ci: summarize static checks and retain complete reports (#2242)

Keep static feedback concise while retaining complete diagnostic reports,
visible lint errors, module-boundary checks, and the 3% duplication gate.
Upload complete reports after static failures and restore their outputs
from Turbo caches. Preserve unambiguous, checkout-relative clone locations.

Bind both lint caches to the website, documentation configuration and
experiment sources actually scanned by oxlint, including ignore policy.
Guard this ownership by comparing live checker discovery with resolved
Turbo inputs. A real website syntax error now invalidates a cached pass.

Validation: full current-head CI and the separate Go lane passed. Seven
real-checker regressions passed; failure-status and stale-report mutations
were rejected, and restored source passed. Actual Turbo restored both
reports byte-for-byte; full static/docs/workflow checks passed.

Closes #2238.
Refs #2190.

Assisted-by: Kogasa (gensokyo-skills:kogasa-surprise-testing)
This commit is contained in:
Ysopwnable
2026-10-13 01:02:35 +08:00
committed by GitHub
parent 8cbd57cafa
commit d96550f2f6
7 changed files with 407 additions and 4 deletions
+10
View File
@@ -68,6 +68,16 @@ correctness. Select relevant boundary files and real-provider lanes explicitly.
Source projects also contain large capacity regressions; `test:fast` promises
no compiled-runtime prerequisite, not a fixed time budget.
`npm run lint` prints errors and a warning summary; `npm run jscpd` prints
duplication statistics. Complete reports remain in `.cache/static-checks/` and
CI uploads them as `static-check-reports`, including after a failed static gate.
Use `npm run lint:details` or `npm run jscpd:details` for all console diagnostics.
Both duplicate commands scan production and test code with the same 3% limit;
the summary commands preserve checker failures and restore reports from cached
static tasks. Both lint caches include the website, documentation configuration,
and experiment code scanned by the checker; a regression compares the actual
lint inventory with Turbo's resolved inputs.
Routine iterations and rebases need focused regressions and relevant checks.
Before handing off a PR, run `npm run check` and generated-document checks when
applicable; CI owns the full suite and coverage. Use the full local gate for