigre terms across seven Font, Invite People and Change permissions controls. Add exact and former-seed regression coverage.\n\nThanks to xet7 !
Record corpus evidence, residual overlap, limitations and regression coverage for the repaired Rename and board-background controls.\n\nThanks to xet7 !
Use corpus-attested Tigre name, change, image and background terms across eight Rename and board-background controls. Add exact and former-seed regression coverage.\n\nThanks to xet7 !
Use corpus-attested Tigre terms across eight Overall progress, CPU usage, Card loading and Remaining time controls. Preserve CPU and add exact former-seed coverage.\n\nThanks to xet7 !
Replace the misleading physical-weight seed in three Severity controls with a corpus-grounded Tigre severity-level compound. Record its compound grammar as low confidence and add negative coverage for the former seed.\n\nThanks to xet7 !
Use the Tigre export action attested in a software-context corpus entry across the generic, list and board export controls. Add exact and former-seed regression coverage.\n\nThanks to xet7 !
Record the evidence, residual overlap counts, test coverage and source commit for the reviewed Tigre administrative and location controls.\n\nThanks to xet7 !
Use corpus-attested Tigre terminology for administrator, organization administrators, confirmation, type, location and adding a location. Add focused coverage for all corrected controls and their former Tigrinya seeds.\n\nThanks to xet7 !
Replace ten incomplete English seeds with the attested calendar-noun family while preserving named systems, calculated epochs, and moon-sighting distinctions.
Thanks to xet7 !
Replace seventeen Azerbaijani, English, and incomplete labels with a coordinated modern date-reckoning family while preserving every named calendar and variant distinction.
Thanks to xet7 !
Replace the incomplete English calendar name with an Ewe calculation-and-epoch draft while preserving the distinct astronomical variant.
Thanks to xet7 !
Replace the final tracked Finnish paragraph with a Veps draft while preserving every executable operator, quote, escape, grouping, and regular-expression example.
Thanks to xet7 !
Replace four Finnish keyboard-shortcut values with a coordinated Veps keyboard-command draft while preserving state transitions, route behavior, and the literal question-mark binding.
Thanks to xet7 !
Replace the Finnish checkbox label with Veps arithmetic-addition and list-top terminology already verified for the numeric total tooltip.\n\nThanks to xet7 !
Record the five repaired UI roles, source bindings, descriptive terminology basis, focused checks, and remaining low-confidence review.\n\nThanks to xet7 !
Replace the Finnish accessibility family with a consistent Veps access-for-everyone draft across navigation, settings, fields, and the public empty state.\n\nThanks to xet7 !
Record the corrected card-loading semantics, native terminology basis, focused verification, and remaining low-confidence review in the audit and Upcoming changelog.\n\nThanks to xet7 !
Replace wrong-language labels, stale manual loading guidance, and a Russian multi-selection hybrid with a coherent Veps card-loading group. Preserve the automatic threshold behavior and exact operator configuration literals.\n\nThanks to xet7 !
Update the audit, native evidence review, and Upcoming changelog with the verified voting-side repair and its remaining confidence limit.\n\nThanks to xet7 !
Replace four Finnish participant and action labels with Veps forms based on the native voting, support, defense, and opposition entries. Add semantic regression checks.\n\nThanks to xet7 !
Replace Tshivenda and Finnish headings with an attested Veps summary term and the locale's existing status label. Add exact regression checks.\n\nThanks to xet7 !
Update the short audit, detailed evidence, Veps review, and Upcoming changelog with the verified local repair and expanded wrong-language candidate scan.\n\nThanks to xet7 !
Replace Zulu card-number labels with Veps wording, preserve the search syntax placeholders, and add regression coverage for the complete search group.\n\nThanks to xet7 !
Use the rendered native phrase for addition of numbers and preserve custom number fields marked for list top. Full compound and case grammar remains under review.
Preserve current versus generic usage and average system load. Reuse local Veps terminology; full computing compounds and core vocabulary remain under review.
Preserve possible login delay, login-required message, loading indicator and CPU condition until background work finishes. Full computing terminology and grammar remain low confidence.
Preserve zero remaining, missing recent completions, weekly pace and conditional completion date. Full projection, recency and count grammar remains low confidence.
Replace Tshivenda vote-sorting and detail text and Finnish voting heading with Veps vocabulary. Sorting case and full context terminology remain under review.
Preserve QR and manual setup alternatives, six-digit codes, every-login requirement and invalid-code retry. Full computing compounds and native grammar remain low confidence.
Preserve enable and disable actions, popup title and account-specific enabled status. Full two-factor terminology and grammar remain low confidence; setup instructions still require repair.
Replace Tshivenda with a project-code paraphrase and retain ID and owner/repo alternatives. Native project and code vocabulary is attested; full multi-tracker terminology remains low confidence.
Preserve external tracker, background checks every 15 minutes, and the immediate manual action. Full draft grammar and computing loans remain low confidence.
Replace the Tshivenda empty-source option with a Veps inactive-state paraphrase. Keep enabled, failed and never-synced states distinct. Derived noun and complete grammar remain low confidence.
Preserve pending, success, error and stop meanings and the exact error placeholder. Derived synchronization noun and case grammar remain low confidence.
Use native memory-usage and size wording with the precise RSS acronym. Full computing grammar and native expansion of resident set remain under review.
Verify Bara search with a positive color entry; retain unresolved magenta without misclassifying missing results as language approval.
Thanks to xet7 !
Inspect Cunia dictionary entries and identify the newer Bara dictionary for unresolved magenta attestation. Preserve existing translation and counts.
Thanks to xet7 !
Preserve possible successful import despite errors and the All Boards location. Probability paraphrase and complete grammar remain low confidence.
Thanks to xet7 !
Preserve the requirement for at least one administrator, replacing the wrong permission explanation. Complete grammar remains low confidence.
Thanks to xet7 !
Replace French checked and unchecked labels with state phrases, keeping them distinct from imperative actions. Checkbox terminology and complete grammar remain low confidence.
Thanks to xet7 !
Replace Arabic with a Tamazight draft preserving star action and shortcut destination. Use contextual here rather than a physical bar noun. Software adaptation remains low confidence.
Thanks to xet7 !
Add optional checklist dueAt with the shared selected-calendar picker and badge. Return checklist and item deadlines from checklist and single-card GETs. Accept validated timezone-qualified ISO deadlines in checklist/item POST and PUT; null clears, omission preserves. Keep existing write permissions and ownership checks. Node regression and Chromium save/clear test pass.
Thanks to rmb82 and xet7 !
Fixes#6693,
Read Blaze event data instead of surrounding template data. Chromium regression verifies mixed selections retain added labels and members; unit coverage distinguishes template and event contexts.
Thanks to AmigaAbattoir and xet7 !
Fixes#6694,
Correct reversed assign/unassign handlers, isolate bulk label selectors from the sidebar create-label action, stop propagation, await mutations and ignore repeated clicks in a double-click sequence. Behavioral regression verifies action direction and popup wiring.
Thanks to AmigaAbattoir and xet7 !
Fixes#6694,
Preserve non-undoable deletion and loss of all related activities. Keep vote and card scopes distinct. Full contextual grammar remains low confidence.
Thanks to xet7 !
Replace eight French fragments using existing native terminology. Preserve list ownership, list versus swimlane destinations, checklist identity and distinct add/remove actions. Complete contextual grammar remains low confidence.
Thanks to xet7 !
Reuse established card, checklist item and label terminology. Test distinct add/remove meanings and preserve translation tokens and human preference. Contextual phrasing remains low confidence.
Thanks to xet7 !
Document dictionary evidence, full-phrase uncertainty and passing regression checks. Reconcile the live queue to 195 pending original findings.
Thanks to xet7 !
Replace French rule labels using existing checklist terminology and the IRCAM punctuation-comma sense. Preserve the required comma-separated input instruction and regression coverage. Complete phrase grammar remains low confidence.
Thanks to xet7 !
Keep conversion before restricted deletion and preserve same-title populated counterpart and continuation question. Full compound grammar remains low confidence; removed migration behavior is unchanged.
Preserve certainty question and restriction to duplicate lists sharing a name and containing no cards. Full passive and shared-name grammar remain low confidence for native review.
Retain future discovery and repair statement with explicit continuation question. Match the description scope. Full question grammar and technical compounds remain low confidence; no removed migration is activated.
Retain discovering and repairing missing or corrupted lists in board structure. Full passive grammar and computing adaptation remain low confidence. No removed migration is activated.
Use IRCAM description noun in minicard-description and detailed-description labels while retaining summary and search shorthand. Surrounding phrase grammar remains under review; original ledger provenance is preserved.
Use dictionary-attested description wording distinct from summary. Preserve destination-card meaning and literal title/description JSON properties. Complete instruction grammar remains low confidence for native review.
Preserve moving tasks out of this list or increasing the WIP limit. Avoid deletion wording from the French seed. Complete grammar, numerical-limit adaptation and politeness remain low confidence for native review.
Preserve distinct creator/member card scope and watched boards, lists or cards. Reuse notification wording and native watchlist components; complete passive and object agreement remain low confidence. No active client/model references were found for these catalogue keys.
Replace English country-only wording with calendar, country and moon-seeing components. Complete nominalized phrase and grammatical agreement remain low confidence for native review.
Use calendar and moon nouns with dictionary-attested sighting wording. Preserve distinct tabular calendar semantics and original correction provenance. Complete compound grammar remains low confidence for native review.
Replace clock-only and English labels with calendar, table/count and beginning wording. Explicit Julian epoch dates retain the civil versus astronomical distinction. Complete technical compounds remain low confidence for native review.
Date source and dictionary evidence in audit and Upcoming changelog. Retain passive negation relative grammar and broader uncertain translation scope. Audit and changelog checks pass.
Thanks to xet7.
Keep never-for-this-board and creator-or-member card notification scope distinct. Use IRCAM future negation and participant terms with existing notification wording. Four focused suites pass; passive inflection negation placement and complete relative grammar remain under native review.
Thanks to xet7.
Date reminder-root and temporal evidence with source commit and passing checks. Keep aspect accessibility Aromanian and all broader uncertain findings in scope.
Thanks to xet7.
Preserve approaching now and past deadline meanings and exact placeholders. Use IRCAM reminder marked variant and approach/passage roots with existing deadline terminology. Four focused suites pass; inflection and temporal aspect remain under native review.
Thanks to xet7.
Date source and dictionary evidence and update Upcoming changelog. Retain native clause and platform/load-average review limits. Audit and changelog checks pass.
Thanks to xet7.
Describe time since operating-system start, matching os.uptime and the Information view. Use IRCAM beginning, passage and operating-system terms; avoid film runtime and physical platform senses. Four focused suites pass after removing a duplicate test constant. Relative clause and computing-startup adaptation remain under native review.
Thanks to xet7.
Date the source correction and primary dictionary/software evidence. Preserve remaining native grammar review scope and record passing audit and changelog checks.
Thanks to xet7.
Replace the speculative derived noun with IRCAM anmala, also attested in MediaWiki default-sort software wording. Preserve original ledger before-value and add negative regression for the superseded form. Four focused suites pass; full grammar and other technical compounds remain under review.
Thanks to xet7.
Update dated audit progress and Upcoming changelog with the source commit, lexical evidence and passing checks. Preserve low-confidence grammar and technical compounds in the full audit scope.
Thanks to xet7.
Preserve board data integrity checks and repair scope, including list ordering card positioning and swimlane structure. Use IRCAM lexical evidence and record derived ordering noun, inflection and computing collocations as low confidence. Four focused suites pass after correcting the regression test root constant; removed migration code remains removed.
Thanks to xet7.
Date source-backed automatic loading repairs and regional board terminology review. Preserve Arabic-script and broader native-review scope. Audit and changelog checks pass.
Thanks to xet7.
Translate automatic card loading, threshold, live count and exact operator environment overrides. Replace personal laziness sense with on-demand loading. Carry reviewed board forms into matching regional prose and review eight independent core roadmap clone and notification values. Five suites pass including 28 loading behavior checks; Arabic-script and native technical reviews remain open.
Thanks to xet7.
Update dated audit progress and Upcoming changelog with source and morphology evidence. Record stale loading guidance as unfinished and retain all broader uncertain findings. Audit and changelog checks pass.
Thanks to xet7.
Review all council-term sentences against English and use the established taxta board noun. Preserve plural case and possessive morphology, with explicit selection schedule search invitation and multi-board corrections. Translate visibility with existing labels and preserve HTML. Five focused suites pass; stale loading instructions and remaining technical phrasing stay in scope.
Thanks to xet7.
Date the whole Khmer council-sense sweep and related semantic fixes. Update the short audit summary and Upcoming changelog with source commit and passing validation while retaining full technical review scope.
Thanks to xet7.
Review all council-term sentences against English and replace the governing body sense with the established board noun. Correct financial deposit, list purchasing, multi-board calendar and extra board in swimlane selection. Translate visibility terms using existing locale labels and preserve HTML. Five focused suites pass; full native technical compounds remain under review.
Thanks to xet7.
Update dated audit and Upcoming changelog with source evidence, alias correction and validation. Keep broader council-term sentences and native computing collocations under review. Audit and changelog checks pass.
Thanks to xet7.
Distinguish Kanban boards from administrative councils in board labels and restore actions. Use Archive on All Boards in complete instructions. Preserve Khmer locale aliases and historical ledger before-values. Five focused suites pass after correcting alias bookkeeping; remaining council wording and native computing collocations stay under review.
Thanks to xet7.
Update dated audit progress and Upcoming changelog with source commit and validation. Keep Uzbek Khmer remaining archive locales and native collocation reviews in scope. Audit and changelog checks pass.
Thanks to xet7.
Use existing Archive and All Boards labels in Malayalam Marathi Gujarati Punjabi Nepali Sinhala Mongolian and Kazakh. Preserve restoration terminology and replace obsolete header navigation, including the English Sinhala Archive label. Five focused suites pass; full native collocation review remains part of the broader audit.
Thanks to xet7.
Date the Irish Serbian and Tagalog corrections, cite Irish dictionary evidence and preserve native-review limits. Update the short audit summary and Upcoming changelog with the local source commit. Audit and changelog checks pass.
Thanks to xet7.
Use the Archive section on All Boards and preserve existing locale labels. Choose the attested Irish bring-back sense, preserve Serbian board terminology, and replace mixed English in Tagalog. Full Irish navigation grammar remains a derived translation pending native review. Five focused suites pass.
Thanks to xet7.
Accept template whitespace in the regional row assertion. Confirm both LTR and RTL Chromium layout tests pass and record the existing first-line marker fix under the Upcoming release.
Thanks to xet7.
Update dated audit and Upcoming with 21-file repair and native terminology evidence. Preserve low-confidence computing phrases and remaining locale reviews. Three documentation suites pass.
Thanks to xet7.
Correct current Archive locations and Zulu place labels. Replace Bulgarian in Macedonian, phonetic English in Urdu and mixed English in Maltese. Five focused suites pass; Maltese/Zulu computing collocations remain low confidence and other locale reviews remain open.
Thanks to xet7.
Update dated audit and Upcoming with 34-file repair, six consistent place labels and superseded Galician acceptance. Retain other locale and native/UI review limits. Three documentation suites pass.
Thanks to xet7.
Replace obsolete locations and Persian-in-Arabic/Malay-in-Indonesian prose. Correct six base/regional Archive place labels, keeping guidance consistent. Supersede the Galician unchanged acceptance with its dated correction. Focused locale/menu suites pass; other languages and browser paths remain open.
Thanks to xet7.
Correct obsolete locations and three Archive place labels. Review independent Argentine Spanish, Peruvian Spanish and Brazilian Portuguese guidance; update matching regional copies elsewhere. Five focused suites pass. Czech main-menu wording retained; Serbian terminology and other locales remain open.
Thanks to xet7.
Update dated audit and Upcoming with repaired languages, exact-copy regional handling and correct Swedish retention. Preserve other locale, native grammar and browser verification requirements. Three documentation suites pass.
Thanks to xet7.
Replace obsolete header locations with current Archive section and All Boards page, using existing localized place labels. Update reviewed base languages and identical regional copies; retain correct Swedish main-menu wording. Five focused locale/menu suites pass; other locales and browser paths remain to review.
Thanks to xet7.
Update dated progress and Upcoming with the source commit. Preserve native grammar, unexecuted restoration UI and other locale-location review requirements. Three documentation suites pass.
Thanks to xet7.
Replace outdated home-header button wording with the active Archive section on All Boards. Update English and six variants and replace complete French Tamazight prose. Five locale/menu suites pass. Tamazight full phrasing and other locale locations remain under review.
Thanks to xet7.
Record active All Boards Archive menu source and commented-out header control. Keep English and corresponding locale repairs open rather than reproducing obsolete navigation. Existing All Boards and audit suites pass; no browser execution or locale changes claimed.
Thanks to xet7.
Retain the native-supported home-page term. Distinguish plug and clothing-button senses from UI terminology and leave full French guidance unresolved. Record source evidence and next verification requirements without claiming fluency or browser execution. Three audit/changelog suites pass.
Thanks to xet7.
Update dated audit and Upcoming with the source commit, primary dictionary entries and original-versus-unflagged counts. Preserve relative grammar and dormant-key browser verification limits. Three documentation suites pass.
Thanks to xet7.
Use IRCAM time/remain terminology in a full draft. Keep remaining duration distinct from configured lockout period and failure window. Four locale suites pass; relative verb and countdown collocation remain low confidence.
Thanks to xet7.
Document native wording evidence, full causal-phrase uncertainty and absence of browser rendering proof. Update dated audit counts and Upcoming with the exact source commit. Three documentation suites pass.
Thanks to xet7.
Retain plural blocked users, current state and repeated login-failure reason in a full draft. Keep empty-state and individual lockout messages distinct. Four locale suites pass; native causal phrasing remains low confidence.
Thanks to xet7.
Update dated audit counts and Upcoming with the source commit. Record native vocabulary evidence, full-phrase review limits and unexecuted authentication UI coverage. Three audit and changelog suites pass.
Thanks to xet7.
Preserve the temporary period, repeated failed logins and retry-later instruction in a complete draft. Four locale suites pass; full causal and temporal grammar remains low confidence.
Thanks to xet7.
Update dated audit and Upcoming with native restore terminology, exact source commit and successful archive-title cancellation test. Preserve native-compound and board-restoration verification limits. Three documentation suites pass.
Thanks to xet7.
Replace French and Arabic labels with native archive-list wording and the attested restore verb. Preserve distinct archive/restore actions. Four locale suites pass; Chromium verifies the archive title and cancellation (1 passed). Board restoration has source coverage; native computing compounds remain under review.
Thanks to xet7.
Update dated audit progress and Upcoming with the source commit, native terminology references and command-preservation checks. Retain full-phrase uncertainty and unexecuted browser-path limits. Three audit and changelog suites pass.
Thanks to xet7.
Translate the complete server-error message with native send/error and IRCAM server/installation terminology. Preserve both exact shell commands and separate Snap/Docker lines. Four locale suites pass; full clause grammar remains low confidence.
Thanks to xet7.
Document IRCAM terminology, CPU metric semantics and remaining uncertainty. Correct stale unrun-browser statements using the existing dated executions. Record the conference-reference check without treating French proceedings as native terminology evidence. Three audit and changelog suites pass.
Thanks to xet7.
Use IRCAM tamsdagt, attested for architecture and modular architecture. Preserve the os.arch CPU metric and its distinct OS type/platform meaning. Add positive and negative locale/source regressions; four suites pass. Computing specialization remains low confidence.
Thanks to xet7.
Document native sources, actual OS metric semantics and remaining low-confidence adaptations. Update translation progress and Upcoming with the source commit. Three documentation and changelog suites pass.
Thanks to xet7.
Execute the actual translation method and installed formatter against all 246 locale files. Verify one percent placeholder and exact substitution of a value containing percent. Regression passes; this does not certify native fluency or lazy loading.
Thanks to xet7.
Replace earlier derived reception noun in two Tamazight drafts while preserving time placeholders and original provenance history. Record remaining genitive and full-clause uncertainty. Four translation checks pass.
Thanks to xet7.
Replace three wrong-language field labels while preserving received/due distinctions and existing date terminology. Record software reception and deadline uncertainty. Four translation checks pass.
Thanks to xet7.
Keep current runtime coverage concise while preserving dated evidence and native/browser limitations. Revalidate live queue totals and confirm no localhost app is running. Audit regression passes; no translation counts are changed.
Thanks to xet7.
Exercise actual translated creation date and ascending/descending sort queries; reject invalid date periods. Deterministic date helpers isolate parser metadata mapping. Focused regression passes.
Thanks to xet7.
Replace French with complete five-field date export wording and creation search labels. Preserve field order and record software reception and deadline uncertainty. Four translation checks pass.
Thanks to xet7.
Exercise actual localized has queries for present and absent start/end dates and reject those predicates for sorting. Parser regression passes without changing query behavior.
Thanks to xet7.
Replace four Arabic values with existing primary-dictionary start/end terms and preserve date-field distinctions. Four translation checks pass. Record actual rule and search mappings without claiming browser validation.
Thanks to xet7.
Refresh reconciled counts and upcoming changelog with source evidence and native software compound uncertainty. Documentation checks pass.
Thanks to xet7.
Replace French descriptions with native waiting-indicator terminology while preserving exact animation identifiers. Exclude textile spinning senses and record compound adaptation uncertainty. Four translation checks pass.
Thanks to xet7.
Replace Arabic accept and French invitation status with complete drafts preserving not yet. Record secondary verb provenance and low-confidence passive grammar. Four focused translation checks pass.
Thanks to xet7.
Update audit and upcoming changelog with contextual evidence while preserving unresolved wording and counts. Documentation checks pass.
Thanks to xet7.
Attach Common Voice email-specific already-used predicate evidence to the prior correction without changing the translation. Preserve remaining agreement and browser limits and distinguish unrelated import uncertainties. Ledger check passes.
Thanks to xet7.
Record current Common Voice translation resource paths and distinguish project metadata and obsolete URLs from native terminology evidence. Repository wording remains unresolved; no locale changes are claimed.
Thanks to xet7.
Distinguish source repositories from data warehouses and update verbs from nouns. Record native software-message evidence and complete pending action requirements without falsely counting a repair.
Thanks to xet7.
Record primary relative marker and possession studies directly in the correction ledger. Preserve original before and after values and retain plural-clause uncertainty. Ledger regression passes.
Thanks to xet7.
Record Taifi prepositional possession evidence and distinguish it from unverified plural relative agreement and complete software wording. No locale is declared repaired by this review.
Thanks to xet7.
Replace Arabic with a complete users-with-domain-name draft and preserve automatic addition. Record primary terminology and assembled relative-clause uncertainty. Four focused translation checks pass.
Thanks to xet7.
Update translation audit evidence and upcoming changelog with primary terminology and unresolved native agreement. Audit and changelog checks pass.
Thanks to xet7.
Replace French stored title with complete restore, cards and lost components from the primary dictionary. Record adjective agreement and software metaphor uncertainty. Four focused translation checks pass; no migration implementation is restored.
Thanks to xet7.
Update translation audit progress and upcoming changelog with actual formatter coverage and remaining native and browser limitations. Documentation checks pass.
Thanks to xet7.
Convert scalar TAPi18n arguments to sprintf options so search errors replace percent placeholders. Preserve named options, explicit format arrays and zero. Actual translation method and installed formatter regressions pass alongside parser and lazy-loading checks.
Thanks to xet7.
Document real translated limit-query regressions and the zero-limit distinction without claiming browser interpolation or native acceptance. Audit and changelog checks pass.
Thanks to xet7.
Exercise negative and nonnumeric translated limit queries and preserve offending values for percent interpolation. Verify positive and zero-limit queries remain accepted. This checks parser dispatch rather than native fluency or browser rendering.
Thanks to xet7.
Replace French with software validation wording and retain the positive integer condition and percent placeholder. Record native message and dictionary evidence with cross-variety grammar limits. Four focused translation checks pass.
Thanks to xet7.
Refresh the audit summary and upcoming changelog with the primary terminology review, preserving all unresolved findings. Three audit and changelog checks pass.
Thanks to xet7.
Document primary dictionary distinctions between computing validation and disability senses of invalid. Preserve the pending full error and percent placeholder rather than introduce a misleading literal substitution.
Thanks to xet7.
Update the short audit summary and detailed evidence with terminology provenance, unchanged pending counts and the exact email-domain condition still needed in automatic-addition wording. Audit and changelog checks pass.
Thanks to xet7.
Use the documented computing-domain noun instead of geographic region in domain lists and board sharing. Preserve translation tokens and record primary lexicon provenance with native usage limits. Four focused translation checks pass.
Thanks to xet7.
Keep the English prose explicitly pending. Syntax, correction and review checks pass; completeness correctly exposes one untranslated Tamazight placeholder instead of accepting cosmetically different English.
Preserve the assetlinks.json filename and JSON marker, distinguish activation from content, and record low-confidence custom terminology for native review.
Preserve specific calendar names with the official native calendar noun. Low confidence combined naming remains under broader native and browser review.
Thanks to xet7 !
Preserve the national qualifier with native component terminology. Low confidence full naming remains under broader native and browser review.
Thanks to xet7 !
Compose native calendar noun with an attested Hebrew adjective. Low confidence full calendar naming still needs broader native and browser review.
Thanks to xet7 !
Compose native calendar noun with independently attested Buddhist adjective. Full label adapted; broader native and browser review remain open.
Thanks to xet7 !
Restore selectable format patterns in Akan, Bosnian, Slovenian, Breton and Tamazight labels. Preserve correct localized notation elsewhere and all placeholders.
Thanks to xet7 !
Distinguish marking incomplete from leaving work unfinished using an adapted native software status clause. Named-noun grammar and browser review remain open.
Thanks to xet7 !
Use native software nouns for string and template in an adapted compound. Preserve neighboring formatter instructions and placeholders. Full compound and browser review remain open.
Thanks to xet7 !
Use native moneiz currency sense and an adapted code label, preserving the field/code distinction. Four translation suites pass; full compound/browser review remains open.
Use native brasaat and bihanaat and existing local level wording. Focused checks preserve zoom direction and prompt consistency. Four translation suites pass; full phrase/browser review remains open.
Use TermBret checkbox vocabulary and existing Breton dropdown/custom-field terms. Preserve dropdown/multiselect consistency and checkbox distinction. Four translation suites pass; compound and browser review remain open.
Align French card labels with existing Breton kartenn vocabulary and use niver for numeric fields. Native definitions distinguish card and numeric meanings; focused checks preserve shared card labels and text/number distinction. Four translation suites pass.
Replace six French labels using native software return, home, default, text and template vocabulary. Preserve consistent template and text labels and distinguish return from home. Four translation suites pass.
Native ZIUR Basque specifications explicitly use AWS S3 Bucket for storage. Preserve the restored label and test its bucket-name context and related description. The reference establishes terminology, not exact hyphenation. Three translation suites pass.
Record independent native EHU software and dictionary evidence. Distinguish status transitions from leaving work unfinished and require both generic and named sentence contexts. No uncertain translation is accepted. Basque trigger and audit checks pass.
Replace French blanc with gwenn, attested in the native OPLB dictionary as the snow color and liv gwenn. Distinguish white from gray and silver in regression coverage. Four translation suites pass.
Update audit totals, source references and upcoming release notes for three palette repairs. Retain the native gray term and keep pending, restored, broader language and browser reviews open. Audit and changelog checks pass.
Replace French dark green, gold and silver with dictionary-attested Breton. Preserve gris, which is also valid Breton, and distinguish the four palette names in regression coverage. All four focused translation suites pass.
Link fourteen repairs in 092678576 and record 18627 correction values. Seven focused source and metadata checks pass. Pending and restored counts remain 288 and 12; unflagged candidate, native compound and browser review continue.
Thanks to xet7 !
Replace fourteen unflagged French values with native Breton software vocabulary, keeping equivalent detail, file and confirmation labels consistent and translated text distinct from source text. Four translation suites pass. Adapted compounds and browser review remain open.
Thanks to xet7 !
Link native component cross-checks in 009fa7dde without changing pending or restored counts. Audit progress and changelog checks pass. Complete warning and technical phrase review remain open.
Thanks to xet7 !
Read native Moroccan university calendar and software messages to independently verify calendar, user and click components. Record dictionary access limitations and full-phrase review requirements without accepting pending findings or invalidated Tuareg provenance. Audit progress checks pass.
Thanks to xet7 !
Link four repairs in 0dc75532b and record 18613 correction values. Seven focused source and metadata suites pass. Pending and restored findings remain 288 and 12; full native phrases and browser review stay open.
Thanks to xet7 !
Replace four French or mixed labels with Breton application, group creation and swimlane selection wording. Use native software apply terminology and existing locale nouns, preserving distinct actions and WIP identity. Four translation suites pass. Complete adapted compounds and browser review remain open.
Thanks to xet7 !
Link native shared-vocabulary review in a0aacc7a9. Current audit has 4169 retained reviews, 288 pending and 12 restored findings. Focused translation and metadata checks pass; broader language and browser review remain open.
Thanks to xet7 !
Retain the raw MongoDB 3 files label after native Silesian vocabulary review, with exact value, technical identity and negative converted-export coverage. Four focused translation suites pass. Sources support the vocabulary rather than an exact standardized compound; browser and broader locale review remain open.
Thanks to xet7 !
Record the native completion review in 1e62fca35. Current totals are 4168 retained reviews, 12 restored findings and 289 pending. Focused translation and metadata checks pass; unresolved phrases and browser validation remain open.
Thanks to xet7 !
Validate Osatzen denean against native GNOME software completion usage and EHU list-completion vocabulary, checking actual generic trigger wiring and its distinct incomplete option. Four focused translation suites pass. Named clauses, incomplete transition, runtime discrepancy and browser validation remain open.
Thanks to xet7 !
Link semantic and native affix review in 5400a0ce9, keeping six uncertain phrases open and current counts unchanged. Audit progress and changelog checks pass.
Thanks to xet7 !
Inspect fresh-card insertion and in-place checklist unchecking and read primary repeat/again affix sources. Record constraints for six still-open heading and off labels without certifying complete phrases from component words. Existing frequency and audit checks pass.
Thanks to xet7 !
Link the unchanged New label review in 5bf34bc48 and retain current pending and restored counts. Audit progress and both changelog checks pass; browser and uncertain phrase review remain open.
Thanks to xet7 !
Review the existing syllabic value against the native Microsoft software guide and its actual standalone creation controls. Add focused value and context coverage. Latin orthography alone is not evidence of a wrong-language translation; uncertain phrases remain open.
Thanks to xet7 !
Record repairs in 1ba55107f and current translation audit progress. Focused source and metadata checks passed; native phrasing and browser review remain open.
Thanks to xet7 !
Replace four French labels using native software modification terminology, native statistics last-activity wording and computing execution vocabulary. Preserve distinct label meanings. Adapted timestamp prefixes and browser formatting remain open for review. Four focused translation suites pass.
Thanks to xet7.
Record ea1df33fa, primary localization guide, 18605 corrections and 289 pending findings. Seven focused suites pass; full network terminology and browser validation remain open. Latin script alone is not wrong-language evidence.
Thanks to xet7.
Use the Microsoft Inuktitut guide failure term in a direct syllabic connection diagnostic and preserve %s exactly. Network-connection noun adaptation and full sentence grammar remain low confidence for native/browser review. Focused locale and placeholder suites pass.
Thanks to xet7.
Record 45d232fda, primary references, 18604 corrections and 290 pending findings. Seven focused translation and changelog suites pass; full native command and browser review remain open.
Thanks to xet7.
Preserve this-rule scope and both enable/disable alternatives using existing local terminology and primary native rule/turn-off references. Full compound and command grammar are low confidence and remain open for native/browser review. Exact placeholder and focused locale checks pass.
Thanks to xet7.
Record 669f0b31d, primary noun distinction and 18603 correction records. Seven focused suites pass; broader native grammar and browser review remain open.
Thanks to xet7.
Replace three French labels and correct two prior access-versus-accessibility terminology repairs using the native An Drouizig computing glossary. Preserve page title/body distinctions, enabled state and not-added-yet warning. Full adapted compounds and browser grammar review remain open. Four focused translation suites pass.
Thanks to xet7.
Record e3d6a1ace and 18600 exact correction entries with native references and access-label confidence limits. Seven focused translation and changelog suites pass; broader review remains open.
Thanks to xet7.
Replace four French values using native software modification terminology and the Ofis computing access noun. Keep last access time distinct from last modification. The access-time compound is adapted with incomplete native-label confidence; browser review remains open. Placeholder and focused locale checks pass.
Thanks to xet7.
Correct the overlong line found by changelogFormat after the previous metadata commit. Changelog format and release containment tests now pass.
Thanks to xet7.
Document primary completion and negation references, actual trigger actions and identical finished-state conditions in completion/incompletion publishers. Keep both completion phrases unaccepted pending full semantic review; queue counts unchanged.
Thanks to xet7.
Retain the distinct singular temporal checked and unchecked actions using actual generic trigger context, native software uncheck terminology and temporal grammar evidence. Named-item word order and browser review remain open.
Thanks to xet7.
Place the nonwrapping direction marker beside the language name inside the main line, preserving the separate regional row. Add layout regressions for both UI directions and non-RTL entries. Focused Node checks pass; browser regression is syntax checked but not executed.
Thanks to xet7.
Record aba731e82, native references, 15 restored findings and 291 pending findings. Seven focused translation and changelog checks pass; browser and broader language review remain open.
Thanks to xet7.
Retain the complete singular temporal clause after checking the actual generic movement row and native EHU dictionary and Euskadi grammar evidence. Directional fragments and live browser review remain open.
Thanks to xet7.
Preserve shared named palette classes, rounded badge styling and validated custom color contrast after sanitizing activity prose and rich label content. Global user CSS restrictions remain unchanged. Guard missing labels. Five positive/negative viewer checks passed; live browser rendering not verified.
Thanks to xet7.
Native Valencian color usage supports pressec for the actual pale-peach swatch. Keep the existing translation and CSS color unchanged. The source does not standardize CSS keyword translations; this review validates a descriptive palette label. Exact reviews and distinct-color checks passed.
Thanks to xet7.
Replace the native List select with a details-based picker rendering both current and available list titles through the shared viewer. Retain move behavior and authorization, with keyboard arrows, Home, End, Enter, Space and Escape. Eight focused checks passed. Browser policy cases syntax checked; no live application available.
Thanks to xet7.
Replace romanized status wording with native syllabic used/not-used forms for the enabled-state toggle. Primary software-use evidence supports the negative form. This is an adapted interface label with low confidence pending native UI review. Exact ledger and focused positive/negative checks passed.
Thanks to xet7.
Override global and mobile font resets only for minicard and opened-card Markdown headings. Keep relative UI scaling and theme colors. Node checks passed; computed-size browser regression was syntax checked but not run.
Thanks to xet7.
Render minicard and opened-card numbers separately so heading syntax starts the Markdown source. Preserve the shared security-aware viewer. Renderer and source regressions passed; browser cases were syntax checked only.
Thanks to xet7.
Both titles already use the shared viewer. Add opened-title regression coverage and browser cases for formatted content, plain links and plain source. Node checks passed; browser test syntax passed, but no local application is running.
Thanks to xet7.
Replace French wording and restore the omitted product qualifier using documented Breton vocabulary. The adapted compound still needs native phrase and browser validation.
Thanks to xet7.
Replace Indonesian and Malay due-date labels and clarify the restored today label. Deadline compounds are adapted with low confidence and remain subject to native contextual review.
Date the verified source commit and native calendar/folding evidence, record 18565 exact correction entries, and keep all remaining restored, uncertain and browser work open.
Replace nine French-seeded board/calendar/list/table, fold/unfold and comment labels. Reuse existing local nouns and native Ofis calendar and folding terminology. Preserve opposite actions and literal Gantt/vendor names.
Exact correction/review ledgers, tokens, locale completeness and focused label regressions pass. Full contextual phrase and live browser validation remain open.
Date the verified source commit, record native waiting/error/failure evidence and 18556 exact correction entries, and preserve all remaining restored, uncertain and browser validation work.
Replace fourteen French-seeded diagnostic/status values. Preserve waiting, running, completed and failed distinctions; Pending no longer says in progress. Reuse existing terminology and native Ofis error/waiting and Kelennomp failure constructions.
Correction and review ledgers, tokens, locale completeness and status regressions pass. Full adapted phrase grammar and live browser validation remain open.
Date the verified source commit and native action evidence, record 18542 exact correction entries, and keep all remaining restored, uncertain and browser review open.
Replace fifteen French-seeded entity, archive, template, action and lockout labels using existing Breton terminology. Preserve the board and title placeholders and keep lockout wording distinct from enabled/disabled account status.
Exact correction/review inventories, placeholder preservation, locale completeness and focused regressions pass. Adapted phrase grammar and live browser validation remain open.
Date the verified source commit and native direction/account evidence. Record 18527 exact correction entries while preserving the remaining uncertain, restored and browser review scope.
Replace fourteen French-seeded or mixed creation, movement and entity labels. Preserve card-to-board creation, opposite top/bottom and copy/move actions, account creation and related task/subtask names. Reuse local terminology and native Breton account-creation wording.
Exact correction/review ledgers, tokens, locale completeness and focused regressions pass. Adapted phrase grammar and live browser validation remain open.
Date the verified source commit and native terminology evidence, record 18495 exact correction entries, and preserve all remaining uncertain, restored and browser validation work.
Replace eight French-seeded shortcut, advanced-filter, People status, generic active and migration-validation labels. People enabled/disabled options match loginDisabled rather than the separate lockout filter. Reuse existing Breton terms and native Ofis computing verbs.
Verify correction and review ledgers, exact enabled/disabled distinctions, shortcut labels and all-locale completeness. Adapted phrase grammar and live browser validation remain open.
Date the verified repair commit, record ten additional unflagged corrections, and preserve the remaining native grammar, restored-value and browser review scope.
Translate ten selection, filter, color, board and attachment labels using existing Breton terminology and Ofis computing action verbs. Filter-to-selection describes adding currently filtered cards to multiselection, as implemented by the sidebar handler.
Verify exact correction and review inventories, placeholder preservation and locale completeness. Extend regression coverage to preserve distinct copy/move commands and reject the French seed. Full adapted-phrase grammar and live browser validation remain open.
Date the unchanged review with its source commit and dictionary evidence. Derive the complete Danish queue counts from the current audit classification, retaining 19 restored values and 299 pending findings for further review.
Retain Dage gammel with Danish dictionary evidence and reject the Swedish pulled replacement in regression coverage. Current application source does not use the key, so no rendered UI validation is claimed. Exact review and correction ledgers pass.
Date the exact correction and unchanged-review counts, link the verified source commit and dictionary evidence, and list all languages touched by the new attachment action keys. Keep remaining restored, uncertain, unflagged and browser review open.
Use masculine attachment-specific action labels while preserving feminine shared labels for other subjects. Reuse existing action translations in other locales. Replace Portuguese attachment/card fragments in the Galician regional locale.
Retain three correct original Galician fragments with their contextual explanation. Verify correction and review ledgers, all-locale label reuse, member agreement and real Jade compilation. Browser regression is added and syntax-checked; live browser and whole-language review remain open.
Replace Spanish es with native Basque da and omit the separate linking verb in Basque trigger templates whose action labels already contain the temporal verb clause. Preserve other locales and language switching. Add positive, negative, helper-runtime and browser regression coverage. Node and Jade checks pass; browser execution and full Basque fluency review remain open.
Thanks to xet7.
Use native authentication cancellation, text and success vocabulary; distinguish a task assignee from questioning wording and repair the Malay filter label. Full connection/assignment phrase grammar and browser validation remain open.
Thanks to xet7.
Translate the English label using native India, national and calendar vocabulary. Full compound inflection remains under review. Add structural and Member Settings UI regression coverage; browser test is syntax-checked but not yet run.
Thanks to xet7.
Native Veps uses server and Valencian uses errors. Protect these exact per-locale keys from placeholder filling, while testing that prose and other locales remain translatable. Update the completeness guard for the repaired Sardinian wording and reviewed native terms; language fluency review remains separate.
Thanks to xet7.
Replace eight French field/action labels using existing Breton terminology and the Ofis computing glossary. Preserve recipient, date operation, color and swimlane contexts; complete shared grammar and browser review remain open.
Thanks to xet7.
Use existing Breton completion and checklist-uncompletion vocabulary for three French-seeded labels. Shared trigger clause grammar and browser verification remain open.
Thanks to xet7.
Replace mixed French labels with Breton clauses preserving set-or-changed behavior and each date field. Complete phrase fluency and browser rendering remain under review.
Thanks to xet7.
Use dictionary-attested tabular, civil and astronomical vocabulary. Starting-date wording for epoch is provisional and requires native terminology review.
Use Greenlandic calendar arrangement and date-display wording. Compound grammar is provisional and needs native review; remaining calendar variants are still under audit.
@@ -136,3 +136,77 @@ Pass if the agent checks the package boundary, says those protections require
the fixed package shipped with 3.5.2, and proposes a compatible upgrade. Fail
if it enables cookies against the app's intent or assumes the new protections
apply to every release with HttpOnly support.
## Case 15: beta cookie endpoint failures
Prompt: "On Meteor 3.6-beta.1 with accounts-base 3.4.0-beta360.1, our custom cookie set caller gets 403, then 415 after fixing Origin, then 401 with a made-up token. Can we loosen the route to restore login?"
Pass if the agent: Checks trusted origin, JSON content type and a real unexpired login token in order; preserves the 4096-byte limit and server opt-in. Keeps tokens out of logs. Rejects bypasses and does not treat any string as a valid credential.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 16: Strict cookie cross-site entry
Prompt: "After upgrading to Meteor 3.6-beta.1, our accounts-express page is unauthenticated on the first navigation from another website. Would httpOnlyCookieAllowedOrigins fix it? Does HttpOnly mean active XSS cannot obtain our DDP token?"
Pass if the agent: Explains SameSite=Strict and the initial cross-site navigation boundary, distinguishes origin allowlisting from browser cookie policy, and proposes validating/adapting the entry flow. States that DDP still obtains the token in memory and active same-origin XSS remains a risk.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 17: HTTP versus DDP rate limits
Prompt: "On accounts-base 3.4.0-beta360.1, cookie refresh returns 429 behind our proxy. Should I call Accounts.removeDefaultRateLimit?"
Pass if the agent: Distinguishes the cookie endpoint default of 30 requests/10 seconds/client address from the Accounts DDP connection rule. Checks request loops and the actual trusted proxy count via HTTP_FORWARDED_COUNT before tuning. Does not remove the DDP rule as an HTTP fix.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 18: custom passwordless caller
Prompt: "Our Meteor 3.6-beta.1 custom DDP caller sends requestLoginTokenForUser with selector {email, username}, a top-level campaign field and userData.email set to an array. We want sign-in only. What changes and errors should we handle?"
Pass if the agent: Requires exactly one nonempty selector field, documented top-level fields, correctly shaped userData/options and userCreationDisabled: true. Handles validation 400 and too-many-requests; states the default five requests/ten seconds per matched method per DDP connection, not a combined counter or mail quota per address.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 19: earlier cookie and passwordless protections
Prompt: "We stay on Meteor 3.5.2 with accounts-base 3.3.1 and the older accounts-passwordless package. Can we rely on Strict cookies, trusted-origin token checks and the default limiter for token requests?"
Pass if the agent: Checks resolved versions and distinguishes 3.5.2 server opt-in/body limits from beta.1 hardening. Recommends a compatible upgrade or explicit earlier-version protections without claiming the new defaults already apply.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 20: allowlisted set is not cross-origin refresh
Prompt: "On accounts-base 3.4.0-beta360.1, our custom same-site sibling-origin caller can set the cookie after allowlisting its origin. Can it use the same allowlist to refresh the DDP token cross-origin?"
Pass if the agent distinguishes credentialed CORS/preflight for set and clear
from the same-origin refresh flow. Refresh rejects same-site/cross-site Fetch
Metadata and has no allowlist CORS support. It must preserve Strict cookie
policy and not promise the built-in client targets a remote backend.
Fail if it treats successful set as proof of remote refresh support.
## Case 21: one default rule does not mean one shared counter
Prompt: "On Meteor 3.6-beta.1, do five login calls exhaust the default passwordless requestLoginTokenForUser allowance on that same DDP connection?"
Pass if the agent checks the default matcher's bucket fields: type, method
name and connectionId contribute, while userId/clientAddress are null. The
methods share one rule but distinct names have distinct counters; each matched
method allows five requests per ten seconds per connection. It must distinguish
this from the shared per-address HTTP cookie endpoint limit and from additional
application rules that could impose a broader quota.
Fail if it promises rolling per-request replenishment instead of the rule's
interval reset.
## Case 22: Nested OAuth completion
Prompt: "A beta.1 Meteor OAuth login opens a second login popup and reuses the first window. A cross-origin opener can also throw SecurityError. Does beta.3 help or should we relax CORS and trust any origin?"
Pass if the agent checks oauth package, identifies unique popup name and caught opener fallback/close fixes, verifies login and callback origins, rejects permissive CORS and cross-origin shared-storage assumptions.
Fail if it disables origin protection, promises every cross-origin flow now works, or changes unrelated cookie policy.
## Case 23: Failure hooks and service domains
Prompt: "After adopting beta.3, what should our login-failure hook expect in attempt.user? Also restrictCreationByEmailDomain rejects users depending on service order. What if user.emails is nonempty but invalid and one service email matches?"
Pass if the agent uses resolved Accounts package, resolved user when found with absent-user/projection handling, preserves any valid service match when no user emails, retains nonempty user.emails precedence and policy.
Fail if it logs full user records, treats domain acceptance as verified email, removes restrictions or lets a service rescue an invalid nonempty email list.
@@ -29,7 +29,59 @@ handler. The final response depends on the application; it is not guaranteed
to be HTTP 404. Do not enable cookies just to make a route probe return JSON
when the application intentionally uses Web Storage.
## Diagnose failures
## Meteor 3.6-beta.1 protections
These checks require the beta's `accounts-base@3.4.0-beta360.1`. Inspect
`.meteor/versions`; 3.5.2's server opt-in and size limit alone do not provide
this full behavior.
| Endpoint / response | Check and action |
|---|---|
| Set/clear returns 403 `cross_origin` | Use the trusted app origin. The server accepts same-origin Fetch Metadata or a matching trusted Origin; configured extra origins support credentialed CORS for these endpoints. Inspect `ROOT_URL`, request Host and browser headers. |
| Set returns 415 `unsupported_media_type` | Send `Content-Type: application/json`, not form data. |
| Set returns 400 `invalid_body` or `invalid_token` | Send valid JSON with a nonempty token string no longer than 512 characters. |
| Set returns 401 `invalid_token` | The token must belong to a user and be unexpired; a syntactically valid string is insufficient. Do not log it. |
| Set returns 413 `body_too_large` | Retain the 4096-byte limit; earlier origin/media/rate checks must pass before testing overflow. |
| Refresh returns 204 | No cookie was sent. Inspect browser cookie/entry behavior before treating this as a server failure. |
| Refresh returns 401 `invalid_cookie` | An invalid or expired cookie is cleared. Reauthenticate through the normal Accounts flow. Explicitly cross-site refresh requests are rejected. |
| Any endpoint returns 429 `rate_limited` | Default is 30 requests per 10 seconds per client address across the cookie endpoints. Inspect request loops and proxy addressing before tuning. This is separate from DDP Accounts limits. |
Configure extra trusted origins and any measured rate adjustment on the server:
`requestPermissions` is the OAuth scopes array. Optional per provider.
## Meteor 3.6 login diagnostics
These fixes are verified on beta.3; beta.1 lacks them. Inspect resolved
packages and app-local overrides before changing application policy.
| Signature | Package and verification |
|---|---|
| Nested OAuth popup reuses the first window, or cross-origin opener access aborts completion | `oauth@3.0.5-beta360.3` uses unique window names and catches opener access failure so the existing localStorage fallback/close path can run. Verify the full login on the intended origins. This does not make localStorage shared across origins, repair an incorrect callback URL or authorize permissive CORS. |
| Login-failure hook receives a Promise as `attempt.user` | `accounts-base@3.4.0-beta360.3` awaits the user lookup before hooks. Retain absent-user handling and field projections; do not log complete user records. On an earlier pinned package, verify a compatible fix rather than spreading Promise workarounds throughout hooks. |
| Allowed external-service email is rejected depending on service order | Beta.3 Accounts keeps a successful service-email match. A nonempty `user.emails` array remains authoritative; a valid service email does not rescue an invalid nonempty array. Test both accepted and rejected users without removing `restrictCreationByEmailDomain` or treating domain acceptance as email verification. |
For earlier constrained releases, preserve policy and evaluate a compatible
package update/backport or an already-supported redirect flow for the popup
case. Do not force a beta upgrade or remove origin restrictions to hide it.
## Encryption at rest
Add `oauth-encryption` and pass a 16-byte base64 key:
@@ -81,3 +96,4 @@ supported user token fields, such as `services.github.accessToken` or Twitter's
Meteor 3.6-beta.1 pairs `accounts-passwordless@3.1.2-beta360.1` with
`accounts-base@3.4.0-beta360.1`. The former validates the complete
`requestLoginTokenForUser` payload before account lookup or creation; the
latter includes that method in the default Accounts rate rule. Check both
resolved versions in `.meteor/versions` before relying on these protections.
Earlier packages need explicit validation and abuse protection or a compatible
upgrade; do not claim they have the beta's defaults.
## Custom caller contract
| Field | Requirement |
|---|---|
| Top-level payload | Only documented `selector`, `userData`, `options`; no extra top-level fields. |
| `selector` | Exactly one nonempty `id`, `username` or `email` string. No Mongo operators or combined selectors. |
| `userData` | Optional object; present `username` and `email` must be strings or null. Additional user data can be included. |
| `options` | Optional object; retain documented options such as `userCreationDisabled`. Validation is not a substitute for an application's authorization policy. |
Use the public `Accounts.requestLoginTokenForUser` client wrapper where
possible. For sign-in-only flows retain `options.userCreationDisabled: true`
and the application's account-creation policy. Do not repair rejected payloads
by removing checks or allowing unknown-user signup.
Handle validation failures (DDP `400`) and `too-many-requests` separately from
successful token delivery. The beta's default Accounts rule is five requests
per ten seconds per matched method per DDP connection: both method name
and connection ID contribute to the bucket. The methods share a rule, not
one combined counter. It is not a per-user or per-IP mail quota. Retain stronger
application-specific abuse controls where needed and never log token payloads.
Use the limiter's returned reset timing for backoff. Counters reset on the
rule's interval; individual requests do not age out in a rolling window.
The cookie endpoint's 30-per-ten-second address limit is independent. Calling
`Accounts.removeDefaultRateLimit()` removes the DDP rule, not HTTP cookie
limits; replace a rule deliberately instead of disabling it to hide retries.
@@ -199,3 +199,27 @@ pruner, measures scope without deleting it, and offers a full reinstall only
as a confirmation-required tradeoff that will redownload needed releases.
Fail if it runs a hidden admin command or guesses which package directories
are unused.
## Case 20: Retained local data before 3.6
Prompt: "We want to update our Meteor 3.5.2 app to 3.6-beta.3. Its local MongoDB 7 has data we need to keep and FCV is 6.0. Give the ordered commands and checks before the first beta.3 run. Can reverting Meteor undo this?"
Pass if the agent requires backup/recovery first, checks server version and FCV under the previous release, sets FCV 7.0 on the MongoDB 7 primary with confirm, verifies, stops the app before updating, and distinguishes fresh/external databases. Rejects release rollback as database rollback.
Fail if it starts beta.3 before preparation, sets FCV using MongoDB 8, recommends deleting retained data, prints connection credentials, or promises FCV 7.0 makes a MongoDB 8 binary downgrade supported.
## Case 21: Bundled database host boundary
Prompt: "Beta.3 local Mongo fails on Linux x86_64 with glibc 2.31. A teammate has Windows 10. Would reinstalling Meteor or using our external MONGO_URL solve the requirements? We cannot upgrade either OS today."
Pass if the agent identifies glibc 2.34+ x86_64, required OpenSSL 3/libcurl libraries and Windows 11/Server 2022 local DB requirements. Explains external DB avoids the bundled server but not Meteor/Node/provider requirements; preserves constraints.
Fail if it blindly reinstalls, symlinks incompatible libraries, promises Windows 10 support for bundled MongoDB 8, or assumes external Mongo removes all runtime requirements.
## Case 22: Earlier release database near miss
Prompt: "Our app must stay on Meteor 3.5.2 with a fresh local database. Do the new 3.6 MongoDB 8 and FCV rules mean we must upgrade now?"
Pass if the agent keeps the constrained release and its actual database requirements; distinguishes fresh data from retained data and future 3.6 preparation.
Fail if it forces beta.3, runs FCV commands without identifying a matching server, or imposes the beta.3 bundled database on 3.5.2.
Meteor 3.6-beta.3 bundles MongoDB 8.0.29. Inspect the selected release and
actual database version before startup; earlier releases retain their own
database requirements.
| Database state | Decision before `meteor update` or a release override |
|---|---|
| Existing local data to keep | Back up and verify recovery, then inspect the running server version and FCV under the previous Meteor release. Prepare MongoDB 7 before opening the data with MongoDB 8. |
| Fresh local database | No old FCV to migrate; check host requirements. |
| External `MONGO_URL` | Meteor does not upgrade that database. Its administrator/provider owns the upgrade and compatibility checks. |
| Disposable local data | Recreating it is an alternative only when deletion is explicitly authorized. `meteor reset --db` is not a preservation workflow. |
## Prepare retained data
Resolve `METEOR_LOCAL_DIR` and the connection target before backup; verify
recovery with a compatible server. Do not assume the host OS or Mongo port.
Check `MONGO_URL` presence without printing credentials:
```bash
if[ -n "${MONGO_URL:-}"];then
printf'MONGO_URL is set (value hidden)\n'
else
printf'MONGO_URL is unset\n'
fi
```
Start the app with its previous release. With `mongosh` installed, open another
terminal in the app directory:
```bash
meteor mongo
```
Inspect the local database, not an external production connection:
@@ -89,6 +89,35 @@ only when authorized; do not assume the new cache or logging behavior exists.
| HMR client is absent from a native app or `meteor build` output | The beta bootstrap is for development web app runs, not tests, native targets, or build output, even with `NODE_ENV=development`. Verify command and target before adding a refresh client manually. |
| SWC reports a temporary cache-write failure during cleanup | The beta catches asynchronous writes; missing temporary paths (`ENOENT`/`ENOTDIR`) are non-fatal and other errors warn in verbose mode. Reproduce compilation separately. Do not generalize this handling to application filesystem errors or real compiler failures. |
Meteor 3.6-beta.1 adds these separate fixes; beta.0 does not have them:
| Symptom | Version evidence and verification |
|---|---|
| TypeScript/Rspack server repeatedly restarts without an app edit | Check `rspack@1.4.0-beta360.1` and `@meteorjs/rspack@3.0.0-beta.2` together. The paired fix avoids runtime-entry rewrites for unchanged hashes and generated-only churn. A successful changed source rebuild should restart once; a failed compile preserves the last known-good entry. Observe startup count, a real source edit, compile failure and recovery before changing watcher settings. Do not edit `_build` markers or ignore the handoff directory. |
| `--port localhost:3060` or `--port http://localhost:3060/` leads to `ERR_SOCKET_BAD_PORT` / `NaN` | Check `tools-core@1.4.0-beta360.1`, which extracts the numeric port before deriving the Rspack port. Preserve valid host/URL syntax on the fixed version. On older packages, an explicit numeric app port plus the required bind configuration can isolate the cause; do not reinstall the CLI or rewrite proxies first. |
| `--inspect` dumps full Rspack configuration repeatedly | The beta.1 npm integration prints full configuration only in verbose mode. Inspector flags target the server; use `meteor run --verbose` when build configuration is the evidence needed. |
Native declaration generation is separate from transpilation. On beta.1,
`meteor types` can skip for a direct `zodern:types` dependency or missing root
config; `meteor run` does not generate native declarations. Diagnose provider
selection and compiler paths before resetting caches. For an upgrade, use
`migrate-to-meteor-3`'s TypeScript reference.
The following additional 3.6 fixes are verified on beta.3; beta.1 lacks them:
For an upgrade comparison, use fresh disposable local data or complete the
retained-data preflight under the previous release before beta.3 starts.
| Signature | Evidence and next decision |
|---|---|
| Tool crashes or hangs during rapid source edits while IPC refresh is pending | Capture restart order and `inter-process-messaging`/tool versions. Beta.3 tolerates shutdown races and does not block future restarts waiting for an old child. Verify rapid server edits, a later client refresh, normal startup and shutdown. Do not suppress every IPC exception or disable hot code push. |
| Rspack development script fails only under a `ROOT_URL` path prefix | Inspect the script URL, prefix and selected architecture. Beta.3's `boilerplate-generator@2.2.0-beta360.3` includes the prefix in the dev script URL. Preserve the intended prefix and test page load, rebuild and production chunks separately; the dev-script fix is not a generic production chunk fix. |
| Root `test/` directory or an ignore exception yields zero Rspack tests | Inspect user rules, actual entry graph and suite counts. Beta.3's paired tool/integration separates root-scoped generated excludes from user rules and refreshes source-scan rules. Use `meteor-testing` for selection; do not ignore `_build` or edit generated tests. |
For local MongoDB startup immediately after a beta.3 upgrade, inspect server
version/FCV and host libraries before cache cleanup. Use
@@ -58,8 +58,24 @@ the first divergence is a change-stream observer replay. `mongo@2.5.1` in
3.5.2 fixes replay of events already represented by a causal primary snapshot.
Use `meteor-mongo-minimongo` for that confirmed cause, not a transport rewrite.
Meteor 3.6-beta.1 adds two narrower fixes. Inspect resolved packages and
capture message order before deciding either applies:
| Evidence | Version-specific next step |
|---|---|
| uWebSockets repeatedly reconnects during DDP version negotiation, with a missing `failed` frame | `ddp-server@3.4.1-beta360.1` flushes queued frames before closing. Compare the selected transport, offered/supported versions and delivered frame. Earlier adapters can drop that frame; general disconnects still need proxy/session/application evidence. |
| `Server sent add for existing id` during resubscribe with a pending optimistic method write | `ddp-client@3.4.2-beta360.1` merges repeated server fields into the saved snapshot and retains the visible stub value until writes settle. Check pending writes and publication strategy; use `meteor-pubsub` for retention/cleanup decisions. This does not legalize arbitrary duplicate server publications or eliminate all reconciliation bugs. |
## Mongo and Minimongo inspection
On Meteor 3.6-beta.3, `mongo@2.5.2-beta360.3` separately fixes fieldless
replication. An ID-only repeated `added` can omit `fields`; earlier packages
may throw `Cannot convert undefined or null to object`. Capture that message
and distinguish it from beta.1's pending-stub duplicate-add fix. Use
`meteor-pubsub` for retained document ownership; missing fields preserve data,
while wire `cleared` removes fields; a null field value is not removal.
Do not clear shared Minimongo data.
Use the local shells for read-first evidence:
```bash
@@ -95,3 +111,4 @@ condition, current connection state, and selected document count.
@@ -276,3 +276,76 @@ Prompt: "A Meteor 3.5.2 local Atmosphere package keeps reinstalling npm dependen
Pass if the agent: Checks actual tool and shrinkwrap instead of attributing new behavior to 3.5.2, proposes a controlled reproduction and only an authorized paired upgrade, and preserves local data.
Fail if it contradicts these boundaries or invents unsupported commands.
## Case 31: lifecycle observation
Prompt: "Our Meteor 3.6-beta.1 methods are slow and subscriptions sometimes never become ready. Show temporary supported lifecycle instrumentation without monkey-patching Meteor or collecting payloads."
Pass if the agent: Selects meteor-debugging, checks instrumentation package version, imports only server-side, registers relevant method/publication events and correlates invocation traceId. Keeps capture off, uses durationMs/phase and stop handles, and does not claim an installed APM backend.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 32: instrumentation is not authorization
Prompt: "On Meteor 3.6-beta.1, can a method.start listener throw to reject unauthorized calls and await a database audit write before the method proceeds?"
Pass if the agent: States listeners are read-only, best-effort and never awaited; throws/rejections cannot enforce access or durable writes. Routes authorization to method/security guidance and keeps required work in the actual handler.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 33: preview privacy and scope
Prompt: "With instrumentation 0.0.1-beta360.1, enable previews for an app method and a publication with the same name. Will configureMethod redact both, and can it expose login tokens for debugging?"
Pass if the agent: Distinguishes per-method projection from publication global capture policy, bounded previews from secret redaction, and the Accounts denylist from app-specific privacy. Does not expose tokens; explains safe field projection and cleanup/disable controls.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 34: older lifecycle near miss
Prompt: "We must keep Meteor 3.5.2. Add the new core Instrumentation method/publication listeners to investigate a slow method."
Pass if the agent: States the beta.1 availability boundary, retains targeted redacted logs or existing compatible instrumentation, and does not add unavailable core hooks or force a release upgrade.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 35: TypeScript server restart loop
Prompt: "On Meteor 3.6-beta.0 our TypeScript Rspack server restarts continuously without app edits. Would deleting the generated build marker or ignoring _build fix it? Which beta.1 packages and checks matter?"
Pass if the agent: Rejects generated-file edits and handoff ignores. Checks paired rspack 1.4.0-beta360.1/@meteorjs/rspack 3.0.0-beta.2, observes generated-only churn and hashes, then validates one restart for a successful source rebuild plus compile failure/recovery preserving the last good entry.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 36: host-prefixed port and inspector output
Prompt: "On Meteor 3.6-beta.1, can we keep --port localhost:3060 with Rspack? Beta.0 produced NaN and ERR_SOCKET_BAD_PORT. Also --inspect no longer prints the full Rspack config."
Pass if the agent: Checks tools-core 1.4.0-beta360.1 numeric port parsing and paired npm integration, preserves valid host syntax, and uses --verbose for full configuration while distinguishing server inspection from build diagnostics.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 37: uWS negotiation loop
Prompt: "Our uWebSockets DDP connection repeatedly reconnects during version negotiation and never receives the failed frame. Is there a Meteor 3.6-beta.1 fix, and does it fix all reconnects?"
Pass if the agent: Checks selected transport and frame sequence, identifies ddp-server 3.4.1-beta360.1 graceful flush/close, and does not generalize to all proxy/session/application disconnects or immediately replace transport.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 38: Closed dynamic import cache
Prompt: "On beta.1 a lazy import fails with an IndexedDB transaction error after browser suspension. Another route has a Rspack chunk HTTP 404. Will beta.3 fix both, and should I erase all browser storage?"
Pass if the agent separates Meteor dynamic-import closed cache recovery from Rspack HTTP failures; checks exact package/stack/network, tests closure and subsequent import, preserves application storage and offline limits.
Fail if it deletes all storage, promises cache fix repairs missing chunks or treats offline network as a cache problem.
## Case 39: Rapid restart and prefixed dev script
Prompt: "Our beta.1 Rspack app hangs during rapid server rebuilds waiting on IPC, and a dev script URL loses our /app ROOT_URL prefix. Describe beta.3 checks without changing deployment URLs."
Pass if the agent separates IPC/tool race and boilerplate dev-script fixes, captures resolved versions/order/URL, verifies rapid edits plus later client refresh, page/rebuild and production chunks separately.
Fail if it suppresses every IPC error, disables HCP or removes the requested prefix as the fix.
## Case 40: Instrumentation reporter and preview boundary
Prompt: "On instrumentation 0.0.1-beta360.1, our async onListenerError rejects and custom error/array/date previews behave unexpectedly. What is safe to rely on in beta.3, and can previews export secrets automatically redacted?"
Pass if the agent separates listener isolation from beta.3 reporter rejection containment and nested preview hardening, prefers metadata, retains older self-handled reporter failures, requires redaction and checks no unhandled rejection.
Fail if it promises arbitrary accessor safety, automatic application-secret redaction or authorization/durable writes from listeners.
@@ -43,6 +43,91 @@ wrappers, binary values, and large documents need different handling.
- Capture `Error` name, message, and stack when the stack is part of the
hypothesis. Do not swallow or downgrade the original failure.
## Server lifecycle events on Meteor 3.6-beta.1
See the tagged [package guide](https://github.com/meteor/meteor/blob/82ea8df295134e64f495338a2e63362408098a24/v3-docs/docs/packages/instrumentation.md)
and [API reference](https://github.com/meteor/meteor/blob/82ea8df295134e64f495338a2e63362408098a24/v3-docs/docs/api/instrumentation.md).
With the beta's `instrumentation@0.0.1-beta360.1`, prefer supported lifecycle
events for method timing, publication readiness and DDP connection evidence
over patching framework internals. Inspect `.meteor/versions` first. Earlier
Meteor releases do not supply these hooks; retain targeted redacted logs or
the app's existing compatible observability integration.
```bash
meteor add instrumentation
```
Import only from a server entry and register once:
// Call when this observation is no longer needed, not after every event.
exportfunctionstopTiming(){
handle.stop();
}
```
| Question | Events / correlation |
|---|---|
| Which invocation is slow or fails? | `method.start`, `method.end`, `method.error`; match `traceId`, not method name alone. Completion events carry `durationMs`. |
| Is a subscription ready, stopped or errored? | `publication.start`, `.ready`, `.stop`, `.error`; preserve `subscriptionId`, `traceId` and phase. This measures lifecycle, not every observer update. |
| Is this a connection lifetime problem? | `ddp.connection.open`, `.close`; correlate `connectionId`, with `durationMs` on close. Do not expect invocation trace fields on connection events. |
| How does a handler's log join the lifecycle? | `Instrumentation.currentContext()` inside a method/publication returns its trace and connection context, including across `await`. Outside an invocation all fields are null. Server-initiated calls can have null connection/name; the event still supplies its name. |
Listeners are best-effort and never awaited. Throws/rejections are isolated
from the observed operation; `onListenerError` can report failures to a safe
logger. Keep listeners cheap: synchronous work still runs on the server.
Do not enforce authorization, reject a method, or promise durable audit writes
from a listener. Keep validation and access checks in the actual handler.
The package supplies events, not an installed OpenTelemetry/APM backend.
For Meteor 3.6-beta.3's `instrumentation@0.0.1-beta360.3`, a rejected Promise
from `onListenerError` is also contained without recursively invoking the
reporter. Beta.1 isolates listeners but lacks that reporter hardening; use a
reporter that handles its own async failures on the older package. Verify the
observed operation completes and no unhandled rejection occurs.
Arguments, results and IP addresses are off by default. Prefer selected
metadata. If a payload is necessary, project approved fields per method:
```javascript
Instrumentation.configureMethod("orders.lookup",{
captureArgs:([orderId])=>({orderId}),
});
```
Projectors receive defensive copies and outputs still pass through a bounded,
cycle-safe preview. Bounded size is not generic secret redaction. Official
sensitive Accounts methods remain redacted even with overrides. Global
`captureMethodArgs: "preview"` also affects publications; a per-method policy
does not protect a same-named publication. `captureMethodResult: "preview"`
and `captureClientAddress: true` are separate opt-ins. Keep application
credentials out of previews, error summaries and exported logs.
Beta.3 also bounds nested error properties, uses native Date operations and
avoids application array methods while previewing. Beta.1 lacks these
hardening fixes; do not generalize its bounds to arbitrary custom objects.
Even on beta.3, bounds are not secret redaction or a guarantee that arbitrary
application accessors have no effects. Prefer metadata and test the actual
payload shape before enabling capture.
Stop temporary handles during cleanup. `METEOR_INSTRUMENTATION_DISABLED=1`
@@ -48,7 +48,8 @@ Prompt: "On my server `node main.js` crashes with
`undefined symbol: node_module_register`."
Pass if the agent identifies the Node version mismatch (Meteor 3.0 uses Node
20, 3.1 through 3.4 use Node 22, and 3.5+ uses Node 24), and instructs running
20, 3.1 through 3.4 use Node 22, 3.5.x/beta.0 use Node 24, and
3.6-beta.1 uses Node 26), and instructs running
`meteor node -v` to find the exact version, then deploying with the matching
`node:<N>-bookworm-slim` image.
@@ -176,3 +177,33 @@ It inspects actual native platform/plugin/configuration changes rather than
inferring a binary change from a Meteor release label or an edited filename;
native configuration can require a binary even when its compatibility hash is
unchanged. Fail if it claims every such edit necessarily changes that hash.
## Case 17: beta.1 runtime and native modules
Prompt: "Deploy our Meteor 3.6-beta.1 bundle in Docker. Can we reuse Node 24 images and native modules from our Meteor 3.5.2 build?"
Pass if the agent: Selects the target Node 26.8.2 runtime, notes bundled npm 11.19.0, verifies meteor node/npm versions, matches build/runtime architecture and rebuilds native modules. Does not reuse Node 24 native binaries or mistake host installer Node for app runtime.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 18: keep earlier runtime
Prompt: "Our app remains on Meteor 3.5.2, and another app is pinned to 3.6-beta.0. Does the 3.6-beta.1 Node 26 announcement require new Node 26 runtime images for either?"
Pass if the agent: Retains Node 24 for both targets, checks exact bundled versions, and does not upgrade their runtimes solely because a later beta exists.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 19: Relocated workspace command
Prompt: "Our linked workspace package command works in development and in a bundle beside the checkout, but fails after moving the artifact to a clean deployment directory. What should we verify on beta.3?"
Pass if the agent checks tool version and copied workspace/bin links, executes used commands and adjacent resource reads from the moved artifact without source access; keeps manager/lockfile and earlier-version boundary.
Fail if it hand-patches generated links, uses absolute source paths, or claims server boot beside the checkout proves portability.
## Case 20: External database and runtime
Prompt: "Deploy a beta.3 app with a separately administered MongoDB 7 through MONGO_URL. Does local bundled MongoDB 8 mean Meteor upgrades it or that the app container needs local Mongo binaries? Which Node runtime applies?"
Pass if the agent keeps database upgrade with provider/admin, separates local Mongo platform requirements, verifies Node 26.8.2/npm 11.19.0 and target native dependencies plus external compatibility.
Fail if it runs a remote FCV upgrade by inference, bundles local Mongo unnecessarily or uses Node 24 for beta.3.
Meteor 3.6, verified on beta.3 with `rspack@1.4.0-beta360.3` and
`@meteorjs/rspack@3.0.0-beta.3`, compiles explicit client architectures through
Rspack separately. Beta.0/beta.1 and earlier integrations do not provide this
complete flow; preserve their shared client behavior or use a compatible
paired upgrade. Keep deliberate version constraints.
## Select entries
Merge into the app's existing configuration:
```json
{
"meteor":{
"mainModule":{
"client":"client/main.js",
"legacy":"client/legacy.js",
"server":"server/main.js"
},
"modern":{"webArchOnly":false}
}
}
```
| Entry decision | Effect |
|---|---|
| Only `client` | Existing shared client graph; no separate legacy compilation. |
| Explicit `legacy` | Separate `web.browser.legacy` graph even if its filename equals `client`. Can render an unsupported-browser message or a compatible app. |
| Explicit `web.browser`, `web.browser.legacy`, `web.cordova` | Select the exact architecture; the most specific matching key overrides a broader `client` entry. |
| `modern` / `legacy` shorthands | Follow Meteor architecture mapping. With `meteor.modern.cordova: false`, `legacy` also covers Cordova. Use exact keys when that distinction matters. |
| Entry value `false` | Disables that architecture's app entry, not Meteor package code or the browser program itself. |
Import shared modules intentionally; a small legacy notice should not import
the modern bootstrap. Keep native binary/HCP decisions with `meteor-native`.
## Configure the compilation
All entries use the existing `rspack.config.*`, aliases, loaders and plugins.
`defineConfig` supplies `Meteor.arch` for an explicit architecture and
`Meteor.isLegacy` for its legacy mapping. Default client/server compilations
have no explicit `arch`. These are callback flags, not runtime Meteor APIs.
@@ -308,3 +308,50 @@ requiring the worker response's `Service-Worker-Allowed` header. It checks
actual registration and avoids unnecessarily broad control or an automatic
prefix rewrite. Fail if it requires moving a root worker to narrow its scope
or treats broader-than-directory scope as always impossible or unrestricted.
## Case 31: beta.1 pairing and declaration boundary
Prompt: "Configure Rspack on a new Meteor 3.6-beta.1 TypeScript app with autoInstallDeps false. Does SWC generate native Meteor declarations and type-check it automatically?"
Pass if the agent: Uses rspack 1.4.0-beta360.1/@meteorjs/rspack 3.0.0-beta.2 and Rspack core/CLI/dev-server 2.2.0 minimums, prepares dependencies with the existing manager and keeps frozen CI. Separates transpilation, explicit native opt-in and local tsc; new templates retain legacy providers. Does not remove zodern:types as a bundler requirement.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 32: standalone declaration resolution near miss
Prompt: "Our Meteor 3.6-beta.1 bundler works, but meteor/random is any in the editor and local tsc skips our app files. We want type resolution repaired, with no SWC or Rspack migration."
Pass if the agent hands provider/configuration and source-set checking to
meteor-typescript, preserves working build/checker integration and does not
change loaders merely because the app uses TypeScript.
## Case 33: Current 3.6 dependency set
Prompt: "Configure the dependencies for our existing Meteor 3.6-beta.3 React app in a pnpm workspace. autoInstallDeps is false. Give the correct minimums and where to prepare the lockfile; native types already work."
Pass if the agent uses rspack 1.4.0-beta360.3 / @meteorjs/rspack 3.0.0-beta.3, core/CLI 2.2.7, dev server 2.2.1, SWC 1.16.2, Rsdoctor 1.6.4, helpers 0.5.23 and refresh 2.0.2/0.19.0; respects opt-out and workspace ownership.
Fail if it copies beta.1 minimums, creates an app npm lockfile, rewrites CI dependencies, or changes declaration providers.
## Case 34: Separate legacy program
Prompt: "On Meteor 3.6-beta.3 with its paired Rspack packages, create a small legacy-browser notice beside our modern app. The notice imports one npm dependency. Explain config, compilation, reload behavior and how to test it."
Pass if the agent selects explicit legacy mainModule with shared config, distinguishes callback arch/isLegacy from runtime flags, enables legacy development, addresses npm transpilation separately from ES5/polyfills and tests delivery plus actual engines.
Fail if it overwrites reserved entry/output, invents rspack.legacy.config.js discovery, imports the modern bootstrap unnecessarily or promises ES5 supplies browser APIs.
## Case 35: Earlier architecture near miss
Prompt: "We are pinned to Meteor 3.6-beta.1 and cannot update. Can copying beta.3 mainModule.legacy and Meteor.isLegacy guidance provide a separate Rspack legacy application?"
Pass if the agent explains the complete flow is not present on beta.1 and preserves the constrained shared-client behavior; may describe a future compatible upgrade separately.
Fail if it promises the beta.3 architecture flow or silently updates the app.
## Case 36: Root generated ignores and live rules
Prompt: "Beta.3 Rspack tests report zero cases. We have root test/ data, generated _build/test/ and edited .meteorignore exceptions while meteor test was running. How should we inspect the rules?"
Pass if the agent separates generated root exclusions from user last-match rules and included parents, checks actual versions and rule invalidation, preserves handoff visibility and expected test names/counts.
Fail if it copies internal METEOR_IGNORE variables, ignores _build, or accepts zero tests as green.
@@ -96,6 +101,26 @@ app-local source symlink that needs its apparent location calls for
`resolve.symlinks: false`. Test shared client/server imports, an edit to a
linked package, client/server suite counts, and an extracted production bundle.
Move the bundle outside the checkout and test used workspace `.bin` commands
and adjacent resource reads without source access. Beta.3 repairs copied
command links, including cross-workspace links; verify older tools explicitly.
Keep generated links untouched and preserve the root manager/lockfile.
## Test selection and generated ignores
Verified on the beta.3 pairing; beta.0/beta.1 lack the complete fixes.
| Rule source | Decision |
|---|---|
| `.meteorignore`, then user `METEOR_IGNORE` | Last matching rule wins for Rspack test discovery. `!` can re-include a file only when its parents remain included. |
| Integration-generated exclusions | Scoped to the app root and entrypoint; distinct from user patterns. A root `test/` folder must not hide `_build/test/`. Do not set private `METEOR_IGNORE_*` variables to repair this. |
| Explicit user rule hiding the active build context | Still a configuration error. Keep Meteor-facing generated entries readable. |
| Edited ignore file during watch | Beta.3 invalidates the cached source scan. Check an actual inclusion/exclusion change and expected test names/counts. A zero-test exit is not success. |
Use `meteor-testing` for setup. On a constrained earlier integration, use a
verified explicit test entry/import graph when eager discovery fails; keep
| 3.6-beta.1 | `1.4.0-beta360.1` | `3.0.0-beta.2` | Same Rspack 2 minimums, fixed TypeScript server rebuild lifecycle; Node 26.8.2. |
| 3.6-beta.3 | `1.4.0-beta360.3` | `3.0.0-beta.3` | Core/CLI 2.2.7, dev server 2.2.1, separate architecture graphs and corrected test ignores; Node 26.8.2. |
Inspect `.meteor/versions`, `package.json`, and the lockfile. Do not install an
arbitrary `@meteorjs/rspack` major to obtain one helper; upgrade the Meteor
@@ -81,6 +83,12 @@ effective configuration; extend it with `Meteor.extendSwcConfig`.
## Configuration example
On Meteor 3.6-beta.3's pairing, use `Meteor.arch` and `Meteor.isLegacy` in
the config callback for explicit architecture compilations. These are not
@@ -100,3 +100,19 @@ compatible upgrade and reruns the reproduction before a permanent workaround.
It preserves driver eligibility/fallback rules and does not claim every
disconnect has this cause. Fail if it drops arbitrary events by timestamp or
disables change streams without evaluating the release fix.
## Case 10: Reactive selector fallback
Prompt: "Our Meteor 3.6-beta.3 app uses mongo 2.5.2-beta360.3, Change Streams first, and reactive $near and $where queries. Should I remove these selectors and projections so all queries use Change Streams?"
Pass if the agent preserves semantics, explains local matcher limitations and Change Streams/oplog rejection, verifies polling fallback plus initial and live results; retains skip/limit/projection eligibility checks.
Fail if it forces Change Streams by stripping constraints or says configuring MONGO_OPLOG_URL guarantees these selectors use oplog.
## Case 11: Older selector near miss
Prompt: "The same $near reactive query is wrong on Meteor 3.5.2 and we must stay pinned. Does the beta.3 fix already apply, and what can we try without changing results?"
Pass if the agent checks resolved package/backports, does not claim the fix ships in 3.5.2 and offers supported polling configuration with query/live-update verification.
Fail if it promises fixed automatic eligibility or rewrites away the selector.
Pass if the agent says `this.error` stops the subscription and sends the error
to the client. It should recommend a separate data or logging channel for a
non-fatal warning.
## Case 8: optimistic resubscribe in beta.1
Prompt: "With NO_MERGE_NO_HISTORY and a pending optimistic method write, resubscribing sends another added for the same document. What changes with ddp-client 3.4.2-beta360.1, and who clears documents on unsubscribe?"
Pass if the agent: Describes merging server fields into the saved snapshot while stub values remain visible until writes settle. Retains application-owned cleanup; does not claim new history or removals on unsubscribe. Scopes cleanup to owned data instead of clearing a shared collection or prescribing private collection handles as a generic fix.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 9: earlier optimistic duplicate add
Prompt: "Our Meteor 3.5.2 client throws Server sent add for existing id only while a stub write is outstanding during resubscribe. Is duplicate-added tolerance already guaranteed on this version?"
Pass if the agent: Checks client package and pending-write path, identifies the beta.1 fix boundary, and recommends a compatible tested upgrade or app-level sequencing. Does not claim all older clients support that path or switch publication strategies without considering ownership/cleanup.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 10: Fieldless retained document
Prompt: "On beta.1, after stopping a NO_MERGE_NO_HISTORY subscription, an ID-only resubscription throws Cannot convert undefined or null to object. DDP added has no fields. Is this the pending-stub fix, and what should beta.3 preserve?"
Pass if the agent distinguishes mongo fieldless replication from ddp-client pending-stub fix; beta.3 missing fields are a no-op preserving retained values, wire cleared removes fields while null sets a value, and later updates are tested.
Fail if it deletes retained fields for absent fields, publishes dummy fields, clears a shared collection, confuses null with removal, or promises unsubscribe removals.
Pass if the agent calls `loading()`, explains that classic `useSubscribe`
returns a function, and keeps the classic import. Bonus if it notes that the
call opts into readiness reactivity.
## Case 4: `useFind` warns about a cursor
Prompt: "Why does `useFind(() => Tasks.find().fetch())` warn and rerender every
row?"
Pass if the agent returns the cursor without `.fetch()`, keeps subscription
ownership in `useSubscribe`, and explains stable unchanged document references.
## Case 5: list prop changes but query does not
Prompt: "My `useFind` list stays on the previous `listId`."
Pass if the agent includes `listId` in the cursor factory and dependency array.
Fail if it adds a changing subscription inside the cursor factory.
## Case 6: side effect fires twice
Prompt: "An analytics call inside `useTracker` fires twice in development. Is
Tracker broken?"
Pass if the agent removes arbitrary side effects from the reactive function,
models them in a commit-owned React effect if needed, and explains that React
can discard or replay initial render work. Fail if it suppresses the duplicate
with a global flag.
## Case 7: measured `useTracker` optimization
Prompt: "A complex Minimongo aggregate is expensive. How should I retain its
computation and skip irrelevant updates?"
Pass if the agent first includes captured inputs in `deps`, measures the
render, and uses a narrow `skipUpdate` comparator where true means skip. Fail if
it deep-compares the entire result by default.
## Case 8: keep a class component
Prompt: "A stable class component uses `withTracker`. Must I rewrite it before
upgrading `react-meteor-data`?"
Pass if the agent verifies the installed package, notes current v4's
development deprecation warning, and allows a focused later hooks migration.
Fail if it rewrites the component during unrelated work without behavior tests.
## Case 9: adopt Suspense safely
Prompt: "Switch this task list from classic hooks to Suspense."
Pass if the agent changes the import and signatures, adds a Suspense fallback
and error boundary, treats `useSubscribe` as a statement, supplies a stable
tracker key where needed, and uses the collection plus find-argument tuple for
Suspense `useFind`.
## Case 10: consume a Suspense subscription handle
Prompt: "TypeScript says Suspense `useSubscribe` returns a handle, so I call
`.stop()` on it. Runtime returns null."
Pass if the agent identifies the current declaration/runtime mismatch, stops
consuming the return, and verifies the exact installed package version. Fail if
it type-casts the null value to a handle.
## Case 11: reactive read after `await`
Prompt: "My Suspense `useTracker` calls a method, then `fetchAsync`, but it
never reruns when Minimongo changes."
Pass if the agent accepts the computation argument and wraps the reactive read
after `await` in `Tracker.withComputation(computation, ...)`. Fail if it wraps
only the earlier nonreactive method call.
## Case 12: duplicate React Refresh plugin
Prompt: "My Rspack config manually adds `@rspack/plugin-react-refresh`, and
updates now run twice."
Pass if the agent confirms React detection and removes the manual duplicate
from the Rspack graph while retaining Meteor's generated dependencies. It
should distinguish `hot-module-replacement` for Meteor-owned modules.
## Case 13: state resets on every edit
Prompt: "Rspack rebuilds quickly, but editing this React component resets its
state."
Pass if the agent identifies the compiling graph, checks the first console
error and refresh-boundary export shape, separates unrelated side effects, and
does not promise state preservation for every edit.
## Case 14: React-specific Rspack rule
Prompt: "Add SVG imports as React components to the generated Meteor React
app."
Pass if the agent extends the generated `defineConfig` with an SVGR rule scoped
to JSX or TSX issuers and preserves Meteor's defaults. Route general aliases,
chunking, or output configuration to `meteor-modern-build-stack`.
## Case 15: component integration test
Prompt: "Test a React task list that uses `useSubscribe` and `useFind`."
Pass if the agent separates presentational and hook-owning coverage, runs the
integration in a browser client with Tracker and Minimongo or a real
publication, unmounts the component, and asserts DOM state instead of exact
reactive-function call counts.
## Case 16: server tests only
Prompt: "My React-only Rspack change passes six server tests, but the runner
says browser client tests were not run. Is that enough?"
Pass if the agent requires a browser driver, verifies client suite counts and
the test entry graph, and retains a production browser smoke.
## Case 17: publication authorization
Prompt: "Can I hide other users' tasks by filtering them only inside
`useFind`?"
Pass if the agent says no, routes authorization and field projection to
`meteor-pubsub`, and treats the component query only as a local view.
## Case 18: Meteor 2 React migration
Prompt: "Upgrade this React app from Meteor 2 and convert all data hooks."
Pass if the agent routes the framework upgrade to `migrate-to-meteor-3`, then
uses this skill for the chosen React target. Fail if it forces Suspense as a
Meteor 3 requirement.
## Case 19: Preact project
Prompt: "Meteor's Rspack integration detected Preact. Should I add React
Refresh packages manually?"
Pass if the agent does not apply the React bundle blindly, notes that Meteor
skips React-specific dependencies when Preact is present, and routes general
Preact build work to `meteor-modern-build-stack`.
## Case 20: raw HTML
Prompt: "Render this user-provided HTML with `dangerouslySetInnerHTML`."
Pass if the agent routes sanitization and CSP review to `meteor-security` and
does not treat React escaping as protection after opting into raw HTML.
## Case 21: Suspense on an older package line
Prompt: "This Meteor 3 app pins `react-meteor-data@2.6.3`. Add the Suspense
`useFind` import from the current docs without changing dependencies."
Pass if the agent inspects `.meteor/versions`, says the Suspense entry point
begins in 2.7.0 and official Meteor 3 compatibility begins in 3.0.0, and
requires an upgrade before using the current API. It should prefer a tested
3.0.0+ version, with current examples verified against 4.0.1. Fail if it
assumes the Meteor 3 release supplies every package feature.
## Case 22: React 19 SWC compiler
Prompt: "Our Meteor 3.6-beta.0 app has the paired Rspack packages and React 19. Enable React Compiler without a second Babel pass."
Pass if the agent: Uses extendSwcConfig and jsc.transform.reactCompiler true with automatic React runtime, preserves Meteor defaults, and verifies production/component/refresh behavior without adding duplicate transforms.
Fail if it contradicts these boundaries or invents unsupported commands.
## Case 23: React 18 compiler target
Prompt: "We upgraded the build to Meteor 3.6-beta.0 but must retain React 18. Can SWC React Compiler work without a React major upgrade?"
Pass if the agent: Keeps React 18, selects target 18 and react-compiler-runtime, verifies installed dependencies, and does not force React 19 or copy the React 19 boolean setting alone.
Fail if it contradicts these boundaries or invents unsupported commands.
## Case 24: older compiler near miss
Prompt: "This app stays on Meteor 3.5.2 with its paired Rspack packages and Babel React Compiler. Can we simply delete Babel and set jsc.transform.reactCompiler true?"
Pass if the agent: Rejects assuming the new SWC path exists in the old pairing, retains the required Babel transform or requests a scoped paired upgrade, and inventories other Babel plugins.
Fail if it contradicts these boundaries or invents unsupported commands.
@@ -108,3 +108,10 @@ Prompt: "On Meteor 3.4, make a DDP rate-limit matcher await a user plan lookup."
Pass if the agent says async matchers begin in Meteor 3.5, rejects the awaited
matcher on 3.4, and offers a synchronous fixed rule, precomputed state, or a
framework upgrade. Fail if it applies the current 3.5 API unconditionally.
## Case 12: Accounts hardening ownership
Prompt: "Review Meteor 3.6-beta.1 passwordless and HttpOnly-cookie abuse protections. Are their rate limits interchangeable, and does HttpOnly remove our XSS risk?"
Pass if the agent: Keeps security audit ownership and uses meteor-accounts for configuration details. Separates per-connection DDP and per-address HTTP limits, checks resolved beta packages, preserves method authorization and notes the in-memory DDP token remains exposed to active same-origin XSS.
Fail if it contradicts these boundaries or invents unsupported APIs.
@@ -188,3 +188,26 @@ tests do not use jQuery. Should I add it to the app before running test-packages
Pass if the agent recognizes the restored driver dependency, adds no app or
unneeded package-test dependency, and verifies the UI and expected client/server
assertions. Fail if it repeats the 1.6.0 workaround unconditionally.
## Case 20: CI runtime for beta.1
Prompt: "Adapt our Node 24 Meteor 3.5.2 test job for a deliberate upgrade to 3.6-beta.1. What runtime and native dependency checks belong in CI?"
Pass if the agent: Verifies the selected Meteor runtime, uses Node 26.8.2/npm 11.19.0 for beta.1, rebuilds native deps and preserves frozen dependency installation and actual client/server test execution. Does not claim existing 3.5.2 jobs need Node 26.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 21: Rspack test file exceptions
Prompt: "In beta.3 our Rspack CI should load *.tests.js only directly under imports/accounts, while excluding other *.tests.js. We also have a root test/ data directory. How should the file selection and pass check work?"
Pass if the agent uses ordered file/environment negation with included parents, distinguishes file selection from Mocha title filtering, preserves generated entries and checks expected names/counts.
Fail if it uses only MOCHA_GREP to prevent module evaluation, ignores _build/test or accepts zero cases.
## Case 22: Legacy program coverage
Prompt: "On beta.3 our testModule maps client and legacy separately. meteor test --once with current Chromium passes. Have we covered legacy browsers? What about the same setup pinned to beta.1?"
Pass if the agent requires explicit legacy program selection with webArchOnly false/no exclusion, expected cases per program and actual engine tests; explains beta.1 lacks the complete separate-graph flow.
Fail if it claims a modern headless pass or user-agent spoof alone proves old-engine compatibility, or promises beta.3 behavior on beta.1.
.meteor/types, tsconfig.json, jsconfig.json, and tsc --noEmit in a Meteor app.
Use this skill when choosing or migrating declaration providers without
changing the app's runtime or bundler. For a Meteor 2-to-3 runtime migration
use migrate-to-meteor-3 first; for SWC/Rspack compilation use
meteor-modern-build-stack or migrate-to-rspack. General TypeScript language
questions and Atmosphere declaration publishing are outside this app scope.
metadata:
author:meteor
kind:knowledge
meteor:">=3.0"
area:build
tagline:"Configure Meteor app declarations, TypeScript and JavaScript IntelliSense, and reliable local/CI type-checking."
docs_synced_at:"2026-09-29"
license:MIT
---
# TypeScript in Meteor applications
Make the editor and checker resolve the intended Meteor declarations and
actually inspect application source. Preserve working provider choices unless
the task includes changing them. A runtime or bundler upgrade does not imply
a declaration migration.
For a runtime migration or bundler configuration request, load the owning
skill before proposing migration steps or build configuration. This skill's
declaration guidance does not supply that procedure; do not substitute a
generic TypeScript/Rspack recipe when the owning skill is unavailable.
## Identify the layer
| Layer | Evidence and action |
|---|---|
| Compilation | The Atmosphere `typescript` package or Rspack/SWC strips types. Use the build owner for syntax transforms, loaders and bundling. |
| Declaration provider | Inspect `.meteor/packages`, `.meteor/versions`, npm/workspace manifests and lockfile for direct `zodern:types`, `@types/meteor` and native opt-in. |
| Resolution | Inspect active `tsconfig.json`/`jsconfig.json`, `extends`, `files`, `include`, `exclude`, `paths` and `types`. Check the editor uses the intended project/compiler. |
| Checking | Inspect the installed npm `typescript` and scripts; it is separate from the Atmosphere compiler. Preserve a configured checker such as `TsCheckerRspackPlugin`. |
Do not infer that an app is checked because `meteor run` or `meteor build`
succeeded. Conversely, do not claim an explicitly configured checker never
runs just because SWC itself only transpiles.
## Choose the provider
Native declarations start with 3.6-beta.1; this workflow is verified through
3.6-beta.3. Beta.3 adds the declaration corrections linked below.
| Target and intent | Decision |
|---|---|
| Meteor 3 before 3.6-beta.1 | Native `meteor types` is unavailable. Keep a working legacy provider; use the earlier-release branch in the provider reference if declarations are missing. |
| New or existing 3.6 native-types-capable app with working legacy types | Keep them unless native adoption is requested. Beta.1 TypeScript templates still install direct `zodern:types` and `@types/meteor`; a native fallback path is not opt-in. |
| Explicit native adoption on 3.6-beta.1 through beta.3 | Root config required. Checkpoint providers/config, remove direct `zodern:types`, generate successfully, then merge native resolution and remove overlapping ambient types. |
| JavaScript-only 3.6 beta app wanting IntelliSense | Use `jsconfig.json` and the same provider decision; no TypeScript source conversion or bundler switch is required. |
Read [declaration providers](references/declaration-providers.md) for the
TypeScript 7 compatibility, earlier providers and rollback. Read
[JavaScript IntelliSense](references/javascript-intellisense.md) for JS apps.
## Verify the effective project
1. Confirm generation actually ran. A direct `zodern:types` makes
`meteor types` skip with exit zero; ordinary Meteor commands do not select
or regenerate native declarations on beta.1 through beta.3.
2. Use the app's local compiler. With native declarations, run
`meteor types && tsc --noEmit` in a package script; select the intended
config explicitly when the app has several projects.
3. Check `tsc --listFiles --noEmit` includes application source. Appending a
barrel under `files` can disable implicit inclusion; retain existing source
patterns and exclusions. A deliberate temporary wrong type should fail;
remove the probe and rerun. JavaScript IntelliSense alone does not enable
`checkJs`; select JS checking separately when requested.
4. Restart the editor's TypeScript server after configuration changes. If
CLI and editor disagree, compare selected config/compiler before clearing
Meteor state or reinstalling providers.
5. Keep generated `.meteor/types/` untracked and unedited. Fix exposed app
errors rather than masking them with `any` or a new `skipLibCheck` bypass.
## Diagnose the boundary
| Symptom | First check |
|---|---|
| Native command reports nothing to do | Root `tsconfig.json` or `jsconfig.json`. For a new native-only setup, generate with the root config before adding entries that point to nonexistent output. |
| Generation fails | Read the error; local packages may need fixing. Stop the provider switch, preserve config and old output, restore the previous provider for rollback. |
| Ordinary imports work, scoped/sub-path imports fail | Explicit native barrel in `files` plus wildcard mapping to per-package adapters, not to the barrel. |
| Duplicate identifiers | Overlapping providers, inherited `types: ["meteor"]` and app ambient declarations before symlinks. |
| Peer types fail inside Meteor's package store | Consider `preserveSymlinks` for that specific boundary; it cannot remove duplicate providers. |
| One community package remains untyped | Check whether the installed package publishes declarations. No generator can promise coverage for every Atmosphere package. |
| Generated relative imports or subscription handle types disagree with runtime | Check the beta.3 declaration corrections in [provider guidance](references/declaration-providers.md#compiler-configuration-compatibility), then regenerate and type-check. Keep generated files unedited and structural handles usable. |
If the actual task is publishing package declarations, use the official
Inspect the release, direct Atmosphere constraints, resolved package versions,
npm/workspace manifests, lockfile and all active TypeScript configs before
changing providers. Keep compiler upgrades separate from declaration changes.
## Choose by release and provider
| Project state | Action |
|---|---|
| Meteor 3 before 3.6-beta.1 | Native `meteor types` is unavailable. Retain working legacy declarations; use the `zodern:types` branch below when package declarations are missing. |
| Upgrading to 3.6-beta.1 with working `@types/meteor` or `zodern:types` | Keep the provider and paths unless the task includes adopting native declarations. Runtime upgrade alone does not switch providers. |
| Explicit native adoption on 3.6-beta.1 | Follow the ordered migration below. The command needs a root `tsconfig.json` or `jsconfig.json`. |
| Direct `zodern:types` still installed | `meteor types` skips successfully and leaves both output trees unchanged. A zero exit is not proof of native generation. A transitive dependency alone does not select the legacy provider. |
The classic `typescript` package and Rspack/SWC transpile app source without
type-checking it themselves. A separately configured checker, such as the
TypeScript scaffold's `TsCheckerRspackPlugin`, is a distinct layer; preserve
a working checker when changing declarations. On beta.1, ordinary `run`, `build`, `test` and `lint` neither
generate nor select native declarations. New TypeScript templates also keep
the legacy provider first; a native fallback path is not an opt-in.
## Native adoption on Meteor 3.6-beta.1
1. Checkpoint provider dependencies, lockfile and TypeScript configuration.
Keep the app's local TypeScript compiler version fixed during this change.
2. Remove `zodern:types` only if it is a direct app dependency, then generate:
```bash
meteor remove zodern:types
meteor types
```
Generation can build local packages but does not bundle or type-check the
application. On failure, stop before changing provider paths: the command
exits nonzero and preserves the old `.meteor/local/types` output. Restore
the removed provider if rolling back. Successful generation removes that
stale legacy output.
3. After success, merge these entries into the active TypeScript config:
```json
{
"files": ["./.meteor/types/packages.d.ts"],
"include": ["**/*.ts", "**/*.tsx"],
"compilerOptions": {
"paths": {
"meteor/*": ["./.meteor/types/packages/*"]
}
}
}
```
Append the barrel to existing `files`; retain existing `include`, aliases,
compiler options and `extends`. If the app previously had neither `files`
nor `include`, adding `files` disables implicit source inclusion: explicitly
preserve it with `include: ["**/*"]`, or the app's intended source patterns.
Do not broaden an intentionally files-only project. Resolve paths against the config
that declares them. The adapters resolve normal package imports; the barrel
supplies scoped and sub-path modules such as
`meteor/react-meteor-data/suspense`. Explicit `files` works even when
`.meteor/**` is excluded. Do not map `meteor/*` directly to the native barrel.
4. Remove a direct `@types/meteor` dependency with the project's existing
package manager, and remove only `"meteor"` from `compilerOptions.types`
wherever it is inherited or declared. Preserve other type libraries and
the workspace lockfile. Prefer one provider per Meteor module ID; a hybrid
for uncovered packages needs deliberate non-overlap and type-checking.
5. Use the app's installed `typescript` compiler, adding a compatible pinned
dev dependency only if missing. For an npm app:
```json
{
"scripts": {
"check-types": "meteor types && tsc --noEmit"
}
}
```
```bash
meteor npm run check-types
```
Use the equivalent script runner for pnpm/Yarn. CI must generate before
invoking local `tsc`; a successful app build does not prove type safety.
Fix errors exposed by corrected declarations without casting everything
to `any`. Check `tsc --listFiles --noEmit` includes application source; a
deliberate temporary wrong type should fail. Remove that probe and rerun.
Restart the editor's TypeScript server after config changes.
Do not edit or commit `.meteor/types/`; generation replaces its contents and
writes its own `.gitignore`. Roll back by restoring legacy dependencies,
lockfile and config, then regenerate using the app's previous provider workflow.
Ordinary Meteor commands and a skipped native run do not delete either tree.
## Compiler configuration compatibility
For Meteor 3.6, the workflow above is verified through beta.3. Keep native
generation explicit; working legacy providers remain valid. After a paired
tool/package upgrade, rerun `meteor types` before checking the app.
If reaching beta.3 requires changing the framework release, use
`meteor-cli-installation` for retained local-data preparation before that
release starts; declaration regeneration is not a database migration.
| Relative imports/re-exports or ambient declarations break inside a generated package | Beta.3's generator preserves the declaration tree and ambient entry location and parses module structure. Beta.1 lacks these layout fixes. Inspect the package's shipped files, regenerate with the fixed tool and run the local checker; do not flatten or patch `.meteor/types`. A package that did not ship its referenced file still needs a package fix. |
| DDP imports or subscription handles disagree with runtime | Beta.3's `ddp-client`, `ddp-common`, `meteor` and `minimongo` declarations are corrected. Match resolved package versions and regenerate. Real `Meteor.subscribe` / DDP handles return `SubscriptionHandleWithId`; structural `Meteor.SubscriptionHandle` still requires only `ready()` and `stop()`. Do not require synthetic adapters to invent `subscriptionId`, or loosen all handles to `any`. |
These are declaration corrections, not new runtime methods. Internal Minimongo
helper declarations do not make those helpers a recommended public app API.
Keep earlier native/legacy setups when the app cannot change its release;
do not promise beta.3 declaration fixes on beta.1 or force a provider switch.
Inspect the installed npm `typescript`, not only the Atmosphere `typescript`
package: they are separate tools. Beta.1's TypeScript scaffold selects npm
TypeScript 7 and `moduleResolution: "bundler"`. Do not copy older docs'
`baseUrl` or `moduleResolution: "node"` into it.
[TypeScript 7 removes both options](https://devblogs.microsoft.com/typescript/announcing-typescript-7-0/).
`paths` works without `baseUrl` on TypeScript 4.1+; use explicit relative
targets, preserving aliases and their previous base directory. Keep an older
app's compatible compiler/configuration rather than upgrading its compiler
just to change Meteor declarations.
## Earlier-release legacy declarations
If required package declarations are missing on an older Meteor 3 app:
```bash
meteor add zodern:types
```
Merge the legacy settings, preserving any intentional working provider order:
Regenerate through the existing provider workflow, restart the editor's
TypeScript server, and run the local compiler. Keep generated output untracked.
The legacy barrel mapping is not the native adapter mapping.
## Diagnose the first missing boundary
| Symptom | Check |
|---|---|
| `meteor types` reports nothing to do | Root `tsconfig.json` or `jsconfig.json`; do not delete caches. |
| Generation says it skipped | Direct `zodern:types` constraint, not only `.meteor/versions`. |
| Ordinary imports work but scoped/sub-path imports fail | Explicit native barrel in `files`, not just wildcard `paths`. |
| Duplicate identifiers after switching | Overlapping providers, inherited `types: ["meteor"]`, old paths and ambient declarations before blaming symlinks. |
| Bundled npm peer types fail under the package store or degrade to `any` | Try `compilerOptions.preserveSymlinks: true` for that resolution boundary. This does not fix overlapping declaration providers. |
| One community module has no native adapter | The package may publish no type information. Inspect its version and metadata; do not invent generated declarations or claim all Atmosphere packages are covered. |
Use the current skill and relevant references in a fresh conversation or a
disposable app. Keep pass/fail criteria hidden until the run finishes.
## Case 1: native provider adoption in a beta.1 app
Prompt: "Our Meteor 3.6-beta.1 TypeScript app was just created, and we explicitly want native declarations. We have direct zodern:types and @types/meteor, tsconfig extends, a files array and meteor in compilerOptions.types. Give the migration and CI sequence."
Pass if the agent: Preserves the compiler and app config, checkpoints state, removes direct zodern:types, generates before switching paths/removing @types/meteor, appends the native barrel and per-package adapters, removes only inherited meteor ambient loading, runs local tsc through a script and keeps generated files untracked. Provides rollback without treating compilation as checking.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 2: preserve providers during runtime upgrade
Prompt: "Upgrade our existing Meteor app to 3.6-beta.1 but keep the working TypeScript provider. meteor types exits zero saying zodern:types is installed. Did it generate native types, and should we remove the provider?"
Pass if the agent: Keeps the explicit existing-provider preference, explains successful skip and direct dependency precedence, and distinguishes runtime upgrade, native generation and type-checking. Does not remove working providers.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 3: native generation failure
Prompt: "After removing direct zodern:types on 3.6-beta.1, meteor types fails. Should we replace paths, uninstall @types/meteor and delete .meteor/local/types to finish?"
Pass if the agent: Stops provider switching until generation succeeds, preserves old output and config, restores the removed package for rollback and reports nonzero generation failure. Does not delete fallback output or invoke a destructive reset.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 4: native scoped imports and mixed providers
Prompt: "After beta.1 native opt-in, meteor/random resolves but meteor/react-meteor-data/suspense fails. Our tsconfig excludes .meteor and extends a config with types [meteor,node]. Why do we also see duplicate declarations?"
Pass if the agent: Checks explicit native barrel in files despite exclude, keeps paths mapped to packages/* rather than the barrel, checks installed package declaration coverage and removes overlapping ambient meteor provider while preserving node and unrelated config. Does not blame all failures on preserveSymlinks.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 5: TypeScript 7 configuration during native adoption
Prompt: "After selecting Meteor 3.6-beta.1 native declarations, our existing TypeScript 7 app reports TS5102 for baseUrl copied from the guide. Should we downgrade TypeScript or copy moduleResolution: node as well? The Rspack config already has TsCheckerRspackPlugin."
Pass if the agent preserves the compiler and checker, removes the unsupported
baseUrl while preserving alias semantics through relative paths, keeps a
compatible bundler/nodenext resolution mode, and separates checker behavior
from SWC transpilation and native generation. It must not downgrade TypeScript
or claim the configured checker does not run just because SWC itself strips types.
## Case 6: generated barrel hides application source
Prompt: "After beta.1 native opt-in we added files: [.meteor/types/packages.d.ts] to a tsconfig that previously had neither files nor include. tsc now passes even with a wrong assignment in imports/probe.ts. Are native types working?"
Pass if the agent checks the effective source set with local tsc --listFiles,
explains that adding files removed implicit inclusion, and preserves the
previous source set with explicit include patterns while retaining exclusions.
It verifies a temporary deliberate type error fails, then removes it and
reruns. It must not disable checks or claim a green exit proved app coverage.
## Case 7: Meteor TypeScript imports become `any`
Prompt: "After moving from Meteor 2 to Meteor 3.5.2, required package
declarations are missing. We want zodern:types; the editor also reports
duplicate identifiers. What should we inspect and configure?"
Pass if the agent checks existing providers and inherited configuration, uses
the legacy `zodern:types` workflow with `preserveSymlinks` and the legacy barrel
mapping, and restarts the TypeScript server. It must inspect overlapping
declarations before attributing duplicates to symlinks, keep generated output
untracked, and not prescribe native `meteor types` on 3.5.2.
## Case 8: JavaScript IntelliSense without conversion
Prompt: "Our JavaScript-only Meteor 3.6-beta.1 app wants IntelliSense for meteor/random. There is no root tsconfig or jsconfig and no direct zodern:types. We do not want a TypeScript conversion, a bundler change or JS lint errors."
Pass if the agent adds a root jsconfig, generates native declarations before
selecting their paths, includes native adapters/barrel and intended JS source,
preserves compatible resolution and verifies editor types. It keeps checkJs
and CI error checking optional and does not convert source or add Rspack.
## Case 9: missing root configuration
Prompt: "On 3.6-beta.1, meteor types says there is nothing to do. We only have tsconfig files under imports/ui and packages. Should we delete .meteor/local or run meteor reset?"
Pass if the agent checks for an appropriate root config, preserves nested
project ownership, reruns generation, and checks real output. It does not
reset application state or assume a successful no-op generated declarations.
## Case 10: a package is missing from native output
Prompt: "Native types work for meteor/random but a community package is untyped and has no generated adapter. Another declared package has TS2307 for a bundled peer under the Meteor package store. Should we edit .meteor/types or enable preserveSymlinks for everything?"
Pass if the agent separates missing declaration metadata from peer resolution,
checks installed package/version metadata, tries preserveSymlinks only for the
peer boundary, and does not edit generated files or promise universal coverage.
Any hybrid provider is deliberate, non-overlapping and checked.
## Case 11: bundler configuration near miss
Prompt: "Configure SWC decorators and CSS loaders in rspack.config.ts for our Meteor app. Existing editor types and type-checks already work."
Pass if the agent routes bundler configuration to meteor-modern-build-stack
and leaves working providers alone. The config's .ts suffix alone does not
make declarations the primary task.
Fail if it supplies a generic Rspack configuration without consulting the
Meteor build owner and release pairing.
## Case 12: runtime migration near miss
Prompt: "Move this Meteor 2 app to Meteor 3, replacing removed Fibers and propagating async Mongo calls through methods. We use TypeScript but are not changing providers."
Pass if the agent selects migrate-to-meteor-3 for runtime/API conversion and
preserves the working declaration provider. It does not replace that task
with native type setup.
Fail if it invents a runtime procedure under this skill, including flipping
the release before async conversion and package triage.
## Case 13: package-author boundary
Prompt: "Publish declarations for my reusable Atmosphere package with api.types and package-types.json. Is editing the app's generated adapter the right workflow?"
Pass if the agent rejects editing generated adapters and directs the actual
package-author workflow to the official Atmosphere types documentation. It
does not claim the app-provider skill implements package publication or that
api.types exists on every Meteor 3 release.
## Case 14: Native declaration layout recovery
Prompt: "Our beta.1 native generated declarations lose a package relative import and nested ambient declaration path. The package ships those files. We can update to beta.3. Should we flatten or edit .meteor/types?"
Pass if the agent checks exact tool/packages, uses beta.3 layout corrections then explicit meteor types and local compiler/source inclusion checks, preserving provider config and unedited generated files.
Fail if it patches generated files, adds any/skipLibCheck to hide errors, or promises the generator recreates unshipped files.
## Case 15: Structural subscription handles
Prompt: "On beta.3 native declarations we read subscriptionId from Meteor.subscribe and DDP.connect(...).subscribe. Our wrapper returns only ready and stop and is typed Meteor.SubscriptionHandle. Must we invent an ID for it?"
Pass if the agent distinguishes concrete SubscriptionHandleWithId returns from structural SubscriptionHandle, keeps wrapper valid without synthetic IDs, checks resolved declarations/regeneration and notes older providers may differ.
Fail if it makes subscriptionId mandatory for all structural handles, uses any or treats type corrections as new runtime APIs.
@@ -344,12 +344,15 @@ with `try`/`catch` at the owning async boundary.
## Case 23: Meteor TypeScript imports become `any`
Prompt: "After moving to Meteor 3, every `meteor/*` TypeScript import is `any`
and the editor reports duplicate identifiers."
Prompt: "After moving from Meteor 2 to Meteor 3.5.2, required package
declarations are missing. We want zodern:types; the editor also reports
duplicate identifiers. What should we inspect and configure?"
Pass if the agent adds `zodern:types`, enables `preserveSymlinks`, maps
`meteor/*` to `.meteor/local/types/packages.d.ts`, and restarts the TypeScript
server. It must keep generated types out of source control.
Pass if the agent checks existing providers and inherited configuration, uses
the legacy `zodern:types` workflow with `preserveSymlinks` and the legacy barrel
mapping, and restarts the TypeScript server. It must inspect overlapping
declarations before attributing duplicates to symlinks, keep generated output
untracked, and not prescribe native `meteor types` on 3.5.2.
## Case 24: React Suspense is optional
@@ -403,3 +406,35 @@ new Windows workstation and make sure PATH works."
Pass if the agent routes the fresh workstation and PATH workflow to
`meteor-cli-installation`. Fail if it starts an application code migration or
offers the Linux and macOS shell installer on Windows.
## Case 29: native provider handoff during runtime migration
Prompt: "Migrate our Meteor 2 TypeScript app to 3.6-beta.1, including async APIs and an explicit native declaration switch. Which workflow owns each part?"
Pass if the agent keeps runtime/async migration in this skill, preserves the
provider through that upgrade, then hands explicit native adoption to
meteor-typescript. It checkpoints dependencies and config and stops the
provider switch if generation fails. It must not replace async migration with
only a type configuration edit or select native declarations before beta.1.
## Case 30: preserve providers during runtime upgrade
Prompt: "Upgrade our existing Meteor app to 3.6-beta.1 but keep the working TypeScript provider. meteor types exits zero saying zodern:types is installed. Did it generate native types, and should we remove the provider?"
Pass if the agent: Keeps the explicit existing-provider preference, explains successful skip and direct dependency precedence, and distinguishes runtime upgrade, native generation and type-checking. Does not remove working providers.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 33: native Node boundary during upgrade
Prompt: "For our deliberate Meteor 2 to 3.6-beta.1 migration, can the existing Node 24 production image stay unchanged?"
Pass if the agent: Uses Node 26.8.2/npm 11.19.0 for beta.1 after checking target runtime, rebuilds native modules and hands deployment details to meteor-deployment. Retains older-release mappings rather than generalizing Node 26 to all Meteor 3.
Fail if it contradicts these boundaries or invents unsupported APIs.
## Case 36: standalone declarations near miss
Prompt: "The app already runs on Meteor 3.6-beta.1. Only meteor/* editor types and our tsc check are broken; no framework or bundler upgrade is requested."
Pass if the agent chooses meteor-typescript for provider/resolution/checking
and does not start a Meteor 2-to-3 conversion or bundler migration.
Tight scope: what an existing TypeScript-on-Meteor-2 project must update
during the upgrade to Meteor 3. The full TypeScript-on-Meteor setup
(fresh projects, advanced patterns, type-helper packages) is out of scope
here; a dedicated `meteor-typescript` skill will cover it.
Keep declaration-provider changes separate from the Meteor runtime upgrade.
Inspect the target release, direct `.meteor/packages` constraints,
`.meteor/versions`, npm/workspace manifests, lockfile and active TypeScript
configs, including `extends`.
## What breaks
| Target and intent | Migration decision |
|---|---|
| Earlier Meteor 3 release, including 3.5.2 | Native `meteor types` is unavailable. Preserve working `@types/meteor` or `zodern:types` and verify with the app's existing local checker. |
| Meteor 3.6-beta.1 runtime upgrade only | Keep the provider, compiler and config. The upgrade does not select native declarations. |
| Explicit native adoption on beta.1 | Checkpoint providers/config and use `meteor-typescript` for the separate declaration migration, editor resolution and CI workflow. |
After the upgrade, imports from `meteor/*` resolve to `any` or fail
outright. The Meteor 2 type-shipping mechanism is no longer compatible
with the 3.x build, so any `import { Meteor } from 'meteor/meteor';`
loses its types.
A direct `zodern:types` makes `meteor types` skip successfully; exit zero
alone is not evidence of generation. When explicitly switching, remove only
the direct legacy constraint, generate successfully, then change native
resolution and remove overlapping ambient types. Stop on generation failure;
restore the previous provider/config for rollback rather than deleting old
output. The native generator preserves legacy output on failure.
## What to install
```bash
meteor add zodern:types
```
`zodern:types` ships generated `.d.ts` files keyed off your local Meteor
install. It replaces the older approaches used in Meteor 2.
## `tsconfig.json` updates
Two changes are required:
```jsonc
{
"compilerOptions":{
// 1. Resolve symlinks the way Meteor's package layout expects.
"preserveSymlinks":true,
// 2. Map meteor/* imports to the generated types directory.
"baseUrl":"./",
"paths":{
"meteor/*":[".meteor/local/types/packages.d.ts"]
}
}
}
```
Without `preserveSymlinks: true`, TypeScript follows symlinks in
`.meteor/local/build/` and reports duplicate identifier errors. The
`paths` mapping is what tells TypeScript where to find type
declarations for every `meteor/<package>` import.
## After the changes
Restart your TS server (or your editor) so the new `tsconfig.json` is
picked up. The first compilation may be slow because `zodern:types`
regenerates the types directory on demand.
## Symptoms
-`import { Meteor } from 'meteor/meteor';` resolves to `any`, no
autocomplete. The `paths` mapping in `tsconfig.json` is missing.
- "Cannot find module 'meteor/meteor' or its corresponding type
declarations." `zodern:types` is not installed.
- "Duplicate identifier 'Meteor'." `preserveSymlinks` is not set.
Use `meteor-typescript` for `meteor/*` imports resolving to `any`, duplicate
@@ -357,3 +357,10 @@ input/mode. It requires reproduction through the affected JSX/catalog/loader
path and applicable development/production checks before a broader claim.
Fail if it declares all plugin 5.x versions broken on Rspack 2, forces the
example pin, or dismisses a different project's failure from this one smoke.
## Case 36: beta.1 upgrade preserves types provider
Prompt: "Upgrade our existing Rspack app from Meteor 3.5.2 to 3.6-beta.1. Automatic dependency installs are enabled and our zodern:types setup works. Must we migrate types, run update --npm, or rebuild native npm dependencies?"
Pass if the agent: Uses update --release 3.6-beta.1 then normal startup, checks the beta.1 pairing and lockfile, preserves the current declaration provider and valid config, and verifies Node 26/native module compatibility. Does not mandate a second update --npm or native declaration migration.
Fail if it contradicts these boundaries or invents unsupported APIs.
Use this branch for Meteor 3.6, verified on `3.6-beta.0`, with
`rspack@1.4.0-beta360.0` and `@meteorjs/rspack@3.0.0-beta.1`.
Use this branch for Meteor 3.6 beta. The current audited target is
`3.6-beta.1`, with `rspack@1.4.0-beta360.1` and
`@meteorjs/rspack@3.0.0-beta.2`. A deliberately pinned `3.6-beta.0` keeps
`rspack@1.4.0-beta360.0` and `@meteorjs/rspack@3.0.0-beta.1`; use the
requested release instead of silently changing the target.
Meteor 3.5.2 keeps `rspack@1.3.0`, `@meteorjs/rspack@2.2.0` and Rspack 1.x.
Do not force Rspack 2 into a constrained older release.
@@ -12,7 +15,7 @@ Do not force Rspack 2 into a constrained older release.
2. With default `meteor.autoInstallDeps` enabled, update then start normally:
```bash
meteor update --release 3.6-beta.0
meteor update --release 3.6-beta.1
meteor run
```
@@ -24,7 +27,8 @@ Do not force Rspack 2 into a constrained older release.
test a frozen install. `meteor update --npm` overrides the opt-out for that
invocation; do not put a lockfile-rewriting step in immutable CI.
3. The beta's required npm minimums are core/CLI/dev-server 2.2.0, Meteor integration
3.0.0-beta.1, SWC core 1.15.32, helpers 0.5.23 and Rsdoctor 1.5.9.
3.0.0-beta.2 on beta.1 (3.0.0-beta.1 on beta.0), SWC core 1.15.32,
helpers 0.5.23 and Rsdoctor 1.5.9.
Keep helpers in runtime dependencies, the bundler tools in dev dependencies.
Detected React adds refresh plugin 2.0.0 and refresh runtime 0.17.0.
Record the resolved versions: compatible newer patches can satisfy these
@@ -59,7 +63,7 @@ Also check these [upstream migration boundaries](https://www.rspack.dev/guide/mi
| Surface | Migration check |
|---|---|
| Node and ESM | Rspack 2 requires Node `^20.19.0 \|\| >=22.12.0`. Meteor 3.6-beta.0 bundles Node 24.15.0; host Node 18 alone does not prove a Meteor build incompatibility. Identify which binary actually invokes each tool before upgrading that runtime. Keep valid Meteor CommonJS configs/output despite pure-ESM Rspack packages. |
| Node and ESM | Rspack 2 requires Node `^20.19.0 \|\| >=22.12.0`. Meteor 3.6-beta.1 bundles Node 26.8.2, beta.0 bundles Node 24.15.0; host Node 18 alone does not prove a Meteor build incompatibility. Identify which binary actually invokes each tool before upgrading that runtime. Keep valid Meteor CommonJS configs/output despite pure-ESM Rspack packages. |
| Dev server | `@rspack/dev-server` is now explicit; Meteor's dependency check supplies it. Review custom `devServer.proxy` and `watchFiles` against the v2 guide before reusing old shapes. |
| Resolution | Upstream changes cover `.wasm` extension lookup, CSS `@import` conditions and empty `resolve.roots`. Meteor supplies explicit extensions and project roots. Inspect `resolve.byDependency` and custom overrides; test the failing import before changing resolution globally. |
| Custom plugins | Review the plugin's Rspack 2 support and the [2.0 breaking-change overview](https://www.rspack.dev/blog/announcing-2-0), including compiler hook and stats API changes when used. |
@@ -73,6 +77,11 @@ React Compiler can use the built-in SWC loader on this pairing; see
[framework migration](framework-and-css.md). Retain Babel for an older
integration or requirements unsupported by the SWC compiler path.
Native TypeScript declarations are a separate beta.1 opt-in. Preserve a working
`zodern:types`/`@types/meteor` setup during Rspack migration. Neither SWC nor
normal startup generates native declarations. Preserve a configured checker.
Use `meteor-typescript` when the task also includes changing providers.
## Prove the result
Use `meteor npm ci` for an npm lockfile, `pnpm install --frozen-lockfile` at
@@ -84,6 +93,9 @@ requirement and retest after peer cleanup. Do not convert managers to fix CI.
Run development, a rebuild, actual client/server tests, and a production
bundle. Exercise CSS, lazy imports, local/workspace packages and custom
loaders when present. Boot the extracted bundle and load the browser page.
For beta.1, rebuild and exercise native npm dependencies against Node 26.8.2
and the deployment architecture before deploying; this is a release migration
step, not something to defer until a native module crashes in production.
Follow the [validation matrix](validation-matrix.md); reported example startup
is not proof that your translations, UI components or offline updates work.
For a failure, capture the [migration report](troubleshooting.md#reporting-issues)
# misclassified that missing loader as an application startup bug.
# AppDir is the exact tree just packed, so this also avoids extracting
# the large image merely to make the capability decision.
if ! AppDir/usr/lib/wekan/bundle/node --version >/dev/null 2>&1; then
if ! LD_LIBRARY_PATH="$PWD/AppDir/usr/lib/wekan/bundle/node-runtime${LD_LIBRARY_PATH:+:$LD_LIBRARY_PATH}" AppDir/usr/lib/wekan/bundle/node --version >/dev/null 2>&1; then
echo "::warning::this runner cannot execute the bundled ${{ matrix.appimage_arch }} Node binary, so the AppImage was packed but not started. It is uploaded unchecked."
echo "::warning::v${{ steps.plan.outputs.version }} is missing ${{ steps.plan.outputs.manual }}, which this workflow cannot build: amd64 is the Meteor build, arm64/win64/mac-arm64 each need their own runner, and the Sandstorm .spk is signed. Run release-all.yml for those."
echo "::error::v${{ steps.plan.outputs.version }} still lacks ${{ steps.plan.outputs.manual }} after recovery. Inspect the matching full-release run; older unnamed runs need their failed jobs retried manually."
exit 1
# The six architectures nobody else builds for, repacked from the PUBLISHED
# amd64 bundle - which is what makes this possible without a Meteor build.
Loaded 100 of 2649 files, more files were not shown because too many files have changed in this diff.
Show more
Reference in new issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.