aglinxinyuan opened a new issue, #7458:
URL: https://github.com/apache/texera/issues/7458
### Task Summary
Six component templates report **0% coverage** despite having substantial,
passing specs, and adding tests to them does not change the number. Together
they account for roughly 230 uncovered lines that no amount of test writing can
reach.
| Template | Spec size | Reported |
|---|---|---|
| `hub-workflow-detail.component.html` | 442-line spec, 7 `detectChanges` |
0% |
| `user-project.component.html` | 262-line spec | 0% |
| `search.component.html` | 242-line spec | 0% |
| `workspace.component.html` | — | 0% |
| `hub-search-result.component.html` | — | 0% |
| `search-results.component.html` | — | 75.8% (partial) |
The common factor is `TestBed.overrideComponent`. Across the repo, templates
whose spec calls it average **18%** coverage with 6 of 8 at zero; templates
whose spec does not average **76.7%**, with only 2 of 73 at zero. Every one of
these specs uses the additive `remove`/`add: { imports }` form to swap
heavyweight child components for stubs — they do **not** override the template
itself, so the real template genuinely renders and is genuinely exercised. It
simply is not attributed.
Two counterexamples (`search-results`, `operator-property-edit-frame`) also
call `overrideComponent` and do report coverage, so `overrideComponent` alone
is not a sufficient explanation and the precise trigger is still unidentified.
### Why this matters
These files look like the largest remaining coverage gaps in the frontend,
so they attract effort that cannot pay off. Two separate coverage sweeps ranked
`hub-workflow-detail` (46 lines) and `user-project` (38 lines) near the top of
the backlog before the pattern was spotted. Anyone working from the Codecov
file list will keep rediscovering them.
### Suggested next steps
- Confirm the mechanism — most likely the component recompilation
`overrideComponent` triggers produces a template with a different identity than
the one the coverage remapper knows about.
- If it cannot be fixed, exclude these templates from the coverage
denominator so the reported figure reflects what is actually testable, and note
the reason.
- Either way, record it somewhere discoverable so the next person ranking
targets does not spend the effort again.
### Task Type
- [ ] Refactor / Cleanup
- [x] DevOps / Deployment / CI
- [ ] Testing / QA
- [ ] Documentation
- [ ] Performance
- [ ] Other
--
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]
