aglinxinyuan opened a new pull request, #8031:
URL: https://github.com/apache/texera/pull/8031
### What changes were proposed in this PR?
Two small frontend files. Measured with the identical spec-file filter on
both sides (two repeated `--include` globs, never a test-name phrase), `rm -rf
coverage` between runs, reading `frontend/coverage/gui/lcov.info`.
| File | Codecov metric | Delta |
|---|---|---|
| `computing-unit-actions.service.ts` | 26/29 = 89.7% → **29/29 = 100%** |
+3 lines |
| `result-panel.component.html` | 50/53 = 94.3% → **53/53 = 100%** | +3
lines |
**+6 fully-covered lines and +5 branch arms.** The before figure on the
template reproduces Codecov's published 94.3% exactly.
This is a small PR and I would rather say so than dress it up. What makes it
worth having is the template's function counter.
### The template's bindings were never entered
`result-panel.component.html` sat at **FNF 7 / FNH 4** — three of its seven
functions uncovered — because the existing spec called `handleEndDrag(...)` and
`handleStartDrag()` **directly on the component**. The bindings themselves
stayed at `FNDA:0`: covered-looking, unexercised. Driving them through the
rendered element is what closes them.
That is the seventh time in this campaign a file has sat at a healthy line
percentage with uncovered functions, and the second time the specific cause was
a spec calling a handler directly instead of firing the binding.
### Verification
18 mutations, **all 18 killed.**
The first draft claimed `survivors: []` alongside "NO DEFECTS FOUND" and
"nothing in the advertised gap was unreachable". Its own mutants were re-run
against the bundle as shipped first, and several survived it; all now die.
Two further corrections to that draft:
- **An arithmetic error.** It reported the service as `23/26 → 26/26`. Lines
101 and 109 carry `BRDA` records but **no `DA`** record — v8 attributes the
`DA` for the whole `modalService.confirm({...})` statement to line 99 — so on
the `union(DA, BRDA)` line set the correct figures are 26/29 → 29/29. Same
delta, different denominator, and worth stating so the numbers reproduce.
- **A miscredited kill.** One mutation's row named a single test with the
second in a parenthetical "also:". Re-derived, it fails two tests outright.
### One vacuity worth naming, though it is not a survivor
The pre-existing test "renders the resize handles when the panel is docked"
still passes a swap of the line-120 `nzDirections` arm on its own, because it
never distinguishes the two directions. It is not a file-level survivor — the
new tests kill that mutant — but the older test does not constrain what its
name suggests.
### Deliberately not included
Nothing. Both files reach 100% on lines, branches and functions, and neither
carries an unreachable partial — which is unusual at this point in the codebase
and is why this bundle was taken at six lines when three sibling bundles at the
same size were dropped.
No production file is touched. The worktree used a real yarn install rather
than a `node_modules` junction.
### Any related issues, documentation, discussions?
Closes #8030
### How was this PR tested?
```
npx ng test --watch=false
--include="**/computing-unit-actions.service.spec.ts"
--include="**/result-panel.component.spec.ts"
```
```
Test Files 2 passed (2)
```
`yarn format:ci` passes. `frontend/junit.xml` and `frontend/coverage/` are
regenerated by every run and are not committed.
### Was this PR authored or co-authored using generative AI tooling?
Generated-by: Claude Code (Opus 5)
--
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]
