lizhimins commented on PR #4063: URL: https://github.com/apache/rocketmq-dashboard/pull/4063#issuecomment-5578827238
Closing this batch of six PRs (#4063 - #4068). This comment is the summary for all six; the other five reference it. ## What we verified Each of these six PRs adds lines that are byte-identical to a PR that btlqql opened earlier the same day and then closed unmerged (2026-09-07 10:43-10:44 UTC): | This PR | Withdrawn original | Test file | Added lines | | :--- | :--- | :--- | :--- | | #4063 | #4015 | `web/src/layouts/navigationSearch.test.ts` | 9, verbatim identical | | #4064 | #4027 | `web/src/utils/messageTraceDiagnostics.test.ts` | 12, verbatim identical | | #4065 | #4013 | `web/src/constants/__tests__/theme.test.ts` | 17, verbatim identical | | #4066 | #4017 | `web/src/utils/resourceCsvImport.test.ts` | 16, verbatim identical | | #4067 | #4021 | `web/src/pages/ops/__tests__/auditPresentation.test.ts` | 8, verbatim identical | | #4068 | #4025 | `web/src/utils/consumerGroupDiagnostics.test.ts` | 10, verbatim identical | We compared the added lines of each pair after normalising trailing whitespace: all six match exactly, not merely in intent. Titles and target files are identical too. In addition, each of these PRs claims `Closes #NNNN` against an issue filed by btlqql, not by you: #4014, #4026, #4012, #4016, #4020, #4024. Those six issues are still open. Please do not claim another contributor's issue as the justification for a PR; this is the second time we have had to point that out (see #3317, closed as a duplicate of #3156, whose issue #3293 was filed by the author of #3156). ## Why the underlying tests were rejected To be clear: this is not "tests are unwelcome". Substantive tests are merged here regularly. These six were reviewed on 2026-09-07 and each fails on the merits: - **#4063 / #4015 (`navigationSearch`)** - mutation-insensitive. Remove the guard under test and the new test still passes, so it only buys a coverage line, not a behaviour guarantee. - **#4064 / #4027 (`messageTraceDiagnostics`)** - trunk already asserts the 1 / 4 / 0 group-count cases. The title says "distinct" but the implementation does not de-duplicate, so the test would freeze semantics that do not exist in the code. - **#4065 / #4013 (`theme`)** - the subject under test is a plain `as const` literal table. Asserting that its hex strings are six-digit hex and that its labels are scoped to the theme namespace is tautological, and duplicates assertions that already exist. - **#4066 / #4017 (`resourceCsvImport`)** - trunk already covers this at `web/src/utils/resourceCsvImport.test.ts:169-177`: empty name, 128-char and 121-char length violations, unsupported character, and a valid name. The new test restates those four cases with different sample strings. The PR body's framing also contradicts what trunk actually asserts. - **#4067 / #4021 (`auditPresentation`)** - the blank/undefined path of a pure presentation function is already pinned by existing cases. No defect, no increment. - **#4068 / #4025 (`consumerGroupDiagnostics`)** - asserts that an empty array has length 0. The branch actually worth testing (`progressIssues` early return) is untouched. ## What would be worth a PR instead The frontend CI job runs `npm ci` followed by `npm run build` and never runs `vitest` (`.github/workflows/ci.yml`, "Frontend Build (Node 20)" job). That means none of this repo's roughly 936 frontend tests gate anything, and two of them are already flaky. A PR that adds `npm run test` to that job would do more for this codebase than any single test file. Note that `ci.yml` itself currently fails at startup and reports no checks at all, so fixing that is a prerequisite - both are real, unclaimed, high-value gaps. If you do want to contribute tests: pick an uncovered branch with observable behaviour, confirm the test fails when the guarded code is deleted, and file your own issue for it. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
