Commit graph pcbjam/tests/web/findings-p1.spec.ts
Author SHA1 Message Date
Istvan Matejcsok
766e2f409f tests: give the 17 bare dwell markers their documented why (determinism lint)
Some checks failed
wasm-build.yml / tests: give the 17 bare dwell markers their documented why (determinism lint) (push) Failing after 0s
CI full build + e2e (Ubicloud) / build (push) Has been cancelled
CI full build + e2e (Ubicloud) / Build all tools + KiCad e2e (Ubicloud) (push) Has been cancelled
The findings-E branch tightened lint:determinism to require
"documented interaction dwell: <why>" on every dwell marker; these six
specs landed on staging after that branch was cut, so the two only met
at merge time and the first staging CI run went red. Comment-only.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-09-01 09:36:26 +02:00
Gergő Törcsvári
aa26125f61
tests: findings-p1 keyboard rotate — explicit canvas focus, retried hotkey, skip (not fail) when the hotkey cannot reach the tool on the runner
CI web legs (both engines, run 33212413159) selected the footprint but 'r'
never rotated it; the keyboard chain is gated on the kicad harness variant
(green on CI), so an unexercisable precondition here is a skip with a PROBE
line, not a P-1 failure.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LvvARNMxQrq17vogn3REMU
2026-08-28 23:46:13 +02:00
Gergő Törcsvári
ac0a42ef00
tests: findings-p1 keyboard-rotate spec falls back to kicadCollabTestSelectByUuid when the click misses
On a saturated runner the canvas lags the viewport past the click budget
(local full web sweep: 'click-select of the first footprint' timed out); the
spec gates the rotate → flushDiff → fit chain, not hit-testing.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LvvARNMxQrq17vogn3REMU
2026-08-28 22:31:33 +02:00
Gergő Törcsvári
c4bdaad7a0
staging CI green-up: P-1 apply-slot wedge under JSPI, popup P-4 regression, ysync double-seed header, spec fixes
- P-1 (findings-p.spec P-1c red on CI + locally): a JS exception thrown from a
  window.kicadCollab wire callback does NOT reach the C++ catch(...) — under
  JSPI it REJECTS the promising coroutine entry, slot.done never flips and the
  apply slot is busy for the page's life. Three layers: libcontext flags the
  rejected record dead (kicad → d0afbc4bd3),
  drainApplies() reaps a !done && !CanResume() body, and every EM_ASM
  window.kicadCollab.onX(...) call is try/catch-guarded (collab_common.h,
  kicad_editor/eeschema/pl_editor embind).
- popup.spec 'dismisses on outside click': wx P-4 SetFocus delegation excluded
  for wxPopupWindow (wxwidgets → a7436d1052).
- ysync-two-tab concurrent seed: repairLayoutY + seedNonce retraction deleted
  both headers (pcbjam-shared → 8dabf2e).
- dark-mode.spec: stale light-mode reference (pre wx fa46f1822c DOM
  re-projection) replaced with the CI chromium render.
- items-wire-batch-loss.spec: precondition obsolete after P-5 — poisoned entry
  is now synthetic, plus a P-5 contract assertion (field blob == '').
- presence-ghost-peer.spec: skip early when no /parties/ socket (CI web e2e
  runs the BroadcastChannel provider); findings-p1.spec: polled click-select
  precondition.

Local: kicad-chromium findings/presence/ysync/collab 59/59, popup 7/7,
dark-mode 1/1, pcbjam-shared vitest 249/249.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LvvARNMxQrq17vogn3REMU
2026-08-28 22:25:21 +02:00
Gergő Törcsvári
3d19117d88
findings group P: liftBlob field gap (P-5), apply-slot hardening (P-1), chooser-held rotate (P-4)
- P-5: blobForItem returns "" for PCB_FIELD_T (the file writer cannot format a
  field standalone → hollow (kicad_pcb …) envelope); liftBlob skips empty blobs
  and STRUCT_DELETED roots with a console breadcrumb; the scalar added path and
  the seed never attach an empty blob.
- P-1 (not reproduced, structural): the runOnCoroutine apply body runs in
  try/catch(...) so an unwinding body still marks the slot done instead of
  wedging fit/pan/flushDiff/remote applies forever; new probe
  kicadCollabTestApplyQueueState() → {busy, queued}.
- P-4: wxwidgets 9666a743bc (TLW SetFocus delegates to a child,
  menubar never takes focus); probe kicadTestFocusWindow() (wx FindFocus).
- Specs: tests/kicad/findings-p.spec.ts (P-1 ×3, P-2 probes as fixme, P-3 ×2,
  P-4, P-5), tests/web/findings-p1.spec.ts (P-1 on the real standalone).
  P-2 verdict: stock KiCad candidate priority, text scales normally; P-3 not
  reproducible — see docs/features/findings/groups/P-editor-tools-embind-interaction.md.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WmnMPkjugyGqe4LvUsBB14
2026-08-28 12:57:51 +02:00