Files
ui/_docs/LESSONS.md
T
Oleksandr Bezdieniezhnykh bd2b718ddf [AZ-463] [AZ-469] [AZ-476] [AZ-477] Batch 6 - flight/responsive/upload/settings tests
- AZ-463 flight selection persistence (FT-P-16) + rehydration
  on boot (FT-P-17) PASS at the wire; 100-cycle leak guard
  (NFT-RES-LIM-07) and 1h SSE soak (NFT-RES-LIM-06)
  scaffolded as RUN_LONG_RUNNING-gated e2e companions.
- AZ-469 browser-support smoke (FT-P-34) runs in both
  Chromium and Firefox via the existing playwright config;
  responsive variants (FT-P-35 480px / FT-P-36 1024px) PASS
  in fast (Tailwind class shape) and e2e (visibility).
- AZ-476 upload 501 MB -> 413: AC-1 user-visible error is
  drift today (uploadFiles silently falls through to local
  mode); it.fails() + control + e2e test.fail. AC-2 no-alert
  PASS via dialog spy.
- AZ-477 settings save 500 / network drop: AC-1+AC-2+AC-3
  all drift today (no try/finally, no error region, deadline
  unmeasurable); 4 it.fails() + control pinning the stuck-
  disabled drift; e2e companions test.fail mirror it.
- LESSONS.md seeded: vi.stubGlobal('URL', {...URL,...})
  destroys the URL constructor and breaks new URL(...) in
  MSW; patch the methods directly instead.

Code review: PASS (0 findings). Fast: 22/22 files, 120
passed / 13 skipped. Static: 24/24 PASS.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-05-11 05:19:35 +03:00

21 lines
860 B
Markdown

# Lessons
Short, actionable retros from past sessions. Newest at top. Each entry: one
line title + ≤6 lines body. The `autodev` orchestrator surfaces the top 3
entries on every invocation.
---
## 2026-05-11 — Don't replace `URL` via `vi.stubGlobal('URL', { ...URL, ... })`
When stubbing `URL.createObjectURL` / `URL.revokeObjectURL` for a JSDOM-backed
test, **patch the methods on the constructor directly**. Never do
`vi.stubGlobal('URL', { ...URL, createObjectURL })` — the spread copies only
own enumerable properties of the `URL` *function object*, not its prototype, so
the global `URL` becomes a plain object. `new URL(...)` then throws / returns
garbage in MSW handlers and the SPA's API helper, and the test silently sees
"no fetch was made" instead of the real failure. Pattern in
`tests/upload_size_cap.test.tsx` is the canonical fix.
---