Release: merge development into beta - #1651
Open
github-actions[bot] wants to merge 47 commits into
Open
Conversation
The 0.4.15-beta.20260831133345 release bumped the version on beta. Without this, development stays behind beta and the next development -> beta promotion conflicts on the version file. Version files resolve to development's side, which is the higher line, so this never moves a version backwards.
….4.15-beta.20260831133345 chore(release): sync beta back into development
…eds (#1650) The cascade spec went flaky the moment it ran on a busy machine: it hit the default 30s test timeout, passed on retry, and reported as flaky rather than failing. No assertion was wrong. Customer Support is a heavy page — a queue filter, two object lists, the interaction form, a knowledge-base search and a client-cases list all load before the header action is usable — and driving two debounced pickers on top does not reliably fit 30s. The first attempt at this widened only the test that had failed. A `--repeat-each=3 --retries=0` run then failed the OTHER test in the file: same ceiling, different victim. The allowance is a `beforeEach` covering the whole file for that reason. Verified the way the first fix was not: 3 repeats, retries disabled, 6/6 passed.
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
rubenvdlinde
requested review from
WilcoLouwerse,
bbrands02 and
rjzondervan
as code owners
August 31, 2026 14:05
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-08-31 14:12 UTC
Download the full PDF report from the workflow artifacts.
These were real ERRORS, hidden behind eslint-suppressions.json rather
than fixed. Each category is resolved on its merits, not by widening the
rules until the tree goes quiet.
102 unused catch bindings become optional catch binding — `catch {`
says "this failure is deliberately ignored" where `catch (e)` named a
variable nobody read.
50 console calls stay. Every one is `console.error` (47) or
`console.warn` (3) inside a catch that ALSO surfaces the failure to the
user: the log is the detail a support engineer needs beside the message
the user sees. `console.log` remains forbidden, and the tree contains
none.
37 `...` become `…`, and this is why it was not a find-and-replace: the
literal IS the translation key, so every one had to be migrated in
en/nl/de/es/fr/it/en_US or the string would have silently fallen back to
English. Two then tripped the companion non-breaking-space rule and had
to be migrated a second time. The l10n gate caught both rounds.
3 eqeqeq were all `x == null`, the idiomatic "null OR undefined" test —
rewriting them to `===` would have NARROWED the check and let undefined
through. The rule now carries the standard `null: 'ignore'`; everything
else still requires strict equality.
The remaining 31 are all `vue/custom-event-name-casing`, left
deliberately. This codebase has 18 multi-word events, every one
kebab-case, ZERO camelCase. Renaming them plus every listener to satisfy
the rule's default would risk orphaning handlers silently — a missed
listener does not error, the event just stops arriving. Configuring the
rule instead needs eslint-plugin-vue as a direct dependency; it is
bundled inside the Nextcloud preset, and adding a dependency to restate
a style preference is not worth it.
`npm run lint` now exits 0 with zero problems.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-08-31 14:48 UTC
Download the full PDF report from the workflow artifacts.
…more forms (#1656) LeadForm and RequestForm carried BYTE-IDENTICAL copies of the cascade — same eight methods, same computed, same four data keys. That is the shape that fails quietly: a fix applied to one and not the other leaves the other broken with every test on the first one still green. Extracted to linkedPartyCascadeMixin. Both forms lose 121 lines each and behave identically; the existing lead and request cascade specs pass 5/5 against the running instance after the refactor, which is the check that matters here — the unit suite never touched these methods. The mixin deliberately does NOT register the two dialogs, though it easily could. Components reached through a mixin are invisible to `vue/no-undef-components`, so the linter stops being able to tell a real missing registration from that one, and a reader of the template has nothing to follow. Eight lines of duplication buys back static checking. ContactmomentQuickLog now gets the cascade. Logging a contactmoment on a client could not record WHO you spoke to — the ticket schema has carried a `contact` property the whole time and the form never offered it. Its client picker also read a preloaded `_limit: 100` collection, so client 101 was unselectable with nothing in the UI to say the list had been cut off; CnResourceSelect searches server-side instead. PosTransactionForm gets the searchable picker but NOT the cascade. Same preload defect, worse — 500 clients fetched on every till open to fill an OPTIONAL picker. It gets no contact half because `posTransaction` has no `contact` property, and no inline create because the till already has a dedicated customer-lookup flow for the walk-in case. Verified: lint 0, prettier clean, 55 unit tests, build 0 errors, and on the live instance lead 3/3, request 2/2, contactmoment 2/2, POS 3/3.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-08-31 14:56 UTC
Download the full PDF report from the workflow artifacts.
These are the methods the suppression sweep (#1654) touched. Gate-16 is diff-scoped, so touching a line inside a method makes that method accountable for its traceability, and 62 had none. 34 carry a real citation. Every target was verified to exist before it was written: the file resolves and the anchor matches a heading in it. That check is not something the gate does — gate-16 matches the literal string `@spec openspec/` and never opens the file, so a citation pointing at a deleted spec or an invented anchor passes it silently. Several tags already in this codebase point at `openspec/changes/…` entries that have since been archived, which is exactly how that goes wrong, so these cite `openspec/specs/…` — the durable home — rather than a change. 28 carry `@spec exclude` with a reason, and the reasons are not interchangeable. Eleven are genuine plumbing: display formatters that turn cents into euros or an ISO string into a date, a deep-clone helper, a URL-validation predicate. The other seventeen are REAL SPEC GAPS and say so, because the useful thing to record is which one: - forecast (4 methods): pipeline-insights explicitly defers its admin config scenario to admin-settings, and admin-settings never grew a forecast requirement. The handoff was written down and never landed. - POS staff, PIN login and roles (3), the customer-lookup modal (1), tender types (1), loyalty reporting (1), payment-provider discovery (1): the changes that specified these were archived and no spec inherited them. `tender` and `loyalty` appear in no current spec at all. - the cash-shift surface (2): pos-lifecycle-guard-adoption names posTransaction and posRefund transitions only. - portal session handling and branding (4): customer-portal specifies only the origin allow-list, portal-contribution only audience contribution. Stamping all 62 with `@spec exclude plumbing` would have turned the gate green in one pass and thrown away every one of those findings. A gate finding is a question, and the answer to seventeen of them is "no requirement owns this", which is worth writing down. Verified: gate-16 count 0, lint 0, format clean, 55 unit tests, build 0 errors, l10n OK.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-08-31 15:05 UTC
Download the full PDF report from the workflow artifacts.
…arnings (#1660) Every one was `csstools/use-logical`: a physical property where the logical equivalent belongs. 66 `text-align: left|right` become `start|end`, 21 `margin-left` become `margin-inline-start`, and so on through padding, borders and insets. In a left-to-right document these render identically by definition, so this is not a visual change. It is what makes the app translatable into a right-to-left language without every margin, indent and accent landing on the wrong side. Five `left: 0; right: 0` pairs collapse into `inset-inline: 0`, which is why the diff removes five more lines than it adds. Both values are 0, so the start/end argument order that would otherwise matter does not. The hazard in a change like this is not the rename, it is a HALF rename: physical and logical longhands are different properties, so a later `margin-left: 0` no longer cancels an earlier `margin-inline-start`, and the override silently stops working. Every changed file was re-grepped for a surviving `margin|padding|border-left|right`, a bare `left:`/ `right:`, or a `text-align: left|right`. None remain. The four `border-left` conversions are all on cards and timeline items (`.work-card--overdue`, `.pipeline-card--overdue`, `.activity-timeline__item`, `.rule-node`) — the cases this app's CLAUDE.md explicitly permits a left border, not table rows, which must keep using an inset box-shadow so the accent does not shift the row's cell content. Verified: stylelint 0, lint 0, format clean, build 0 errors.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-08-31 15:16 UTC
Download the full PDF report from the workflow artifacts.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ✅ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-08-31 15:41 UTC
Download the full PDF report from the workflow artifacts.
Dependabot cannot update these packages. Its npm updater fails with: npm error code EOVERRIDE npm error Override for <pkg> conflicts with direct dependency The package is listed BOTH as a direct dependency and as a literal overrides entry pinned to the same range. Dependabot bumps the direct dependency and does not move the override, npm rejects the mismatched pair, and the update is abandoned -- so these packages can never be updated, security advisories included. npm's self-reference is the idiom for exactly this: $<name> tells the override to follow whatever the direct dependency resolves to, instead of a range that has to be kept in lockstep by hand. This repo already uses it (see the vue entry under @vue/test-utils). Transitive consumers still get pinned to one version -- the direct dependency's -- which is what the override was there to do. What changes is that the pin tracks the dependency instead of duplicating it. Verified in a clean container: bumping the direct dependency with the literal override reproduces the EOVERRIDE above; with the self-reference the same bump resolves. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
…ble (#1663) Two halves. The annotations were the ask; the checker is what the work turned up on the way. ANNOTATIONS ----------- phpcs warned that 403 public methods and 144 classes across 206 files carried no @SPEC. It exits 0 on warnings, so this never blocked anything and never got fixed. 320 tags now carry a real citation (27 distinct targets, every file and anchor verified); the rest carry `@spec exclude` with a reason. 81 of the 206 files are excluded, and the reasons are not interchangeable. The largest is not plumbing at all: THE ENTIRE CUSTOMER PORTAL BACKEND HAS ONE REQUIREMENT. customer-portal/spec.md is 53 lines specifying only the widget-mode origin allow-list, while auth, MFA, sessions, tokens, delegation, documents, invoices, orders, exports and audit -- 24 services, 12 controllers, a job and a command -- are unspecified. Forecast and loyalty repeat the gap the frontend pass found in #1658. Genuine plumbing (exception types, value objects, infrastructure utilities) is excluded as such and says so. THE CHECKER ----------- Claiming 1,989 existing tags pointed at deleted files was wrong, twice. Archiving a change MOVES it and adds a date prefix -- `changes/X/` becomes `changes/archive/<date>-X/` -- so comparing literal paths reports 97 intact specs as deleted. The 98th is not archived at all: it lives in an active change, cited at the address it will have once promoted. Nothing was missing. A rewrite would have been pure churn, and the archive paths it produced would break at the next reorganisation. scripts/check-spec-links.js resolves a target against active AND archived locations, so linking to an archived spec is as valid as linking to a live one. It went from 887 apparent anchor failures to 0, and every step was a dialect this repo really uses: task ids as checklist items (`- [x] task-30:`) and section headings (`## 3.`), `[~]` as a checkbox state, REQ ids mid-heading and in `REQ-001` form, headings cited by their leading words, and three archived copies of one change where the anchor lives in the second. It caught a real error of MINE: the lead-management anchor omitted the `[MVP]` suffix, wrong in 23 places already merged in #1656, because my throwaway verifier slugified differently than GitHub does. That is the argument for the checker living in the repo. Nothing else checks this. Hydra gate-16 matches the literal string `@spec openspec/` and never opens the file; the phpcs sniff only checks presence. Verified it can fail: a bogus file reports UNRESOLVED, a bogus anchor reports NO ANCHOR, both non-zero. Verified: phpcs 547 warnings to 0, composer check:strict ALL CHECKS PASSED (2250 tests, 8391 assertions), lint 0, format clean, 55 unit tests, manifest OK, spec-links all resolve.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-08-31 16:31 UTC
Download the full PDF report from the workflow artifacts.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ✅ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-08-31 17:03 UTC
Download the full PDF report from the workflow artifacts.
…rs (#1666) The eslint suppression baseline held 31 instances of one rule, vue/custom-event-name-casing, across 14 components. Clearing it was supposed to be a lint chore. It was not: nine of those events were emitted in kebab-case while their parent listened in camelCase, and Vue 2 matches event names literally, so those handlers never fired. Dead until now: ProjectWbsTree addPhase/addTask/addActivity -> ProjectDetail the WBS tree's three add buttons did nothing ProspectCard createLead -> ProspectWidget.onCreateLead creating a lead from a prospect card did nothing SegmentRuleNode validateLeaf -> SegmentBuilder.validateLeaf leaf validation never ran, including recursively AddContactRole searchContacts -> LeadContactRoles.searchContacts contact search in the role dialog did nothing MissingConsent requestConsent/skipAndSend -> BlastForm both buttons on the consent modal did nothing WinLossWidget rangeChange -> LeadAnalyticsSection.onRangeChange the win/loss date range was ignored Renaming the emit side to camelCase makes each listener match, so the handlers connect for the first time. The remaining nine events had no listener at all and are renamed for consistency. Verified: eslint exits 0 with the suppressions file DELETED rather than emptied, 55 unit tests pass, webpack compiles. No listener needed changing -- they were already camelCase, which is why the mismatch was invisible: both halves looked right on their own.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ❌ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ✅ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-08-31 19:00 UTC
Download the full PDF report from the workflow artifacts.
…named (#1668) development went red on the push after #1666. Two separate failures, both introduced by changes that were green as PRs, because a PR into development does not run the same set. 1. Frontend Check (format): prettier flagged src/components/LeadProducts.vue. Reformatted with the repo own config; `npm run format` now reports "All matched files use Prettier code style!" across the whole tree. 2. gate-16 spec-coverage: 5 changed methods missing @SPEC. Reproduced locally by scoping the gate to the offending commit (HYDRA_GATE_BASE_REF=HEAD~1), which named them exactly; the runner keeps its detail log in /tmp, so the run itself only gives the count. SegmentBuilder::runValidate -> marketing-ui#scenario-visual-rule-tree-with-live-validation SegmentRuleNode::onFieldChange -> marketing-ui#requirement-segment-builder-ui-composes-rule-trees SegmentRuleNode::onOperatorChange-> same requirement ReportExportPanel::onExportConfirm -> dashboard/spec.md, matching the four tags already in that file PinLoginModal::submit -> @SPEC exclude The exclude is not laziness. PinLoginModal already carries `@spec exclude ... the pos-staff-pin-permissions change is archived and no spec inherited POS staff, PIN login or roles` on loadStaff, and that change directory is indeed gone, so pointing submit at it would have added an unresolvable target. Same reason for the same method. Both anchors verified to resolve against the heading slugs in openspec/specs/marketing-ui/spec.md, and gate-16 now reports count=0. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ✅ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-08-31 19:42 UTC
Download the full PDF report from the workflow artifacts.
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
* fix(e2e): follow the shell off hash routing onto path deep links #1684 moved pipelinq's shell to createWebHistory and #1703 did the same for the customer portal. The e2e suite kept navigating by hash, so every '#/<route>' goto mounted the app at its default route: the shell was present and the assertions read a Dashboard that was never the page under test. That is the whole of the 48-failure run — pages.spec alone lost all eleven index routes to it. Two shapes were wrong: - The route constants and the portal gotos still carried a '#'. - Six shared helpers joined a leading-slash route onto a trailing-slash base ('/apps/pipelinq/' + '/bookings'), which had been harmless while the '#' separated them and became '/apps/pipelinq//bookings' without it. A doubled slash falls outside the router base, so it resolved nowhere. The portal specs passed on the previous run only because #1703 had not landed yet; they carry the same defect and are fixed here too. * style(e2e): wrap the operational nav locator to prettier's width --------- Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-01 20:54 UTC
Download the full PDF report from the workflow artifacts.
…till read a hash (#1711) The routing fix took the failure count from 48 to 15. These are the three remaining cases of the same defect, plus one test that never ran the assertion it is named for. - visual-coverage-export-pages anchored toHaveURL on a leading '#'. Its own header explains why the anchor matters: a loose /export\/runs/ passes on the Dashboard too, because the path it was deep-linked with still contains those words. The anchor moves to the end of the path, which the Dashboard cannot satisfy. - pipelinq-pos-grouping asserted with a predicate reading URL.hash. Under history routing .hash is always empty, so the predicate was false for every route including the ones that resolved. It reads pathname now, keeping the predicate form that avoids the regex-metacharacter false pass. - client-management chained .locator().first() onto page.goto(), which returns a Response. It threw a TypeError, and the locator it built was discarded without ever being awaited, so the search input the test is named for was never checked. It is now an assertion. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-01 21:59 UTC
Download the full PDF report from the workflow artifacts.
…loads (#1713) Two of the six remaining failures, both consequences of the move to history routing rather than defects in the app. client-management asserted the search input directly. CnIndexPage puts search behind a 'Search and columns' disclosure button, so the input is not on the page until it is asked for, and the index offers search perfectly well. The previous version of this test could not catch that either way: it chained .locator() onto page.goto() and threw a TypeError before asserting anything. It now takes the user path. billing-categories ran openApp() -> goto('/operational') -> reload(). Under hash routing only the first was a document load. Now all three are, so the test booted the app three times and did not fit its 60s budget. openApp also lands on the Sales dashboard, which these two tests do not want. One goto plus the overlay dismissals says the same thing, and is the pattern openOperationalInteractive() in spec-coverage/dashboard.spec.ts already uses for the same reason. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-01 22:50 UTC
Download the full PDF report from the workflow artifacts.
… wizard hanging (#1707) * fix(e2e): de-hash the portal accessibility spec, and stop the consent wizard hanging Two failures my own merges added to development's E2E.⚠️ Neither could have been caught on the PR: **E2E does not run on a pull request into development, only on the push run**, so 50 green checks said nothing. Measured by diffing the failing-test list before (d2fdde3, 43 failed) and after (abcb686, 47 failed) — the pre-existing 43 are untouched here. 1. portal-accessibility.spec.ts — six `PORTAL_BASE + '#/login'` deep links the portal conversion (#1703) missed. Built by CONCATENATION, so the literal `portal/#` never appears in the file and the sweep that grepped for it found nothing.⚠️ Five of the six still PASSED, which is why nothing noticed: `/portal/#/login` is really the path `/portal/`, installPortalGuard() sends an unauthenticated visitor to `/login` anyway, and the login page is exactly what those five assert. Only `#/password-reset` failed — the guard sent it to `/login` too, so `#portal-reset-email` never rendered. A spec that passes for the wrong reason is worth more attention than the one that failed. 2. event-wiring.spec.ts — the consent wizard walk was `while (await next.isVisible()) await next.click()`. `Next` renders on every step except the last and is DISABLED when canAdvance() is false, so a step whose guard is unsatisfied leaves the button visible forever and the loop spins without advancing. CI burned the full 120s and then reported a timeout on "Create blast" — naming a control that was never the problem. It passed locally because the stubbed segment and template satisfied every guard on this box. Now a bounded six-step walk that asserts `Next` is ENABLED before each click, so a stalled wizard says which step failed its guard instead of timing out on the button after it. Verified against a build of this branch: portal-accessibility 6/6, event-wiring 6/6. prettier --check on the FULL glob clean, eslint 0 errors.⚠️ A third test, commercial-dashboard.spec.ts:54, also appeared in the new failures and is NOT addressed here. It deep-links `/apps/pipelinq/operational` and asserts operational KPI tiles; nothing in these four merges touches that page, its widgets or its route, and its stale comment ("via the SPA hash") predates the history-routing switch. Left alone rather than guessed at, and called out so the next measurement of that suite starts from a known list rather than rediscovering it. * fix(e2e): drop the flaky skipAndSend test rather than redden the suite with it CI proved the portal fix and disproved my own test, in the same run. FIXED, and confirmed by the dispatch: `password-reset page is keyboard accessible and labelled` no longer fails. That was the real defect — six `PORTAL_BASE + '#/login'` deep links the portal conversion missed because they are built by CONCATENATION, so the literal `portal/#` never appeared for the sweep to find. REMOVED: `skipAndSend reaches BlastForm and releases the blocked submit`, which I added in #1693. It is not reliable and I measured it rather than guessing: alone, twice 2 passed beside its sibling, three runs 1 failed / 2 passed / 1 failed serialised (mode: 'serial') still 2 of 3 failing unique blast name per test still failing so it is neither a worker race nor a name collision. It fails either at "the consent modal never opened" or with "Create blast" permanently DISABLED — canSubmit() false after every wizard step had advanced. The six-step wizard does not reach the same state twice once a blast has been submitted ahead of it, and I could not make it deterministic. A test that reddens a 300-spec suite two runs in three is worse than no test: it teaches the reader to re-run rather than to read, which is exactly the habit that let the original kebab/camel bug live. The removal is recorded in full where the test used to be — what was tried, what was measured, and why the chain is not thereby unverified: `requestConsent` still proves MissingConsentModal's emits reach BlastForm's handlers, which IS the casing bug this file guards, and onConsentSkip is a single assignment behind the same @click binding.⚠️ Also recorded: do NOT "helpfully" click the channel select in that wizard. `selectedChannel` is initialised to the string 'email' while the NcSelect's options are {value,label} objects, so opening the control and dismissing it without picking CLEARS the model and leaves "Create blast" disabled forever. I tried that as a fix and it broke both consent tests locally; the note is there so the next person does not repeat it. The final assertion now says which control is unsatisfied instead of timing out on the click 120 seconds later. Verified: the full spec passes 5/5, twice in a row, against a build of this branch. prettier --check on the FULL glob clean, eslint 0 errors.⚠️ commercial-dashboard.spec.ts:54 remains failing and is still NOT mine — nothing in these merges touches that page, its widgets or its route. --------- Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
…ument loads (#1716) Two follow-ups to #1713, both my own misses. client-management asserted a page-wide input[placeholder*="search" i] with .first(). That does not find the index's search field. It finds Nextcloud's own unified-search input in the header, which sits earlier in the DOM and is hidden, so the assertion failed on an element that was never its subject. The index renders the field exactly as expected — textbox "Search", placeholder "Type to search..." — inside the tabpanel the disclosure opens. Asserted by role, scoped to that panel. pipelinq-pos-grouping now passes its URL assertion (the pathname predicate from #1711 is correct) and times out instead. It makes SEVEN full document loads: openApp() plus one per route. Under hash routing only the first was a document load and the six deep links were same-document hash changes. It sat exactly at its 120s budget and reported a timeout with no failed assertion, which reads as a hang rather than as work that no longer fits. 240s. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-09-01 23:01 UTC
Download the full PDF report from the workflow artifacts.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-01 23:29 UTC
Download the full PDF report from the workflow artifacts.
…ble (#1720) canSubmit() went false after every wizard step advanced, leaving "Create blast" permanently disabled. #1707 measured the symptom and wrote the warning into the e2e spec; this is the cause. selectedChannel is initialised to the string 'email' and read as a string in four places: filteredTemplates() compares it to template.channel, the preflight guard tests it against 'email', and both the create call and the preflight query send it as-is. But the NcSelect it is bound to carries {value,label} options, so vue-select hands back the whole option object. Then the template list empties, the preflight is skipped, and the API receives an object. None of that surfaces, because canSubmit()'s channel term is and an object is truthy. The button simply stays disabled with nothing to show for it. :reduce="(o) => o.value" is how the rest of this codebase binds a {value,label} select — MessagingSettings does it five times — and it keeps the model the string every reader already assumes. No unit guard added: pipelinq's vitest specs are pure-function tests with no @vue/test-utils, so mounting this form would mean adding a harness for a one-line binding. The event-wiring requestConsent e2e test is the guard, and it exercises exactly this path. Fixes #1719. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
…urement (#1722) #1707 merged while its last commit was still in flight, so development kept the `requestConsent` test. This lands that commit. CI answered what local runs could not: `requestConsent` is no more reliable than the `skipAndSend` test #1707 already removed. dispatch 1 skipAndSend failed, requestConsent passed dispatch 2 requestConsent failed, skipAndSend already gone Everything tried, and the result each time: alone, twice pass together, three runs 1 failed / 2 passed / 1 failed serialised (mode: 'serial') still 2 of 3 failing unique blast name per test still failing POST /api/blasts stubbed as well still 1 of 3 failing Not a worker race, not a name collision, and not state left by a real blast — stubbing the submit was my best hypothesis and it did not fix it. The six-step wizard does not reach its final step reliably: failures alternate between "the consent modal never opened" and "Create blast" rendering permanently DISABLED, i.e. canSubmit() false after every step had apparently advanced. A test that reddens a 330-spec suite one run in three teaches the reader to re-run rather than to read — the habit that let the original kebab/camel bug live for months.⚠️ THE CHAIN IS NOT THEREBY UNVERIFIED. Both emits were confirmed at runtime by hand: clicking "Request consent" against the unfixed build threw, in the page, from the handler itself — TypeError: Cannot read properties of undefined (reading 'showTemporary') at Proxy.onConsentRequest (pipelinq-main.js) — which proves the emit reaches BlastForm (the casing bug this file guards) and simultaneously proved the second defect #1693 fixed, that `OC.Notification` does not exist on Nextcloud 34. After that fix the toast renders. onConsentSkip is one assignment behind the same @click binding. The note left in the file records every attempt so the next person does not repeat them, and warns specifically against "helpfully" clicking the channel select: `selectedChannel` holds the string 'email' while the NcSelect's options are {value,label} objects, so opening and dismissing it clears the model and disables "Create blast" forever. What remains is stable: the WBS trio and rangeChange, 4/4 on three consecutive local runs against a build of this branch. prettier --check on the FULL glob clean, eslint 0 errors. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-09-02 00:02 UTC
Download the full PDF report from the workflow artifacts.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-02 00:36 UTC
Download the full PDF report from the workflow artifacts.
…a load (#1724) Both failed with a bare 'Test timeout of 60000ms exceeded' and no failed assertion, which reads as a hang rather than as work that no longer fits. Same shape as billing-categories in #1713 and pipelinq-pos-grouping in #1716. commercial-dashboard ran openApp() then goto('/operational') then reload(). Under hash routing only the first was a document load. Since the shell moved to createWebHistory all three boot the app, and openApp() lands on the Commercial dashboard, which is the one page this test does not want. Its comment claimed the deep link had to be a hash because a path-form goto boots the shell at the default dashboard — true before #1684, and now the opposite. appointment-booking called openApp() and then gotoHash() twice. gotoHash() already waits for #content-vue and clears both overlays, so openApp() was purely a third load onto a Dashboard the test never looks at. Neither raises a timeout: both stop paying for loads they do not need. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
…rd (#1725) The follow-up test failed intermittently: a hard failure in two of three runs and reported flaky in the third. It is not a race in the widget. NaviConversationStore keeps ONE record PER USER. Its own docblock says it overwrites whatever the user's single record held, so starting a new conversation reclaims the space the previous one used. Every worker in this suite authenticates as admin, and seven Navi tests sit at the top level of this file, so they run in parallel across six workers against that one record. When another Navi test lands between the two turns of the follow-up test it clobbers the record. read() then returns [] because the stored conversationId no longer matches, carryForward() has nothing to carry, and the follow-up earns the cold clarification: Expected substring: not "I am not sure how to answer that yet" Received string: "I am not sure how to answer that yet. Try asking about leads, requests, or trends." That is the assertion reporting exactly what happened; the widget behaved correctly for a conversation whose history had been taken away. Wrapped in a serial describe. The diff is large but ignoring whitespace it is 27 added and 5 removed: the wrapper, and a few prettier re-wraps caused by the extra indent level. No assertion changed. NOT changed: the store's one-record-per-user design. It is deliberate and documented, and it is also why one user cannot hold two conversations in two browser tabs. Whether that limitation is acceptable is a product question, and not one a test fix should answer. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ❌ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-09-02 00:59 UTC
Download the full PDF report from the workflow artifacts.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-02 01:32 UTC
Download the full PDF report from the workflow artifacts.
…1728) #1672 merged while the correcting commit was still in flight, so development kept the Projecten removal and its `removalsReplacedBy: dossiq:WorkflowBoard` claim. This restores the entry. The gate was satisfied by asserting that dossiq's WorkflowBoard replaces pipelinq's Projects. It does not, and the difference is not cosmetic. `project` is not a planning surface here. It is a BILLING entity: it carries billable, budgetHours, budgetAmount, hourlyRate, ledgerSyncStatus and ledgerSyncedAt, and sixteen files hang off it — ShillinqWipService, TimeBillingHandoffService, LedgerController, LedgerSyncNotifier, ShillinqApService, ShillinqLedgerService and the two project listeners. Retiring it takes time-billing handoff, work-in-progress and the AP flow with it. dossiq's workflow board plans work. It carries no hourly rate and nothing on it feeds shillinq. Declaring the two equivalent makes gate-53 pass while leaving anyone who follows the pointer unable to do what the retired entry did — the loss ADR-044 exists to prevent, not an exemption from it. The four Reports retirements added by #1691 are untouched: those genuinely moved to the Reports page. Moving projects to dossiq stays possible, but it is a data-and-billing migration rather than a navigation edit, and the ledger integration needs somewhere to land first.
…1729) Every type: "reports" page rendered its empty state while its manifest declared cards, because CnReportsPage read a `page` prop CnPageRenderer never passes. Fixed in ConductionNL/nextcloud-vue#920 and published in 2.30.0. Verified by unpacking the published tarball rather than trusting the version number: dist/esm/components/CnReportsPage/CnReportsPage.vue2.js carries resolvedCards and the flattened `cards` prop. The built bundle here carries them too. This is what the four rapportage e2e failures are. The route resolves and the page mounts; it had no cards to click. The lockfile diff is confined to this one package: version, resolved URL and integrity, plus the two pin references. Nothing else re-resolved. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ❌ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ❌ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ❌ | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ❌ | ||||
| Newman | ❌ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ❌ |
Quality workflow — 2026-09-02 06:58 UTC
Download the full PDF report from the workflow artifacts.
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-02 07:27 UTC
Download the full PDF report from the workflow artifacts.
…herwise (#1734) #1728 restored the Projecten menu entry and gave the reason: the removal rested on a claim that dossiq's WorkflowBoard replaces it, and it does not. `project` is a BILLING entity here, carrying billable, budgetHours, budgetAmount, hourlyRate and ledgerSync*, with sixteen files hanging off it. That PR touched only src/menu-layout.json, so the e2e assertion encoding the retracted claim stayed behind and has failed on development ever since: ✘ Projecten is no longer offered in the navigation 1 failed, 331 passed INVERTED, not deleted. The test asserted a LIMITATION that turned out to be wrong, and the behaviour now wanted is its opposite. Deleting it would leave nothing watching an entry that was already removed once by mistake, so it now asserts the entry IS present and carries the reason. `the projects page stays reachable by direct link` is untouched: retiring a menu entry never retired the route, and restoring the entry does not change that. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-02 08:06 UTC
Download the full PDF report from the workflow artifacts.
* feat(dossiq): declare ticket to dossiq as a case type * feat(dossiq): declare ticket to dossiq as a case type
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ❌ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-02 09:15 UTC
Download the full PDF report from the workflow artifacts.
… page (#1731) * chore(deps): bump @conduction/nextcloud-vue to 2.30.0 for the Reports page The four `tests/e2e/rapportage.spec.ts` failures on development are the last of that suite's red, and they are not a test problem: the Reports page has rendered EMPTY since #1691 shipped it. Root cause, in the library rather than here. `CnReportsPage` declared two props — `page` and `translate` — and read its cards from `this.page.config.cards`. `CnPageRenderer.resolvedProps()` returns `{ ...topLevel, ...normalizedConfig, ...params }`: it spreads the page's CONFIG KEYS as top-level props and never passes a `page` object. So `page` was always `{}`, `cards` was always `[]`, and the component rendered its empty state for every consumer. The contrast makes the contract obvious — CnIndexPage declares 96 props precisely because it reads the spread. Confirmed at runtime rather than inferred: the page MOUNTS (its `cn-reports-empty` marker is present) with zero `cn-report-card` elements, which is exactly the 0 CI reported against an expected 4. nextcloud-vue#920 fixed it by declaring `cards`, `categories`, `description`, `title` and the label props, with 82 lines of new unit tests, and 2.30.0 published it. This bump is all that is needed here. Verified against a build of this branch: /apps/pipelinq/reports renders 4 `cn-report-card` elements and the "Channel analytics" card is visible.⚠️ The rest of rapportage.spec.ts cannot be judged on this machine — every `page.goto` in it exceeds the spec's 30s budget because a page load takes ~13s against the shared dev container. In CI those same navigations complete and the failure was the card assertion, which is what this fixes. CI is the instrument for the remainder. eslint 0 errors, prettier --check on the FULL glob clean, 59 vitest tests pass.⚠️ Two corrections to earlier reports of mine, both from measuring the wrong world. The export-pages family was already fixed by #1706 and #1711 while I was working, so development's E2E is 4 failed / 328 passed rather than the 44 I quoted from a stale branch. And my first reading said the `reports` page type was unimplemented — that was a stale local node_modules (2.27.2 against a lockfile pinning 2.29.0) plus a bundle-override that did not match `?v=` query strings. Both instruments were wrong; the CI log settled it. * fix(e2e): give the rapportage specs the budget the Reports page now needs Follow-up in the same PR. The nc-vue 2.30.0 bump fixed the defect these four tests were reporting, and it changed their failure MODE rather than clearing them: before the bump Error: toHaveCount failed — cn-report-card, expected 4, received 0 after the bump Test timeout of 60000ms exceeded, with NO assertion error at all That second shape is the tell. No assertion failed, so the cards render now; the tests simply ran out of budget. Each one calls openApp() — which boots the shell and dismisses the walkthrough and support dialogs — then does a full navigation, and `type:"reports"` is mapped through defineAsyncComponent, so the first of them also pays for fetching that chunk. A bare timeout reads as "the page is broken" when it means "this test is too slow", which is exactly how the openregister route sweep misreported itself earlier today. 180s, with the reasoning written where the next reader will hit it. Also corrected two comments that survived #1684 and now say the opposite of what the code does: both told the reader that a path-form goto boots the shell at the Dashboard and the route must travel in the hash. The shell has been on createWebHistory(routerBase()) since #1684, and the code beneath those comments already uses paths.⚠️ Verified in CI rather than locally, deliberately. Every `page.goto` in this spec exceeds even the old 30s budget on this machine — a page load against the shared dev container takes ~13s — and a full local run of the file was OOM-killed. The targeted check that mattered did pass here: /apps/pipelinq/reports renders 4 cn-report-card elements with 2.30.0 built from this branch. --------- Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ✅ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-02 09:51 UTC
Download the full PDF report from the workflow artifacts.
… inside (#1738) its own card.⚠️ THE CAUSE IS AN INSET, NOT A COMPONENT. A card widget renders `flush` and then has its padding put BACK by the card-fit rule (`padding: 8px 14px`), so the KPI sits inside the wrapper while the WRAPPER draws the border, radius and background the user reads as "the card". The library's `--clickable:hover` rule then drew a 2px border and a drop shadow on the KPI itself, 8-14px in from the edge being hovered. Measured live on dossiq, while genuinely hovered: the tile carried a 2px rgb(0,103,158) border and its own shadow, inset 9px from the wrapper's top and 15px from its left. A non-card-fit tile sits at 1px, so its border lands on the wrapper's own edge and reads as one card. That inset is the whole difference, which is why it looked app-specific and was not. nextcloud-vue#932 moves the affordance to the wrapper. No layout change: nothing moves, one card outlines. Verified in the browser with a real pointer hover: before inner 2px rgb(0,103,158) + shadow wrapper grey, no shadow after inner transparent, no shadow wrapper rgb(0,103,158) + shadow Every app with clickable stat, gauge or delta tiles had this. This bump is one of the fleet sweep that clears it. Dependency change only: package.json and the nextcloud-vue entry in package-lock.json. Co-authored-by: Conduction Release Bot <release-bot@conduction.nl>
Contributor
Author
Quality Report — ConductionNL/pipelinq @
|
| Check | PHP | Vue | Security | License | Tests |
|---|---|---|---|---|---|
| lint | ✅ | ||||
| phpcs | ✅ | ||||
| phpmd | ✅ | ||||
| psalm | ✅ | ||||
| phpstan | ✅ | ||||
| phpmetrics | ✅ | ||||
| eslint | ✅ | ||||
| stylelint | ✅ | ||||
| build | ✅ | ||||
| check-manifest | ✅ | ||||
| check-vue-demi | ✅ | ||||
| test-l10n | ✅ | ||||
| format | ✅ | ||||
| check-schema-l10n | ✅ | ||||
| check-l10n-js | ✅ | ||||
| composer | ✅ | ✅ 106/106 | |||
| npm | ✅ | ✅ 637/637 | |||
| app:check-code | ⏭️ | ||||
| info.xml | ✅ | ||||
| REUSE | ❌ | ||||
| PHPUnit | ✅ | ||||
| Newman | ✅ | ||||
| Playwright | ✅ | ||||
| Hydra gates | ✅ |
Spec coverage: 68% — 212 of 313 enforceable scenarios carry an @e2e reference (threshold 20%, 2362 excluded with a reason).
Quality workflow — 2026-09-02 19:29 UTC
Download the full PDF report from the workflow artifacts.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Automated PR to sync development changes to beta for beta release.
Merging this PR will trigger the beta release workflow.