pcbjam/tests/kicad/pcbnew.spec.ts

295 lines
14 KiB
TypeScript
Raw Normal View History

2026-03-22 12:48:05 +01:00
import type { Page } from '@playwright/test';
import { test, expect } from './fixtures';
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
import {
clickByTooltip,
findByTooltip,
waitForCanvasStable,
waitForEditorReady,
e2e/CI: dual-engine suites, per-engine screenshots, SwiftShader retired, prod web suite, CI-coverage gate Squash of experiment/ff-big-modules vs main. Big-module routing removed: native-EH shrank kicad_editor below SpiderMonkey's x86-64 code budget (runs 29355049705/29356152413 green on stock Firefox), so BIG_MODULE_SPECS routing and the baseline-only-JIT crutch are gone — kicad-firefox and kicad-chromium both run the full suite, with the module compiled the way real users' browsers compile it. Per-engine screenshots end to end: stableShot/shotPath write test-results/<engine>/<name>.png; baselines move to baseline-screenshots/{chromium,firefox}/ and the whole tools/screenshots pipeline (compare/promote/manifest/spec-map/changelog/Discord) keys on <engine>/<name>. Previously Firefox and Chromium renders of one spec overwrote each other and Firefox renders were never actually gated. Seeded from CI run 29421380806 (92 new firefox baselines, +24 chromium web-suite shots); manifest generated from the baseline tree. One merged playwright.config.ts (kicad/asyncify/coroutine/perf as projects); ~25 dead npm scripts dropped. The web suite is gated in CI for the first time ever (4 rotted specs fixed, 5 broken lib-bridge specs triaged as fixme in docs/features/web-e2e-rot/); cheap lint step after npm ci; last 26 blind-sleep violations fixed. SwiftShader retired: CI Chromium renders WebGL on ANGLE → Mesa llvmpipe (--use-gl=angle --use-angle=gl --ignore-gpu-blocklist; the blocklist flag is mandatory — llvmpipe is blocklisted and WebGL is silently unavailable without it) in BOTH configs. Under WORKERS=4 congestion SwiftShader transiently failed the first post-board-load draw and the recovery cascade ended in a silent permanent Cairo fallback — that engine flip was the "~1.2% changedRatio both directions" occ-export baseline flake. Validated 160/160 across two 80-repeat rigs; full analysis in docs/features/wx-parity-bugs/occ-export-context-eviction.md. Chromium baselines shift slightly on llvmpipe — promote once from the first green run. Deflakes the new coverage exposed: presence baselines settle before capture; presence fixtures declare current file formats; perf gets its own outputDir so CI evidence survives; occ-export settles the board paint before the export dialog; menu-item waits (waitForRenderedByLabel before clickMenuItem) in 4 specs + the TESTING.md rule. Web suite runs the PROD build, in parallel: webServer becomes backend `start` + the standalone's e2e:preview (build-preview.mjs: link-wasm → stash the public/wasm symlink aside during vite build, build-demo.mjs's move — then vite preview as the persistent server). The wasm middleware serves /wasm/* in preview and emits COOP/COEP/CORP itself (a pthread worker script's own response must carry COEP or Chrome kills it with ERR_BLOCKED_BY_RESPONSE). VITE_* flags bake at build time; VITE_ALLOW_USER_OVERRIDE joins turbo globalEnv. fullyParallel + default workers: 5.2m → 1.4m. Determinism fixes the parallel run exposed: shared-page specs become serial groups; locks.spec grabs alice's exact item via the new kicadCollabTestSelectByUuid hook (cross-tab "first footprint" order is not a ysync invariant); quit specs poll page.url() (quit supersedes its own navigation — NS_BINDING_ABORTED on Firefox). Suite: 51 passed / 12 skipped / 0 failed in 1.6m. CI-coverage gate (lint:ci-coverage): every tests/**/*.spec.ts must be reachable from the npm scripts the workflows invoke — scraped from .github/workflows/, resolved through package.json, coverage asked from playwright --list itself. Rules: uncovered-spec + orphan-project (with a documented LOCAL_ONLY_PROJECTS allowlist). Gating next to lint:determinism; 138 spec files / 13 projects accounted for. Product fixes kept from the investigations (reachable on real GPUs too): wx 7799fd1be5 — paint flags clear before dispatch + Invalidate always propagates; kicad 3dcfea5e45 — SwiftShader pass-boundary flush + per-instance font texture + first-frame GL-error drain (GAL recovery recovers instead of falling back to Cairo) + the user-facing eeschema switch navigates again under __EMSCRIPTEN__ (project-sync's FaceRegistered gate had rerouted it into the hidden sync player; caught by the newly-gated web suite). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018eUxiPApHgGiu9NFyQfhAq
2026-07-17 12:10:40 +02:00
waitUntil, stableShot, shotPath } from '../e2e/utils/element-tracker';
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
import { hideCursor } from './utils/screenshot-compare';
/**
* PCBnew WASM E2E Tests
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
*
* pcbnew.html seeds a default KiCad config in preRun, so the first-run setup
* wizard never opens the editor comes straight up and we wait deterministically
* for its canvas + toolbars. The launch shot uses stableShot (stabilizes then
* compares, replacing the old header-region reference diff). The line-drawing test
* proves a segment rendered via a functional before/after pixel diff.
*/
2026-03-22 12:48:05 +01:00
type CanvasMetrics = {
dpr: number;
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
mainCanvas: null | { width: number; height: number; rectWidth: number; rectHeight: number };
glCanvas: null | { id: string; width: number; height: number; rectWidth: number; rectHeight: number; viewport: number[] | null };
2026-03-22 12:48:05 +01:00
};
type RegistryMetrics = {
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
toolbars: Array<{ id: string; typeName: string; width: number; height: number; label: string; name: string }>;
auiParts: Array<{ id: string; subType: string; label: string; width: number; height: number }>;
2026-03-22 12:48:05 +01:00
};
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
type DiffRegion = { x: number; y: number; width: number; height: number };
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
type ScreenshotDifference = {
actualWidth: number;
actualHeight: number;
diffPixels: number;
diffRatio: number;
meanChannelDiff: number;
};
async function compareScreenshots(
page: Page,
beforePng: Buffer,
afterPng: Buffer,
region: DiffRegion
): Promise<ScreenshotDifference> {
return page.evaluate(async ({ beforeBase64, afterBase64, crop }) => {
const loadImage = async (base64: string): Promise<HTMLImageElement> => {
const image = new Image();
image.src = `data:image/png;base64,${base64}`;
await image.decode();
return image;
};
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const [before, after] = await Promise.all([loadImage(beforeBase64), loadImage(afterBase64)]);
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
if (before.width !== after.width || before.height !== after.height) {
return {
actualWidth: after.width,
actualHeight: after.height,
diffPixels: Number.POSITIVE_INFINITY,
diffRatio: Number.POSITIVE_INFINITY,
meanChannelDiff: Number.POSITIVE_INFINITY,
};
}
const canvas = document.createElement('canvas');
canvas.width = crop.width;
canvas.height = crop.height;
const context = canvas.getContext('2d', { willReadFrequently: true });
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
if (!context) throw new Error('2D canvas context unavailable for screenshot comparison');
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
context.drawImage(before, crop.x, crop.y, crop.width, crop.height, 0, 0, crop.width, crop.height);
const beforeData = context.getImageData(0, 0, canvas.width, canvas.height).data;
context.clearRect(0, 0, canvas.width, canvas.height);
context.drawImage(after, crop.x, crop.y, crop.width, crop.height, 0, 0, crop.width, crop.height);
const afterData = context.getImageData(0, 0, canvas.width, canvas.height).data;
let diffPixels = 0;
let totalChannelDiff = 0;
for (let i = 0; i < beforeData.length; i += 4) {
const dr = Math.abs(beforeData[i] - afterData[i]);
const dg = Math.abs(beforeData[i + 1] - afterData[i + 1]);
const db = Math.abs(beforeData[i + 2] - afterData[i + 2]);
const da = Math.abs(beforeData[i + 3] - afterData[i + 3]);
totalChannelDiff += dr + dg + db + da;
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
if (Math.max(dr, dg, db, da) > 16) diffPixels += 1;
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
}
return {
actualWidth: after.width,
actualHeight: after.height,
diffPixels,
diffRatio: diffPixels / (canvas.width * canvas.height),
meanChannelDiff: totalChannelDiff / beforeData.length,
};
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
}, { beforeBase64: beforePng.toString('base64'), afterBase64: afterPng.toString('base64'), crop: region });
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
}
2026-03-22 12:48:05 +01:00
async function getCanvasMetrics(page: Page): Promise<CanvasMetrics> {
return page.evaluate(() => {
const dpr = window.devicePixelRatio || 1;
const mainCanvas = document.querySelector('#canvas') as HTMLCanvasElement | null;
const glCanvas =
Array.from(document.querySelectorAll('[id^="glcanvas-"]'))
.map((canvas) => canvas as HTMLCanvasElement)
.find((canvas) => {
const rect = canvas.getBoundingClientRect();
const style = window.getComputedStyle(canvas);
return style.display !== 'none' && rect.width > 0 && rect.height > 0;
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
}) ?? (document.querySelector('[id^="glcanvas-"]') as HTMLCanvasElement | null);
2026-03-22 12:48:05 +01:00
const mainRect = mainCanvas?.getBoundingClientRect();
const glRect = glCanvas?.getBoundingClientRect();
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const gl = glCanvas?.getContext('webgl2') || glCanvas?.getContext('webgl');
2026-03-22 12:48:05 +01:00
const viewport = gl ? Array.from(gl.getParameter(gl.VIEWPORT) as Int32Array | number[]) : null;
return {
dpr,
mainCanvas: mainCanvas && mainRect ? {
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
width: mainCanvas.width, height: mainCanvas.height,
rectWidth: mainRect.width, rectHeight: mainRect.height,
2026-03-22 12:48:05 +01:00
} : null,
glCanvas: glCanvas && glRect ? {
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
id: glCanvas.id, width: glCanvas.width, height: glCanvas.height,
rectWidth: glRect.width, rectHeight: glRect.height, viewport,
2026-03-22 12:48:05 +01:00
} : null,
};
});
}
async function getRegistryMetrics(page: Page): Promise<RegistryMetrics> {
return page.evaluate(() => {
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const registry = window.wxElementRegistry!;
const toolbars = registry.findAll({ visible: true })
2026-03-22 12:48:05 +01:00
.filter((element) => /ToolBar/.test(element.typeName))
.map((element) => ({
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
id: element.id, typeName: element.typeName,
width: element.width, height: element.height,
label: element.label, name: element.name,
2026-03-22 12:48:05 +01:00
}));
const auiParts = registry.findAllRendered
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
? registry.findAllRendered({ elementType: 'auipart' }).map((part) => ({
id: part.id, subType: part.subType, label: part.label,
width: part.width, height: part.height,
}))
2026-03-22 12:48:05 +01:00
: [];
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
return { toolbars, auiParts };
2026-03-22 12:48:05 +01:00
});
}
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
test.describe('PCBnew WASM', () => {
test.beforeEach(async ({ page }) => {
await page.goto('/kicad/pcbnew.html');
});
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
test('loads PCBnew with sane canvas + toolbar + pane metrics', async ({ page }) => {
await waitForEditorReady(page);
2026-03-22 12:48:05 +01:00
const metrics = await getCanvasMetrics(page);
const registryMetrics = await getRegistryMetrics(page);
2026-06-04 13:20:55 +02:00
expect(metrics.dpr).toBeGreaterThanOrEqual(1);
2026-03-22 12:48:05 +01:00
expect(metrics.mainCanvas).not.toBeNull();
expect(metrics.glCanvas).not.toBeNull();
expect(registryMetrics.toolbars.length).toBeGreaterThanOrEqual(4);
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const mainCanvas = metrics.mainCanvas!;
const glCanvas = metrics.glCanvas!;
2026-03-22 12:48:05 +01:00
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
expect(Math.round(mainCanvas.rectWidth * metrics.dpr)).toBe(mainCanvas.width);
expect(Math.round(mainCanvas.rectHeight * metrics.dpr)).toBe(mainCanvas.height);
expect(glCanvas.rectWidth).toBeGreaterThan(800);
expect(glCanvas.rectHeight).toBeGreaterThan(500);
expect(Math.round(glCanvas.rectWidth * metrics.dpr)).toBe(glCanvas.width);
expect(Math.round(glCanvas.rectHeight * metrics.dpr)).toBe(glCanvas.height);
2026-03-22 12:48:05 +01:00
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const viewport = glCanvas.viewport;
2026-03-22 12:48:05 +01:00
expect(viewport).not.toBeNull();
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
expect(viewport![2]).toBe(glCanvas.width);
expect(viewport![3]).toBe(glCanvas.height);
2026-03-22 12:48:05 +01:00
const verticalToolbars = registryMetrics.toolbars.filter((toolbar) => toolbar.height > 100);
expect(verticalToolbars).toHaveLength(2);
for (const toolbar of verticalToolbars) {
expect(toolbar.width).toBeLessThanOrEqual(40);
}
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const appearancePane = registryMetrics.auiParts.find((part) => part.subType === 'content' && part.label === 'Appearance');
expect(appearancePane, 'Appearance pane present').toBeTruthy();
expect(appearancePane!.width).toBeGreaterThanOrEqual(200);
expect(appearancePane!.width).toBeLessThanOrEqual(240);
2026-03-22 12:48:05 +01:00
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
await hideCursor(page);
await stableShot(page, 'pcbnew-loaded-css.png');
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
await stableShot(page, 'pcbnew-loaded.png');
const canvasCount = await page.locator('canvas').count();
expect(canvasCount).toBeGreaterThan(0);
});
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
test('select draw lines and draw on the board', async ({ page, testLogger }) => {
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
await waitForEditorReady(page);
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
await hideCursor(page);
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
// Wait for the Draw Lines tool to render into the toolbar registry.
await waitUntil(
page,
() => {
const r = window.wxElementRegistry;
if (!r?.findAllRendered) return false;
return r.findAllRendered({ elementType: 'tool' }).some((tool) => tool.tooltip?.includes('Draw Lines'));
},
'Draw Lines tool rendered',
);
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
const drawLinesTool = await findByTooltip(page, 'Draw Lines', { elementType: 'tool' });
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
expect(drawLinesTool, 'Draw Lines tool present').not.toBeNull();
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const isToolChecked = (t: { label?: string } | null | undefined) => (t?.label ?? '').includes('[checked]');
expect(drawLinesTool!.enabled).toBe(true);
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
expect(isToolChecked(drawLinesTool)).toBe(false);
const baselineErrorCount = testLogger.errors.length;
await stableShot(page, 'pcbnew-draw-lines-00-before-tool-click.png');
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
expect(await clickByTooltip(page, 'Draw Lines', { elementType: 'tool' })).toBe(true);
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
await expect.poll(async () => isToolChecked(await findByTooltip(page, 'Draw Lines', { elementType: 'tool' })), {
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
message: 'Draw Lines tool should stay selected after the click',
timeout: 5000,
}).toBe(true);
const glCanvasId = await page.evaluate(() => {
const glCanvas =
Array.from(document.querySelectorAll('[id^="glcanvas-"]'))
.map((canvas) => canvas as HTMLCanvasElement)
.find((canvas) => {
const rect = canvas.getBoundingClientRect();
const style = window.getComputedStyle(canvas);
return style.display !== 'none' && rect.width > 0 && rect.height > 0;
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
}) ?? (document.querySelector('[id^="glcanvas-"]') as HTMLCanvasElement | null);
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
return glCanvas?.id ?? null;
});
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
expect(glCanvasId, 'a visible GL canvas exists').not.toBeNull();
const glSel = `#${glCanvasId}`;
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const glCanvasBox = await page.locator(glSel).boundingBox();
expect(glCanvasBox, 'GL canvas bounding box available').not.toBeNull();
const box = glCanvasBox!;
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
// Move the crosshair onto the canvas (a visible change → deterministic settle).
await page.mouse.move(640, 360);
await waitForCanvasStable(page, glSel);
e2e/CI: dual-engine suites, per-engine screenshots, SwiftShader retired, prod web suite, CI-coverage gate Squash of experiment/ff-big-modules vs main. Big-module routing removed: native-EH shrank kicad_editor below SpiderMonkey's x86-64 code budget (runs 29355049705/29356152413 green on stock Firefox), so BIG_MODULE_SPECS routing and the baseline-only-JIT crutch are gone — kicad-firefox and kicad-chromium both run the full suite, with the module compiled the way real users' browsers compile it. Per-engine screenshots end to end: stableShot/shotPath write test-results/<engine>/<name>.png; baselines move to baseline-screenshots/{chromium,firefox}/ and the whole tools/screenshots pipeline (compare/promote/manifest/spec-map/changelog/Discord) keys on <engine>/<name>. Previously Firefox and Chromium renders of one spec overwrote each other and Firefox renders were never actually gated. Seeded from CI run 29421380806 (92 new firefox baselines, +24 chromium web-suite shots); manifest generated from the baseline tree. One merged playwright.config.ts (kicad/asyncify/coroutine/perf as projects); ~25 dead npm scripts dropped. The web suite is gated in CI for the first time ever (4 rotted specs fixed, 5 broken lib-bridge specs triaged as fixme in docs/features/web-e2e-rot/); cheap lint step after npm ci; last 26 blind-sleep violations fixed. SwiftShader retired: CI Chromium renders WebGL on ANGLE → Mesa llvmpipe (--use-gl=angle --use-angle=gl --ignore-gpu-blocklist; the blocklist flag is mandatory — llvmpipe is blocklisted and WebGL is silently unavailable without it) in BOTH configs. Under WORKERS=4 congestion SwiftShader transiently failed the first post-board-load draw and the recovery cascade ended in a silent permanent Cairo fallback — that engine flip was the "~1.2% changedRatio both directions" occ-export baseline flake. Validated 160/160 across two 80-repeat rigs; full analysis in docs/features/wx-parity-bugs/occ-export-context-eviction.md. Chromium baselines shift slightly on llvmpipe — promote once from the first green run. Deflakes the new coverage exposed: presence baselines settle before capture; presence fixtures declare current file formats; perf gets its own outputDir so CI evidence survives; occ-export settles the board paint before the export dialog; menu-item waits (waitForRenderedByLabel before clickMenuItem) in 4 specs + the TESTING.md rule. Web suite runs the PROD build, in parallel: webServer becomes backend `start` + the standalone's e2e:preview (build-preview.mjs: link-wasm → stash the public/wasm symlink aside during vite build, build-demo.mjs's move — then vite preview as the persistent server). The wasm middleware serves /wasm/* in preview and emits COOP/COEP/CORP itself (a pthread worker script's own response must carry COEP or Chrome kills it with ERR_BLOCKED_BY_RESPONSE). VITE_* flags bake at build time; VITE_ALLOW_USER_OVERRIDE joins turbo globalEnv. fullyParallel + default workers: 5.2m → 1.4m. Determinism fixes the parallel run exposed: shared-page specs become serial groups; locks.spec grabs alice's exact item via the new kicadCollabTestSelectByUuid hook (cross-tab "first footprint" order is not a ysync invariant); quit specs poll page.url() (quit supersedes its own navigation — NS_BINDING_ABORTED on Firefox). Suite: 51 passed / 12 skipped / 0 failed in 1.6m. CI-coverage gate (lint:ci-coverage): every tests/**/*.spec.ts must be reachable from the npm scripts the workflows invoke — scraped from .github/workflows/, resolved through package.json, coverage asked from playwright --list itself. Rules: uncovered-spec + orphan-project (with a documented LOCAL_ONLY_PROJECTS allowlist). Gating next to lint:determinism; 138 spec files / 13 projects accounted for. Product fixes kept from the investigations (reachable on real GPUs too): wx 7799fd1be5 — paint flags clear before dispatch + Invalidate always propagates; kicad 3dcfea5e45 — SwiftShader pass-boundary flush + per-instance font texture + first-frame GL-error drain (GAL recovery recovers instead of falling back to Cairo) + the user-facing eeschema switch navigates again under __EMSCRIPTEN__ (project-sync's FaceRegistered gate had rerouted it into the hidden sync player; caught by the newly-gated web suite). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018eUxiPApHgGiu9NFyQfhAq
2026-07-17 12:10:40 +02:00
const afterToolClick = await page.screenshot({ path: shotPath(page, 'pcbnew-draw-lines-01-after-click.png'), scale: 'css' });
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const startPoint = { x: Math.round(box.x + box.width * 0.28), y: Math.round(box.y + box.height * 0.36) };
const endPoint = { x: Math.round(box.x + box.width * 0.48), y: Math.round(box.y + box.height * 0.47) };
// Place the two line vertices with an explicit, settled motion before each button
// press. KiCad's GAL updates the active tool's world-space cursor from the
// asyncified pointer-move handler; a bare mouse.click() (move+down+up with no dwell)
// fires the button before that handler has run, so the vertex lands at a stale
// position or is dropped and no segment is committed. A short dwell after each move
// lets the position propagate. This is the one interaction whose commit has no
// JS-observable signal to poll (see the render-idle plan / eeschema draw-wires) —
// the dwells are the documented irreducible waits.
2026-06-04 13:20:55 +02:00
await page.mouse.move(startPoint.x, startPoint.y);
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
await page.waitForTimeout(350); // eslint-disable-line -- see comment above
2026-06-04 13:20:55 +02:00
await page.mouse.down();
await page.mouse.up();
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
await page.waitForTimeout(350); // eslint-disable-line -- see comment above
2026-06-04 13:20:55 +02:00
await page.mouse.move(endPoint.x, endPoint.y);
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
await page.waitForTimeout(350); // eslint-disable-line -- see comment above
2026-06-04 13:20:55 +02:00
await page.mouse.down();
await page.mouse.up();
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
await page.waitForTimeout(750); // eslint-disable-line -- see comment above
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
e2e/CI: dual-engine suites, per-engine screenshots, SwiftShader retired, prod web suite, CI-coverage gate Squash of experiment/ff-big-modules vs main. Big-module routing removed: native-EH shrank kicad_editor below SpiderMonkey's x86-64 code budget (runs 29355049705/29356152413 green on stock Firefox), so BIG_MODULE_SPECS routing and the baseline-only-JIT crutch are gone — kicad-firefox and kicad-chromium both run the full suite, with the module compiled the way real users' browsers compile it. Per-engine screenshots end to end: stableShot/shotPath write test-results/<engine>/<name>.png; baselines move to baseline-screenshots/{chromium,firefox}/ and the whole tools/screenshots pipeline (compare/promote/manifest/spec-map/changelog/Discord) keys on <engine>/<name>. Previously Firefox and Chromium renders of one spec overwrote each other and Firefox renders were never actually gated. Seeded from CI run 29421380806 (92 new firefox baselines, +24 chromium web-suite shots); manifest generated from the baseline tree. One merged playwright.config.ts (kicad/asyncify/coroutine/perf as projects); ~25 dead npm scripts dropped. The web suite is gated in CI for the first time ever (4 rotted specs fixed, 5 broken lib-bridge specs triaged as fixme in docs/features/web-e2e-rot/); cheap lint step after npm ci; last 26 blind-sleep violations fixed. SwiftShader retired: CI Chromium renders WebGL on ANGLE → Mesa llvmpipe (--use-gl=angle --use-angle=gl --ignore-gpu-blocklist; the blocklist flag is mandatory — llvmpipe is blocklisted and WebGL is silently unavailable without it) in BOTH configs. Under WORKERS=4 congestion SwiftShader transiently failed the first post-board-load draw and the recovery cascade ended in a silent permanent Cairo fallback — that engine flip was the "~1.2% changedRatio both directions" occ-export baseline flake. Validated 160/160 across two 80-repeat rigs; full analysis in docs/features/wx-parity-bugs/occ-export-context-eviction.md. Chromium baselines shift slightly on llvmpipe — promote once from the first green run. Deflakes the new coverage exposed: presence baselines settle before capture; presence fixtures declare current file formats; perf gets its own outputDir so CI evidence survives; occ-export settles the board paint before the export dialog; menu-item waits (waitForRenderedByLabel before clickMenuItem) in 4 specs + the TESTING.md rule. Web suite runs the PROD build, in parallel: webServer becomes backend `start` + the standalone's e2e:preview (build-preview.mjs: link-wasm → stash the public/wasm symlink aside during vite build, build-demo.mjs's move — then vite preview as the persistent server). The wasm middleware serves /wasm/* in preview and emits COOP/COEP/CORP itself (a pthread worker script's own response must carry COEP or Chrome kills it with ERR_BLOCKED_BY_RESPONSE). VITE_* flags bake at build time; VITE_ALLOW_USER_OVERRIDE joins turbo globalEnv. fullyParallel + default workers: 5.2m → 1.4m. Determinism fixes the parallel run exposed: shared-page specs become serial groups; locks.spec grabs alice's exact item via the new kicadCollabTestSelectByUuid hook (cross-tab "first footprint" order is not a ysync invariant); quit specs poll page.url() (quit supersedes its own navigation — NS_BINDING_ABORTED on Firefox). Suite: 51 passed / 12 skipped / 0 failed in 1.6m. CI-coverage gate (lint:ci-coverage): every tests/**/*.spec.ts must be reachable from the npm scripts the workflows invoke — scraped from .github/workflows/, resolved through package.json, coverage asked from playwright --list itself. Rules: uncovered-spec + orphan-project (with a documented LOCAL_ONLY_PROJECTS allowlist). Gating next to lint:determinism; 138 spec files / 13 projects accounted for. Product fixes kept from the investigations (reachable on real GPUs too): wx 7799fd1be5 — paint flags clear before dispatch + Invalidate always propagates; kicad 3dcfea5e45 — SwiftShader pass-boundary flush + per-instance font texture + first-frame GL-error drain (GAL recovery recovers instead of falling back to Cairo) + the user-facing eeschema switch navigates again under __EMSCRIPTEN__ (project-sync's FaceRegistered gate had rerouted it into the hidden sync player; caught by the newly-gated web suite). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018eUxiPApHgGiu9NFyQfhAq
2026-07-17 12:10:40 +02:00
const afterDrawing = await page.screenshot({ path: shotPath(page, 'pcbnew-draw-lines-02-after-drawing.png'), scale: 'css' });
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
const diffRegion: DiffRegion = {
x: Math.max(0, Math.min(startPoint.x, endPoint.x) - 24),
y: Math.max(0, Math.min(startPoint.y, endPoint.y) - 24),
width: Math.abs(endPoint.x - startPoint.x) + 48,
height: Math.abs(endPoint.y - startPoint.y) + 48,
};
const drawingDiff = await compareScreenshots(page, afterToolClick, afterDrawing, diffRegion);
expect(drawingDiff.diffPixels).toBeGreaterThan(120);
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
expect(drawingDiff.diffRatio).toBeGreaterThan(0.005);
expect(drawingDiff.meanChannelDiff).toBeGreaterThan(0.4);
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
test(determinism): deterministic waits + stableShot screenshots; drop blind sleeps/ifs/retries Make the Playwright e2e + kicad suites deterministic so screenshot flake stops tracing to timing races. - Blind page.waitForTimeout -> condition waits (expect.poll, web-first assertions, waitUntil) + readiness helpers (waitForWxApp, waitForCanvasApp). Remaining sleeps are documented interaction dwells (annotated). - Defensive "if element exists" branches -> loud asserts; label-fallback chains -> normalized clickMenuItemByText. First-run wizard for/if loops removed by seeding calculator/gerbview/pcbnew HTMLs. - Screenshots: new stableShot(page, name) settles the render in-page (canvas hash over rAF) then writes a raw PNG to test-results/ for the existing offline gate (tools/screenshots vs baseline-screenshots). Replaces toHaveScreenshot, which did inline compare + its own baselines and had decoupled the specs from the real gate. scale:'css' pinned. - retries: 0 in both configs. - Guard: tests/tools/lint-determinism.ts (npm run lint:determinism) bans blind sleeps / toHaveScreenshot / inline retries / swallowed catches in specs; documented exceptions carry a marker. Rules in tests/TESTING.md. Assertions, coverage, and renders unchanged (semantic-equivalence reviewed; captures pixel-identical modulo inherent timer/timestamp/3d-raytrace variance). Both suites green at retries:0 (e2e 340, kicad 92); ~35-61% faster. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01BVX1pHMvRPYHdp6ZfEawrk
2026-07-07 10:50:24 +02:00
const realErrors = testLogger.errors.slice(baselineErrorCount).filter((error) => !error.includes('favicon'));
wip: nested-asyncify fix, wxAuiToolBar registration, tests, research docs Main-repo side of a multi-part WIP covering the KiCad WASM tool-selection and nested-Asyncify work. Submodule commits are in kicad@f6e9239aaa (libcontext hygiene) and wxwidgets@bb80f91e8b (auibar registration + dialog diagnostics). ## scripts/common/inject-dyncall-shims.sh Wrap Asyncify.handleSleep / allocateData to save-and-restore Asyncify.currData around each EM_ASYNC_JS sleep. This fixes the nested Asyncify collision where a fiber swap that fired during a modal's event loop clobbered currData, and the modal's later doRewind used the fiber's buffer and hit "RuntimeError: index out of bounds". Root cause documented as Emscripten Issue #9153 (wontfix upstream). Diagnostic-rewind logging (forcedBottomOfCallStack, callStack traces) is retained to help future debugging of Asyncify state corruption. ## tests/ - tests/playwright-kicad.config.ts: add `channel: 'chrome'` for the chromium project so --project=chromium --headed uses system Chrome (real GPU) instead of SwiftShader on ARM Mac. Also switch trace to retain-on-failure + screenshot on-failure for easier E2E debugging. - tests/kicad/pcbnew.spec.ts: replace `tool.checked` assertions with a label-suffix check (`[checked]`) since our auibar registration encodes checked state in the label (no schema change to the registry). - tests/apps/Makefile.wasm: add `coroutine-nested` build target + include it in the all: list. - tests/apps/standalone/coroutine/: kicad_coroutine_harness.h + test app reproducing KiCad COROUTINE semantics against real libcontext. - tests/apps/standalone/coroutine-nested/: nested_test.cpp reproduces the EM_ASYNC_JS-modal + fiber-swap nesting bug in isolation. 8 scenarios from baseline_modal_alone through nested_fibers_inside_modal. - tests/e2e/coroutine.spec.ts + coroutine-nested.spec.ts: Playwright specs that load the standalone apps and assert all case cases pass via [COROUTINE_TEST] SUMMARY log parsing. ## research/ and features/browser-tools/ Three background docs capturing the investigation trajectory: - features/browser-tools/0001-kicad-wasm-tool-activation-investigation.md Early investigation: why tools don't activate; initial dynCall-empty- callback hypothesis. - features/browser-tools/0002-wasm-coroutine-deep-dive.md Deep dive on Asyncify internals, fiber API, QEMU's coroutine-wasm reference implementation. - features/browser-tools/0003-wxauitoolbar-registration-fix.md The narrow fix: why wxAuiToolBar needs a registration block, where to add it, what the fallback plan is. - research/threading_1.md: corrected root-cause analysis after reading runtime logs — nested-Asyncify currData collision, Emscripten #9153. - research/threading_2.md: extended research on alternative approaches (JSPI/WasmFX/state-machines) and why they don't help here. ## Submodule pointer updates kicad: f6e9239aaa (wip: libcontext WASM hygiene cleanup) wxwidgets: bb80f91e8b (wip: wxAuiToolBar element-registry registration + dialog diagnostics) ## Open threads not yet in scope - Firefox/Chrome divergent behavior: "indirect call signature mismatch" traps in Firefox vs renderer crash in system Chrome (tracked in plans/peaceful-hugging-pnueli.md and the research docs). - E2E pixel-diff for Draw Lines fails because the test's diff region does not cover where the line is actually drawn; tool activation works, the line is visible in test-results/pcbnew-draw-lines-02-after-drawing.png. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-04-21 13:58:16 +02:00
expect(realErrors).toEqual([]);
});
});