pcbjam/scripts/kicad/build-kicad-target.sh

770 lines
38 KiB
Shell
Raw Normal View History

#!/bin/bash
feat: pl_editor WASM port + browser file dialog fixes Brings up KiCad's pagelayout_editor (drawing-sheet editor) in the browser, to roughly the same "boots, canvas visible, partially usable in-session" level as the existing pcbnew/eeschema/calculator ports. Build: - docker/build.sh: add pl_editor to the unified app dispatch (case, subdir map, all-loop). - scripts/kicad/build-kicad-target.sh: add pl_editor to the case; upstream target name pl_editor under source subdir pagelayout_editor. - scripts/kicad/build-pl_editor.sh: 7-line thin wrapper matching the pcbnew/eeschema/calculator pattern. - tests/scripts/setup-kicad-wasm.sh: copy_app pl_editor. App glue: - wasm/stubs/nl_pl_editor_plugin_stub.cpp: no-op SpaceMouse plugin so pl_editor_frame.cpp's NL_PL_EDITOR_PLUGIN symbols resolve. Mirrors nl_pcbnew_plugin_stub.cpp. - tests/apps/kicad/pl_editor.html: browser shell. preRun creates /home/kicad and FS.chdir there so file dialogs land somewhere friendly instead of MEMFS root (/dev/, /proc/, etc.). E2E coverage: - tests/kicad/pl_editor.spec.ts: 5 tests — smoke (canvas, no abort), wizard, File menu has Open/Save As, file-dialog folder-navigation regression, canvas + toolbar metrics. - tests/e2e/filedialog-folder-nav.spec.ts: wxWidgets-level twin of the regression test (exercises the underlying widget directly via the standalone filedialog_test app). Submodule bumps: - kicad → feature/pl-editor (WASM gating in pagelayout_editor's CMakeLists + navlib stub). - wxwidgets → feature/pl-editor (wxGenericFileDialog::OnOk navigates into selected directories; wasm/mouse.cpp emits wxEVT_LEFT_DCLICK via timestamp-based double-click detection — the latter benefits every wxWidgets-WASM app). See features/pl-editor/ for the design doc + per-repo diff patches. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-01 10:30:56 +02:00
# Build a KiCad app (pcbnew, eeschema, calculator, pl_editor) for WebAssembly.
#
# Usage:
# ./scripts/kicad/build-kicad-target.sh <app> [options]
#
# Args:
feat: pl_editor WASM port + browser file dialog fixes Brings up KiCad's pagelayout_editor (drawing-sheet editor) in the browser, to roughly the same "boots, canvas visible, partially usable in-session" level as the existing pcbnew/eeschema/calculator ports. Build: - docker/build.sh: add pl_editor to the unified app dispatch (case, subdir map, all-loop). - scripts/kicad/build-kicad-target.sh: add pl_editor to the case; upstream target name pl_editor under source subdir pagelayout_editor. - scripts/kicad/build-pl_editor.sh: 7-line thin wrapper matching the pcbnew/eeschema/calculator pattern. - tests/scripts/setup-kicad-wasm.sh: copy_app pl_editor. App glue: - wasm/stubs/nl_pl_editor_plugin_stub.cpp: no-op SpaceMouse plugin so pl_editor_frame.cpp's NL_PL_EDITOR_PLUGIN symbols resolve. Mirrors nl_pcbnew_plugin_stub.cpp. - tests/apps/kicad/pl_editor.html: browser shell. preRun creates /home/kicad and FS.chdir there so file dialogs land somewhere friendly instead of MEMFS root (/dev/, /proc/, etc.). E2E coverage: - tests/kicad/pl_editor.spec.ts: 5 tests — smoke (canvas, no abort), wizard, File menu has Open/Save As, file-dialog folder-navigation regression, canvas + toolbar metrics. - tests/e2e/filedialog-folder-nav.spec.ts: wxWidgets-level twin of the regression test (exercises the underlying widget directly via the standalone filedialog_test app). Submodule bumps: - kicad → feature/pl-editor (WASM gating in pagelayout_editor's CMakeLists + navlib stub). - wxwidgets → feature/pl-editor (wxGenericFileDialog::OnOk navigates into selected directories; wasm/mouse.cpp emits wxEVT_LEFT_DCLICK via timestamp-based double-click detection — the latter benefits every wxWidgets-WASM app). See features/pl-editor/ for the design doc + per-repo diff patches. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-01 10:30:56 +02:00
# <app> pcbnew | eeschema | calculator | pl_editor (required)
#
# Options:
# --full Full clean rebuild (dependencies + KiCad)
# --clean-kicad Clean only KiCad build directory (not deps)
# --build-deps Build dependencies (default: skip)
# --debug Build with debug symbols (default)
# --release Build optimized without debug symbols
# --diag=... Diagnostic preprocessor flags (gal, coroutine, ctor, all)
# -j N Parallel compilation jobs (default: 1)
#
refactor(build): fold calculator into unified docker/build.sh dispatch Convert the calculator port (commit 0464470) from the parallel/copy-paste pattern to the dispatch pattern used by pcbnew + eeschema. - docker/build.sh: <app> is now required (no pcbnew default); missing, unknown, and --help paths all print "pcbnew | eeschema | calculator | all". Adds calculator to build_app() via kicad_subdir_for() (pcb_calculator inner subdir, calculator.* output basename). Also fixes the EMSDK=/emsdkkicad-wasm-builder typo that snuck in during the calc merge. - scripts/kicad/build-kicad-target.sh: accepts calculator; introduces KICAD_TARGET (pcb_calculator for calc, == APP_NAME otherwise) used for the make target, embind include path, and final-log line. - scripts/kicad/build-calculator.sh: 305-line copy of build-pcbnew.sh collapsed to a 7-line wrapper around build-kicad-target.sh. - tests/scripts/setup-kicad-wasm.sh: copy_app calculator added with the same pcb_calculator subdir mapping for the docker-volume fallback path. - tests/package.json: test:calculator* routed through the shared playwright-kicad.config.ts kicad/calculator.spec.ts (mirrors eeschema). - Delete: docker/build-calculator.sh, tests/scripts/setup-calculator-wasm.sh, tests/playwright-calculator.config.ts. Bumps wxwidgets d1d1627 -> 6fb2eac (origin/wasm-port). The new sha includes "unic/combobox: add GetCurrentSelection() inline default" which calc's kicad/pcb_calculator/widgets/unit_selector.cpp needs to compile. Without this bump the unified dispatch would expose calc as a buildable target but the build itself would fail. Verified: build.sh all completes clean across all three apps; pcbnew/eeschema e2e pass; calc compiles and launches (test stability separate from this refactor). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-29 12:01:44 +02:00
# Each app builds into its own tree: build-wasm/kicad-<app>/.
# Per-app extras live alongside generic stubs:
# - wasm/bindings/<app>_embind.cpp (optional)
# - wasm/stubs/<app>_frame_stub.cpp (optional, app-specific stubs)
# - wasm/stubs/<app>_scripting_stub.cpp (optional, app-specific scripting stubs)
refactor(build): fold calculator into unified docker/build.sh dispatch Convert the calculator port (commit 0464470) from the parallel/copy-paste pattern to the dispatch pattern used by pcbnew + eeschema. - docker/build.sh: <app> is now required (no pcbnew default); missing, unknown, and --help paths all print "pcbnew | eeschema | calculator | all". Adds calculator to build_app() via kicad_subdir_for() (pcb_calculator inner subdir, calculator.* output basename). Also fixes the EMSDK=/emsdkkicad-wasm-builder typo that snuck in during the calc merge. - scripts/kicad/build-kicad-target.sh: accepts calculator; introduces KICAD_TARGET (pcb_calculator for calc, == APP_NAME otherwise) used for the make target, embind include path, and final-log line. - scripts/kicad/build-calculator.sh: 305-line copy of build-pcbnew.sh collapsed to a 7-line wrapper around build-kicad-target.sh. - tests/scripts/setup-kicad-wasm.sh: copy_app calculator added with the same pcb_calculator subdir mapping for the docker-volume fallback path. - tests/package.json: test:calculator* routed through the shared playwright-kicad.config.ts kicad/calculator.spec.ts (mirrors eeschema). - Delete: docker/build-calculator.sh, tests/scripts/setup-calculator-wasm.sh, tests/playwright-calculator.config.ts. Bumps wxwidgets d1d1627 -> 6fb2eac (origin/wasm-port). The new sha includes "unic/combobox: add GetCurrentSelection() inline default" which calc's kicad/pcb_calculator/widgets/unit_selector.cpp needs to compile. Without this bump the unified dispatch would expose calc as a buildable target but the build itself would fail. Verified: build.sh all completes clean across all three apps; pcbnew/eeschema e2e pass; calc compiles and launches (test stability separate from this refactor). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-29 12:01:44 +02:00
#
# Most apps use the same name for the user-facing app, the CMake target, and
# the source subdirectory. Calculator is the exception: app=calculator but the
# upstream target and source subdir are both pcb_calculator (the OUTPUT_NAME
# property in pcb_calculator/CMakeLists.txt emits calculator.{js,wasm}).
feat: pl_editor WASM port + browser file dialog fixes Brings up KiCad's pagelayout_editor (drawing-sheet editor) in the browser, to roughly the same "boots, canvas visible, partially usable in-session" level as the existing pcbnew/eeschema/calculator ports. Build: - docker/build.sh: add pl_editor to the unified app dispatch (case, subdir map, all-loop). - scripts/kicad/build-kicad-target.sh: add pl_editor to the case; upstream target name pl_editor under source subdir pagelayout_editor. - scripts/kicad/build-pl_editor.sh: 7-line thin wrapper matching the pcbnew/eeschema/calculator pattern. - tests/scripts/setup-kicad-wasm.sh: copy_app pl_editor. App glue: - wasm/stubs/nl_pl_editor_plugin_stub.cpp: no-op SpaceMouse plugin so pl_editor_frame.cpp's NL_PL_EDITOR_PLUGIN symbols resolve. Mirrors nl_pcbnew_plugin_stub.cpp. - tests/apps/kicad/pl_editor.html: browser shell. preRun creates /home/kicad and FS.chdir there so file dialogs land somewhere friendly instead of MEMFS root (/dev/, /proc/, etc.). E2E coverage: - tests/kicad/pl_editor.spec.ts: 5 tests — smoke (canvas, no abort), wizard, File menu has Open/Save As, file-dialog folder-navigation regression, canvas + toolbar metrics. - tests/e2e/filedialog-folder-nav.spec.ts: wxWidgets-level twin of the regression test (exercises the underlying widget directly via the standalone filedialog_test app). Submodule bumps: - kicad → feature/pl-editor (WASM gating in pagelayout_editor's CMakeLists + navlib stub). - wxwidgets → feature/pl-editor (wxGenericFileDialog::OnOk navigates into selected directories; wasm/mouse.cpp emits wxEVT_LEFT_DCLICK via timestamp-based double-click detection — the latter benefits every wxWidgets-WASM app). See features/pl-editor/ for the design doc + per-repo diff patches. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-06-01 10:30:56 +02:00
# pl_editor is the standard case but its source subdir is pagelayout_editor.
set -e
if [ -z "$1" ]; then
feat(wasm): kicad_editor — merge the pcbnew+eeschema kifaces into ONE bundle (Part 2) All four editors (PCB / Footprint / Schematic / Symbol) are now runtime --frame choices of a single kicad_editor.wasm (178 MB at -O1 vs 147+82 separate; shared wx/common/boost linked once). One editor per page load, as before; frames pcb / fpedit / sch / symedit. - wasm/editor/: the merged executable target (single_top + both kiface library sets, whole-archive pcbcommon) + the safety-net focus-walk Kiface() dispatch TU. Gated by KICAD_WASM_MERGED_EDITOR (kicad submodule bump carries the fork side: per-engine Kiface/getter binding + ODR renames + dual-kiface launcher). - wasm/bindings/: per-editor collab entries renamed pcbCollab*/schCollab* (JS names unchanged); duplicate kicadOpenFile/kicadCollabOnSave + shared-name registrations guarded behind KICAD_MERGED_EMBIND; new kicad_editor_embind.cpp registers each shared JS name once, dispatching on the live frame. - Build: kicad_editor app (build wrapper, target case arms, 3-object embind compile with the ABI-critical flags, STUB_APP=pcbnew); docker/build.sh "all" = kicad_editor calculator pl_editor gerbview (pcbnew/eeschema stay as explicit debug apps); scripts/kicad/audit-merged-symbols.sh = repeatable ODR-collision audit (run on kicad bumps). - Frontend: Bundle type (bundle ≠ tool); TOOL_BUNDLE maps all four editors to kicad_editor; explicit --frame tokens for pcbnew (pcb) and eeschema (sch); publish list = the 4 real bundles. - Tests/CI: five harnesses load kicad_editor.js with explicit frame tokens; PCBNEW_FAMILY_SPECS renamed BIG_MODULE_SPECS + the 8 eeschema-family specs (they now boot the merged module — SpiderMonkey x86 CI OOM routing); frame-runtime spec covers all four frames from the one bundle. Validated so far: frame-runtime 4/4 (each frame boots with the right title, no aborts, no duplicate embind registration); 24-spec merged-module regression green; 3D raytracer renders. Known pre-existing failure: 3d-viewer title-bar drag deadlock, fixed on main by 7630c7e (2N+8 pthread pre-warm) — picked up by the follow-up rebase. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 14:48:11 +02:00
echo "Error: missing <app> argument (kicad_editor | pcbnew | eeschema | calculator | pl_editor | gerbview)" >&2
exit 1
fi
APP_NAME="$1"
shift
2026-06-02 13:35:37 +02:00
# KICAD_TARGET: the CMake/make target name.
# KICAD_SUBDIR: the source/build subdirectory the target's artifacts land in.
# Most apps share all three names; the exceptions:
# - calculator: target+subdir are both pcb_calculator (OUTPUT_NAME=calculator)
# - pl_editor: subdir is pagelayout_editor (upstream source dir name)
# The footprint/symbol editors have no target of their own — the pcbnew/eeschema
# bundle opens them at runtime via single_top.cpp's --frame flag.
case "$APP_NAME" in
feat(wasm): kicad_editor — merge the pcbnew+eeschema kifaces into ONE bundle (Part 2) All four editors (PCB / Footprint / Schematic / Symbol) are now runtime --frame choices of a single kicad_editor.wasm (178 MB at -O1 vs 147+82 separate; shared wx/common/boost linked once). One editor per page load, as before; frames pcb / fpedit / sch / symedit. - wasm/editor/: the merged executable target (single_top + both kiface library sets, whole-archive pcbcommon) + the safety-net focus-walk Kiface() dispatch TU. Gated by KICAD_WASM_MERGED_EDITOR (kicad submodule bump carries the fork side: per-engine Kiface/getter binding + ODR renames + dual-kiface launcher). - wasm/bindings/: per-editor collab entries renamed pcbCollab*/schCollab* (JS names unchanged); duplicate kicadOpenFile/kicadCollabOnSave + shared-name registrations guarded behind KICAD_MERGED_EMBIND; new kicad_editor_embind.cpp registers each shared JS name once, dispatching on the live frame. - Build: kicad_editor app (build wrapper, target case arms, 3-object embind compile with the ABI-critical flags, STUB_APP=pcbnew); docker/build.sh "all" = kicad_editor calculator pl_editor gerbview (pcbnew/eeschema stay as explicit debug apps); scripts/kicad/audit-merged-symbols.sh = repeatable ODR-collision audit (run on kicad bumps). - Frontend: Bundle type (bundle ≠ tool); TOOL_BUNDLE maps all four editors to kicad_editor; explicit --frame tokens for pcbnew (pcb) and eeschema (sch); publish list = the 4 real bundles. - Tests/CI: five harnesses load kicad_editor.js with explicit frame tokens; PCBNEW_FAMILY_SPECS renamed BIG_MODULE_SPECS + the 8 eeschema-family specs (they now boot the merged module — SpiderMonkey x86 CI OOM routing); frame-runtime spec covers all four frames from the one bundle. Validated so far: frame-runtime 4/4 (each frame boots with the right title, no aborts, no duplicate embind registration); 24-spec merged-module regression green; 3D raytracer renders. Known pre-existing failure: 3d-viewer title-bar drag deadlock, fixed on main by 7630c7e (2N+8 pthread pre-warm) — picked up by the follow-up rebase. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 14:48:11 +02:00
kicad_editor)
# Merged pcbnew+eeschema image (editor-unification Part 2): one executable
# linking BOTH kifaces, frame chosen at runtime by --frame (pcb/fpedit/sch/
# symedit). Target lives in wasm/editor/ (added by the fork's top-level
# CMakeLists under -DKICAD_WASM_MERGED_EDITOR=ON); its binary dir doubles as
# the artifact subdir.
KICAD_TARGET="kicad_editor"
KICAD_SUBDIR="kicad_editor"
;;
2026-06-03 14:23:57 +02:00
pcbnew|eeschema|gerbview)
refactor(build): fold calculator into unified docker/build.sh dispatch Convert the calculator port (commit 0464470) from the parallel/copy-paste pattern to the dispatch pattern used by pcbnew + eeschema. - docker/build.sh: <app> is now required (no pcbnew default); missing, unknown, and --help paths all print "pcbnew | eeschema | calculator | all". Adds calculator to build_app() via kicad_subdir_for() (pcb_calculator inner subdir, calculator.* output basename). Also fixes the EMSDK=/emsdkkicad-wasm-builder typo that snuck in during the calc merge. - scripts/kicad/build-kicad-target.sh: accepts calculator; introduces KICAD_TARGET (pcb_calculator for calc, == APP_NAME otherwise) used for the make target, embind include path, and final-log line. - scripts/kicad/build-calculator.sh: 305-line copy of build-pcbnew.sh collapsed to a 7-line wrapper around build-kicad-target.sh. - tests/scripts/setup-kicad-wasm.sh: copy_app calculator added with the same pcb_calculator subdir mapping for the docker-volume fallback path. - tests/package.json: test:calculator* routed through the shared playwright-kicad.config.ts kicad/calculator.spec.ts (mirrors eeschema). - Delete: docker/build-calculator.sh, tests/scripts/setup-calculator-wasm.sh, tests/playwright-calculator.config.ts. Bumps wxwidgets d1d1627 -> 6fb2eac (origin/wasm-port). The new sha includes "unic/combobox: add GetCurrentSelection() inline default" which calc's kicad/pcb_calculator/widgets/unit_selector.cpp needs to compile. Without this bump the unified dispatch would expose calc as a buildable target but the build itself would fail. Verified: build.sh all completes clean across all three apps; pcbnew/eeschema e2e pass; calc compiles and launches (test stability separate from this refactor). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-29 12:01:44 +02:00
KICAD_TARGET="$APP_NAME"
2026-06-02 13:35:37 +02:00
KICAD_SUBDIR="$APP_NAME"
;;
pl_editor)
KICAD_TARGET="pl_editor"
KICAD_SUBDIR="pagelayout_editor"
refactor(build): fold calculator into unified docker/build.sh dispatch Convert the calculator port (commit 0464470) from the parallel/copy-paste pattern to the dispatch pattern used by pcbnew + eeschema. - docker/build.sh: <app> is now required (no pcbnew default); missing, unknown, and --help paths all print "pcbnew | eeschema | calculator | all". Adds calculator to build_app() via kicad_subdir_for() (pcb_calculator inner subdir, calculator.* output basename). Also fixes the EMSDK=/emsdkkicad-wasm-builder typo that snuck in during the calc merge. - scripts/kicad/build-kicad-target.sh: accepts calculator; introduces KICAD_TARGET (pcb_calculator for calc, == APP_NAME otherwise) used for the make target, embind include path, and final-log line. - scripts/kicad/build-calculator.sh: 305-line copy of build-pcbnew.sh collapsed to a 7-line wrapper around build-kicad-target.sh. - tests/scripts/setup-kicad-wasm.sh: copy_app calculator added with the same pcb_calculator subdir mapping for the docker-volume fallback path. - tests/package.json: test:calculator* routed through the shared playwright-kicad.config.ts kicad/calculator.spec.ts (mirrors eeschema). - Delete: docker/build-calculator.sh, tests/scripts/setup-calculator-wasm.sh, tests/playwright-calculator.config.ts. Bumps wxwidgets d1d1627 -> 6fb2eac (origin/wasm-port). The new sha includes "unic/combobox: add GetCurrentSelection() inline default" which calc's kicad/pcb_calculator/widgets/unit_selector.cpp needs to compile. Without this bump the unified dispatch would expose calc as a buildable target but the build itself would fail. Verified: build.sh all completes clean across all three apps; pcbnew/eeschema e2e pass; calc compiles and launches (test stability separate from this refactor). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-29 12:01:44 +02:00
;;
calculator)
KICAD_TARGET="pcb_calculator"
2026-06-02 13:35:37 +02:00
KICAD_SUBDIR="pcb_calculator"
;;
kicad_tools)
# Merged headless CLI (pcbjam-mcp 0001 tier 3a): both dieted kifaces —
# .lib conversion (--convert-lib), --lint, --erc, --netlist, --bom,
# --plot, --drc in one node image. (Supersedes the retired standalone
# sym_convert / pcb_convert apps.) Its add_executable lives in
# wasm/tools/ (added by the fork's top-level CMakeLists under
# -DKICAD_TOOLS_WASM=ON); the binary dir doubles as the artifact
# subdir of its own kicad-kicad_tools tree. A headless CLI has no 3D
# viewer: skip the wasm/gl1 FFP shim compile (it needs glm, which the
# libs sysroot doesn't guarantee) unless the caller forces it.
KICAD_TARGET="kicad_tools"
KICAD_SUBDIR="kicad_tools"
BUILD_3D_VIEWER="${BUILD_3D_VIEWER:-OFF}"
;;
feat(wasm): occ-split — lazy occ_service worker; kicad_editor drops OCC (−31%) Move OpenCASCADE out of the merged editor image into occ_service: a separate emscripten module (-sASYNCIFY=0, MODULARIZE, in-container -Oz finalize, 2N+8 pre-warmed pthread pool) booted lazily in a dedicated Web Worker on the first STEP export or STEP/IGES model parse. kicad_editor.wasm ~190 MB -> 130 MB; sessions that never touch OCC never fetch its 57 MB. STEP export works in the browser for the first time: the unchanged desktop dialog runs EXPORTER_STEP, whose wasm shadow suspends into globalThis.occService and the export bytes go straight to a browser download (never entering the editor heap). STEP/IGES 3D models parse in the worker via the oce shadow (S3D WriteCache/ReadCache wire). - wasm/occ-service/: service CMake target (hooked from the kicad fork's top-level CMakeLists, wasm/editor pattern), embind entry (occExport/occLoadModel), wxConfig pre-js. - wasm/stubs/{exporter_step,oce_plugin}_stub.cpp: EM_ASYNC_JS worker bridges (callee-shadowing; no caller #ifdefs). - web/standalone: provider installed whenever the kicad_editor bundle boots (cross-face safe); ONE shared worker-boot source occ-worker.js (vite ?raw; the e2e stub reads the same file) — blob worker with locateFile absolutized against the glue URL; export download-name guard. - deps: OCC builds with RapidJSON so its glTF/GLB writer exists — pinned to the vcpkg master snapshot 2025-02-26 (24b5e7a8b27f), the same code official KiCad consumes via vcpkg.json's opencascade[rapidjson]; rapidjson's latest tag (v1.1.0, 2016) is ill-formed under modern clang. - tests: occ-export dialog e2e (lazy-fetch boundary + STEP download bytes), occ-probe incl. a 9-format matrix (step/stpz/brep/xao/ply/stl/glb/u3d/pdf), 3d-viewer-models hard-asserts the worker parse; occ provider stub installed ambiently by the kicad fixtures. Validated against desktop kicad-cli 10.0.4: geometric exact equality (bbox delta 0 um, volume delta 0.0000%) for STEP/GLB/STL/BREP/STPZ across three boards and option sweeps — with desktop OCC 7.9 vs wasm OCC 7.8; PLY/XAO/PDF structurally equal; U3D same-size (quantizer float LSBs differ). Full kicad e2e green on Firefox and Chromium; standalone verified end to end (lazy fetch only on the Export click; export.step 60,628 B ISO-10303-21; loadModel 700 KB STEP -> 569 KB scenegraph cache). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:39:58 +02:00
occ_service)
# Standalone OpenCASCADE 3D service (worker embind module). Target lives
# in wasm/occ-service/ (added by the fork's top-level CMakeLists under
# -DKICAD_OCC_SERVICE_WASM=ON); like kicad_editor, its binary dir
# doubles as the artifact subdir of its own kicad-occ_service tree.
KICAD_TARGET="occ_service"
KICAD_SUBDIR="occ_service"
;;
*)
echo "Error: unknown app '$APP_NAME' (expected: kicad_editor | pcbnew | eeschema | calculator | pl_editor | gerbview | kicad_tools | occ_service)" >&2
exit 1
;;
esac
# Which app's embind bindings to compile + link. Most apps use their own.
# NOTE kicad_tools deliberately does NOT reuse the editors' embind objects even
# though it links both kifaces: each kiface references exactly one embind
# symbol (kicadCollabOnSave), and linking a real bindings TU for it roots the
# whole editor surface from .init_array (~4x binary size — ysync 0009 size
# research). Its own wasm/bindings/kicad_tools_embind.cpp provides a no-op hook.
case "$APP_NAME" in
# occ_service links the pcbnew kiface objects → pcbnew's embind object (its
# own embind entry points live in occ_service_main.cpp, compiled inside the
# CMake target).
feat(wasm): occ-split — lazy occ_service worker; kicad_editor drops OCC (−31%) Move OpenCASCADE out of the merged editor image into occ_service: a separate emscripten module (-sASYNCIFY=0, MODULARIZE, in-container -Oz finalize, 2N+8 pre-warmed pthread pool) booted lazily in a dedicated Web Worker on the first STEP export or STEP/IGES model parse. kicad_editor.wasm ~190 MB -> 130 MB; sessions that never touch OCC never fetch its 57 MB. STEP export works in the browser for the first time: the unchanged desktop dialog runs EXPORTER_STEP, whose wasm shadow suspends into globalThis.occService and the export bytes go straight to a browser download (never entering the editor heap). STEP/IGES 3D models parse in the worker via the oce shadow (S3D WriteCache/ReadCache wire). - wasm/occ-service/: service CMake target (hooked from the kicad fork's top-level CMakeLists, wasm/editor pattern), embind entry (occExport/occLoadModel), wxConfig pre-js. - wasm/stubs/{exporter_step,oce_plugin}_stub.cpp: EM_ASYNC_JS worker bridges (callee-shadowing; no caller #ifdefs). - web/standalone: provider installed whenever the kicad_editor bundle boots (cross-face safe); ONE shared worker-boot source occ-worker.js (vite ?raw; the e2e stub reads the same file) — blob worker with locateFile absolutized against the glue URL; export download-name guard. - deps: OCC builds with RapidJSON so its glTF/GLB writer exists — pinned to the vcpkg master snapshot 2025-02-26 (24b5e7a8b27f), the same code official KiCad consumes via vcpkg.json's opencascade[rapidjson]; rapidjson's latest tag (v1.1.0, 2016) is ill-formed under modern clang. - tests: occ-export dialog e2e (lazy-fetch boundary + STEP download bytes), occ-probe incl. a 9-format matrix (step/stpz/brep/xao/ply/stl/glb/u3d/pdf), 3d-viewer-models hard-asserts the worker parse; occ provider stub installed ambiently by the kicad fixtures. Validated against desktop kicad-cli 10.0.4: geometric exact equality (bbox delta 0 um, volume delta 0.0000%) for STEP/GLB/STL/BREP/STPZ across three boards and option sweeps — with desktop OCC 7.9 vs wasm OCC 7.8; PLY/XAO/PDF structurally equal; U3D same-size (quantizer float LSBs differ). Full kicad e2e green on Firefox and Chromium; standalone verified end to end (lazy fetch only on the Export click; export.step 60,628 B ISO-10303-21; loadModel 700 KB STEP -> 569 KB scenegraph cache). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:39:58 +02:00
occ_service) EMBIND_APP="pcbnew" ;;
*) EMBIND_APP="$APP_NAME" ;;
esac
# Embind linker support (--bind). kicad_tools has no bindings at all (see
# above) — dropping --bind keeps the embind JS/native runtime out entirely.
case "$APP_NAME" in
kicad_tools) EMBIND_LINK_FLAG="" ;;
*) EMBIND_LINK_FLAG="--bind" ;;
esac
# Which app's WASM stub libraries (scripting/frame placeholders) to link.
# kicad_editor links pcbnew's kiface objects, which reference the action-plugin
# scripting placeholders (pcbnewGet*); eeschema's frame stub arrives via CMake
# (target_sources on eeschema_kiface_objects), not this path.
case "$APP_NAME" in
# kicad_tools links both kifaces: eeschema's frame stub comes through this
# path, pcbnew's via CMake PCBNEW_WASM_STUBS.
kicad_tools) STUB_APP="eeschema" ;;
feat(wasm): kicad_editor — merge the pcbnew+eeschema kifaces into ONE bundle (Part 2) All four editors (PCB / Footprint / Schematic / Symbol) are now runtime --frame choices of a single kicad_editor.wasm (178 MB at -O1 vs 147+82 separate; shared wx/common/boost linked once). One editor per page load, as before; frames pcb / fpedit / sch / symedit. - wasm/editor/: the merged executable target (single_top + both kiface library sets, whole-archive pcbcommon) + the safety-net focus-walk Kiface() dispatch TU. Gated by KICAD_WASM_MERGED_EDITOR (kicad submodule bump carries the fork side: per-engine Kiface/getter binding + ODR renames + dual-kiface launcher). - wasm/bindings/: per-editor collab entries renamed pcbCollab*/schCollab* (JS names unchanged); duplicate kicadOpenFile/kicadCollabOnSave + shared-name registrations guarded behind KICAD_MERGED_EMBIND; new kicad_editor_embind.cpp registers each shared JS name once, dispatching on the live frame. - Build: kicad_editor app (build wrapper, target case arms, 3-object embind compile with the ABI-critical flags, STUB_APP=pcbnew); docker/build.sh "all" = kicad_editor calculator pl_editor gerbview (pcbnew/eeschema stay as explicit debug apps); scripts/kicad/audit-merged-symbols.sh = repeatable ODR-collision audit (run on kicad bumps). - Frontend: Bundle type (bundle ≠ tool); TOOL_BUNDLE maps all four editors to kicad_editor; explicit --frame tokens for pcbnew (pcb) and eeschema (sch); publish list = the 4 real bundles. - Tests/CI: five harnesses load kicad_editor.js with explicit frame tokens; PCBNEW_FAMILY_SPECS renamed BIG_MODULE_SPECS + the 8 eeschema-family specs (they now boot the merged module — SpiderMonkey x86 CI OOM routing); frame-runtime spec covers all four frames from the one bundle. Validated so far: frame-runtime 4/4 (each frame boots with the right title, no aborts, no duplicate embind registration); 24-spec merged-module regression green; 3D raytracer renders. Known pre-existing failure: 3d-viewer title-bar drag deadlock, fixed on main by 7630c7e (2N+8 pthread pre-warm) — picked up by the follow-up rebase. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 14:48:11 +02:00
kicad_editor) STUB_APP="pcbnew" ;;
feat(wasm): occ-split — lazy occ_service worker; kicad_editor drops OCC (−31%) Move OpenCASCADE out of the merged editor image into occ_service: a separate emscripten module (-sASYNCIFY=0, MODULARIZE, in-container -Oz finalize, 2N+8 pre-warmed pthread pool) booted lazily in a dedicated Web Worker on the first STEP export or STEP/IGES model parse. kicad_editor.wasm ~190 MB -> 130 MB; sessions that never touch OCC never fetch its 57 MB. STEP export works in the browser for the first time: the unchanged desktop dialog runs EXPORTER_STEP, whose wasm shadow suspends into globalThis.occService and the export bytes go straight to a browser download (never entering the editor heap). STEP/IGES 3D models parse in the worker via the oce shadow (S3D WriteCache/ReadCache wire). - wasm/occ-service/: service CMake target (hooked from the kicad fork's top-level CMakeLists, wasm/editor pattern), embind entry (occExport/occLoadModel), wxConfig pre-js. - wasm/stubs/{exporter_step,oce_plugin}_stub.cpp: EM_ASYNC_JS worker bridges (callee-shadowing; no caller #ifdefs). - web/standalone: provider installed whenever the kicad_editor bundle boots (cross-face safe); ONE shared worker-boot source occ-worker.js (vite ?raw; the e2e stub reads the same file) — blob worker with locateFile absolutized against the glue URL; export download-name guard. - deps: OCC builds with RapidJSON so its glTF/GLB writer exists — pinned to the vcpkg master snapshot 2025-02-26 (24b5e7a8b27f), the same code official KiCad consumes via vcpkg.json's opencascade[rapidjson]; rapidjson's latest tag (v1.1.0, 2016) is ill-formed under modern clang. - tests: occ-export dialog e2e (lazy-fetch boundary + STEP download bytes), occ-probe incl. a 9-format matrix (step/stpz/brep/xao/ply/stl/glb/u3d/pdf), 3d-viewer-models hard-asserts the worker parse; occ provider stub installed ambiently by the kicad fixtures. Validated against desktop kicad-cli 10.0.4: geometric exact equality (bbox delta 0 um, volume delta 0.0000%) for STEP/GLB/STL/BREP/STPZ across three boards and option sweeps — with desktop OCC 7.9 vs wasm OCC 7.8; PLY/XAO/PDF structurally equal; U3D same-size (quantizer float LSBs differ). Full kicad e2e green on Firefox and Chromium; standalone verified end to end (lazy fetch only on the Export click; export.step 60,628 B ISO-10303-21; loadModel 700 KB STEP -> 569 KB scenegraph cache). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:39:58 +02:00
# occ_service links the pcbnew kiface objects → pcbnew's stubs (frame +
# action-plugin scripting placeholders), like the editors.
occ_service) STUB_APP="pcbnew" ;;
*) STUB_APP="$APP_NAME" ;;
esac
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
source "${SCRIPT_DIR}/../common/env.sh"
source "${SCRIPT_DIR}/../common/versions.sh"
source "${SCRIPT_DIR}/../common/functions.sh"
2026-06-01 17:31:03 +02:00
source "${SCRIPT_DIR}/../common/stages.sh"
KICAD_DIR="${PROJECT_ROOT}/kicad"
WASM_LAYER="${PROJECT_ROOT}/wasm"
refactor: collapse dual-mode plumbing — the DOM port is the only WASM build The canvas (wxUniversal) mode is gone (wxwidgets submodule); remove every piece of side-by-side plumbing so there is exactly one build and one test flow: - scripts/build-wxuniversal-wasm.sh -> scripts/build-wx-wasm.sh; no --dom/--enable-universal; builds into build-wasm/wxwidgets - build-wasm-test.sh: no DOM_BUILD / apps-dom rsync mirror / PORT=dom; apps build straight into tests/apps (Makefile.wasm PORT conditionals collapsed; wx.js + wx-dom.js always pre-js) - docker/build.sh, build-kicad-target.sh, env.sh: WX_PORT / -dom / -universal suffixes removed; kicad builds to kicad-<app>, outputs to output/; wx.js/wx-dom.js copied from the real source path (/workspace/wxwidgets/build/wasm — the old build-wasm path never existed and silently failed) - setup-kicad-wasm.sh: single target dir; the perl wx-dom.js injection is gone — the 7 checked-in kicad pages now reference wx-dom.js directly - playwright configs serve apps/; fixtures drop the test-results/dom and logs/wxwidgets/dom namespacing; boot.spec asserts wxDomPort unconditionally; pcbnew.spec uses one reference image; appearance.spec assertions unconditional - compare/update-baseline-screenshots.sh: --port removed - tests/gal-regression/wasm/Makefile: links build-wasm/wxwidgets and carries wx-dom.js as a second pre-js — the gal-webgl suite (30 specs) now actually builds and runs here (it needed host-side boost+glm via scripts/deps; the bundle had been missing, timing the whole spec out) - tests: clickCanvas() dispatches via page.mouse (DOM widgets legitimately cover the canvas; locator actionability refused the click); the comprehensive spec drives wxChoice through its native <select> (browser-owned popup cannot be coordinate-clicked) - docs: README/CLAUDE.md/build.md script names and dirs; features/wx-dom-port README reframed (DOM is THE port), visual-notes bugs 26-28; FindwxWidgets.cmake config label drops 'wasmuniv' - wxwidgets submodule -> 9dbacc9448 (DOM-only port, fork diff shrunk) Gate: full wx e2e suite 292 passed / 1 skipped / 0 failed — first run ever with the gal-webgl specs green (28 scenarios + load + sequential). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-12 10:18:16 +02:00
KICAD_BUILD="${BUILD_ROOT}/kicad-${APP_NAME}"
KICAD_STAMP="${BUILD_ROOT}/stamps/kicad-${APP_NAME}.stamp"
WX_BUILD="${BUILD_ROOT}/wxwidgets"
# Parse arguments - incremental build by default (optimized for development)
NO_CLEAN=1
FULL_CLEAN=0
SKIP_DEPS=1
DEBUG=0
DIAG_LIST=""
while [[ $# -gt 0 ]]; do
case $1 in
--full)
FULL_CLEAN=1
NO_CLEAN=0
SKIP_DEPS=0
shift
;;
--clean-kicad)
NO_CLEAN=0
shift
;;
--build-deps)
SKIP_DEPS=0
shift
;;
--debug)
DEBUG=1
shift
;;
--release)
DEBUG_BUILD=0
export DEBUG_BUILD
shift
;;
--diag=*)
DIAG_LIST="${1#--diag=}"
shift
;;
--diag)
DIAG_LIST="$2"
shift 2
;;
-j)
export JOBS="$2"
shift 2
;;
-j*)
export JOBS="${1#-j}"
shift
;;
*)
shift
;;
esac
done
# Diagnostic preprocessor defines from --diag=<csv> (gal, coroutine, ctor, all).
# These gate the KI_DIAG_* macros in kicad/include/kicad_wasm_diag.h. Output goes
# to stdout ([KICAD_OUT] logs), never errors. Off by default.
DIAG_DEFINES=""
if [ -n "${DIAG_LIST}" ]; then
IFS=',' read -ra _diag_cats <<< "${DIAG_LIST}"
for _cat in "${_diag_cats[@]}"; do
case "${_cat}" in
gal) DIAG_DEFINES="${DIAG_DEFINES} -DKICAD_DIAG_GAL=1" ;;
coroutine) DIAG_DEFINES="${DIAG_DEFINES} -DKICAD_DIAG_COROUTINE=1" ;;
ctor) DIAG_DEFINES="${DIAG_DEFINES} -DKICAD_DIAG_CTOR=1" ;;
all) DIAG_DEFINES="${DIAG_DEFINES} -DKICAD_DIAG_GAL=1 -DKICAD_DIAG_COROUTINE=1 -DKICAD_DIAG_CTOR=1" ;;
"") ;;
*) log_warn "Unknown --diag category: '${_cat}' (valid: gal, coroutine, ctor, all)" ;;
esac
done
log_info "Diagnostic logging enabled:${DIAG_DEFINES}"
fi
log_info "Building app: ${APP_NAME}"
log_info "Using ${JOBS} parallel jobs"
# Step 1: Clean build directories
if [ $FULL_CLEAN -eq 1 ]; then
log_info "Full clean: removing all stamps and build directories..."
rm -rf "${STAMPS_DIR}"/*
rm -rf "${BUILD_ROOT}/deps"/*
refactor: collapse dual-mode plumbing — the DOM port is the only WASM build The canvas (wxUniversal) mode is gone (wxwidgets submodule); remove every piece of side-by-side plumbing so there is exactly one build and one test flow: - scripts/build-wxuniversal-wasm.sh -> scripts/build-wx-wasm.sh; no --dom/--enable-universal; builds into build-wasm/wxwidgets - build-wasm-test.sh: no DOM_BUILD / apps-dom rsync mirror / PORT=dom; apps build straight into tests/apps (Makefile.wasm PORT conditionals collapsed; wx.js + wx-dom.js always pre-js) - docker/build.sh, build-kicad-target.sh, env.sh: WX_PORT / -dom / -universal suffixes removed; kicad builds to kicad-<app>, outputs to output/; wx.js/wx-dom.js copied from the real source path (/workspace/wxwidgets/build/wasm — the old build-wasm path never existed and silently failed) - setup-kicad-wasm.sh: single target dir; the perl wx-dom.js injection is gone — the 7 checked-in kicad pages now reference wx-dom.js directly - playwright configs serve apps/; fixtures drop the test-results/dom and logs/wxwidgets/dom namespacing; boot.spec asserts wxDomPort unconditionally; pcbnew.spec uses one reference image; appearance.spec assertions unconditional - compare/update-baseline-screenshots.sh: --port removed - tests/gal-regression/wasm/Makefile: links build-wasm/wxwidgets and carries wx-dom.js as a second pre-js — the gal-webgl suite (30 specs) now actually builds and runs here (it needed host-side boost+glm via scripts/deps; the bundle had been missing, timing the whole spec out) - tests: clickCanvas() dispatches via page.mouse (DOM widgets legitimately cover the canvas; locator actionability refused the click); the comprehensive spec drives wxChoice through its native <select> (browser-owned popup cannot be coordinate-clicked) - docs: README/CLAUDE.md/build.md script names and dirs; features/wx-dom-port README reframed (DOM is THE port), visual-notes bugs 26-28; FindwxWidgets.cmake config label drops 'wasmuniv' - wxwidgets submodule -> 9dbacc9448 (DOM-only port, fork diff shrunk) Gate: full wx e2e suite 292 passed / 1 skipped / 0 failed — first run ever with the gal-webgl specs green (28 scenarios + load + sequential). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-12 10:18:16 +02:00
rm -rf "${BUILD_ROOT}/wxwidgets"
rm -rf "${BUILD_ROOT}/stubs"
rm -rf "${KICAD_BUILD}"
rm -rf "${SYSROOT}"/*
elif [ $NO_CLEAN -eq 0 ]; then
log_info "Cleaning KiCad ${APP_NAME} build directory..."
rm -rf "${KICAD_BUILD}" "${KICAD_STAMP}"
else
log_info "Incremental build (use --clean-kicad or --full to clean)"
fi
# Step 2: Build dependencies
# Note: --with-occ for OpenCASCADE. The ngspice dep is NOT needed here — the
# editor links only the sharedspice client stub; the engine is the separate
# ngspice_service app (scripts/kicad/build-ngspice_service.sh builds its dep).
if [ $SKIP_DEPS -eq 0 ]; then
2026-06-01 17:31:03 +02:00
kw_stage deps
log_info "Building dependencies..."
"${SCRIPT_DIR}/../deps/build-all-deps.sh" --with-occ
else
log_info "Skipping dependencies (use --build-deps or --full to build)"
fi
# Note: We don't check the KiCad stamp here for incremental builds.
# CMake handles dependency tracking - it will detect changed source files
# and only recompile what's needed. The stamp is created at the end for
# scripts that want to know if KiCad was ever built successfully.
# Step 4: Build wxWidgets (incremental - only recompiles changed files)
2026-06-01 17:31:03 +02:00
kw_stage wxwidgets
log_info "Building wxWidgets..."
refactor: collapse dual-mode plumbing — the DOM port is the only WASM build The canvas (wxUniversal) mode is gone (wxwidgets submodule); remove every piece of side-by-side plumbing so there is exactly one build and one test flow: - scripts/build-wxuniversal-wasm.sh -> scripts/build-wx-wasm.sh; no --dom/--enable-universal; builds into build-wasm/wxwidgets - build-wasm-test.sh: no DOM_BUILD / apps-dom rsync mirror / PORT=dom; apps build straight into tests/apps (Makefile.wasm PORT conditionals collapsed; wx.js + wx-dom.js always pre-js) - docker/build.sh, build-kicad-target.sh, env.sh: WX_PORT / -dom / -universal suffixes removed; kicad builds to kicad-<app>, outputs to output/; wx.js/wx-dom.js copied from the real source path (/workspace/wxwidgets/build/wasm — the old build-wasm path never existed and silently failed) - setup-kicad-wasm.sh: single target dir; the perl wx-dom.js injection is gone — the 7 checked-in kicad pages now reference wx-dom.js directly - playwright configs serve apps/; fixtures drop the test-results/dom and logs/wxwidgets/dom namespacing; boot.spec asserts wxDomPort unconditionally; pcbnew.spec uses one reference image; appearance.spec assertions unconditional - compare/update-baseline-screenshots.sh: --port removed - tests/gal-regression/wasm/Makefile: links build-wasm/wxwidgets and carries wx-dom.js as a second pre-js — the gal-webgl suite (30 specs) now actually builds and runs here (it needed host-side boost+glm via scripts/deps; the bundle had been missing, timing the whole spec out) - tests: clickCanvas() dispatches via page.mouse (DOM widgets legitimately cover the canvas; locator actionability refused the click); the comprehensive spec drives wxChoice through its native <select> (browser-owned popup cannot be coordinate-clicked) - docs: README/CLAUDE.md/build.md script names and dirs; features/wx-dom-port README reframed (DOM is THE port), visual-notes bugs 26-28; FindwxWidgets.cmake config label drops 'wasmuniv' - wxwidgets submodule -> 9dbacc9448 (DOM-only port, fork diff shrunk) Gate: full wx e2e suite 292 passed / 1 skipped / 0 failed — first run ever with the gal-webgl specs green (28 scenarios + load + sequential). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-06-12 10:18:16 +02:00
"${SCRIPT_DIR}/../build-wx-wasm.sh" --no-clean
log_info "Building KiCad ${APP_NAME} ${KICAD_VERSION} for WASM..."
# Step 5: Set build type
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
# Exception model: native WebAssembly exceptions (legacy binary encoding) + wasm setjmp/longjmp,
# single-sourced from scripts/common/env.sh (KiCad and wxWidgets must agree — mixing EH models
# link-fails / traps; both build with exceptions enabled). -matomics -mbulk-memory are required for
# shared memory (pthreads).
KICAD_EH_FLAGS="$DEPS_EH_FLAGS"
log_info "KiCad EH model flags: ${KICAD_EH_FLAGS}"
# Use environment DEBUG_BUILD if set, otherwise check local --debug flag
# NOTE: We use -O1 for debug builds because -O0 produces WASM with too many
# locals for V8/Chrome to compile (error: "local count too large").
# -O1 keeps debug info but optimizes enough to stay under V8's limits.
if [ "${DEBUG_BUILD:-0}" = "1" ] || [ $DEBUG -eq 1 ]; then
BUILD_TYPE="Debug"
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
EXTRA_FLAGS="-g -O1 ${KICAD_EH_FLAGS} -matomics -mbulk-memory"
# CMake defines DEBUG for Config=Debug (kicad/CMakeLists.txt:351). The embind TU (Step 7) is
# compiled OUTSIDE CMake, so it must define DEBUG too — otherwise a DEBUG-gated virtual
# (EDA_ITEM::Show, eda_item.h:471) occupies a vtable slot in the core's emitted vtable that the
# embind TU doesn't account for, shifting every later slot by one. Then every virtual call made
# from the embind TU past that slot (SetWidth/GetPosition/...) reads the wrong vtable offset and
# mis-dispatches at runtime (call_indirect signature-mismatch trap; under native-EH the trap is
# swallowed by the apply coroutine's catch_all → silent hang). See task #54 root-cause analysis.
EMBIND_CONFIG_DEFINES="-DDEBUG"
# -gseparate-dwarf puts debug info in a separate .debug.wasm file
# This keeps the main WASM small (~200MB) while preserving full debug info
# DevTools loads the debug file on-demand when debugging
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
LINKER_DEBUG_FLAGS="-O1 -g -gseparate-dwarf ${KICAD_EH_FLAGS}"
log_info "Building KiCad in DEBUG mode (separate DWARF for smaller main binary)"
else
BUILD_TYPE="Release"
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
EXTRA_FLAGS="-O2 ${KICAD_EH_FLAGS} -matomics -mbulk-memory"
EMBIND_CONFIG_DEFINES="" # Release defines no DEBUG in either TU → vtable layouts already match
# -O0 at link time skips wasm-opt (which can OOM on large WASM files)
# Compilation is still -O2 for optimized code, but we skip post-link wasm-opt
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
LINKER_DEBUG_FLAGS="-O0 ${KICAD_EH_FLAGS}"
log_info "Building KiCad in RELEASE mode (skipping wasm-opt due to memory limits)"
fi
jspi: migration phases 0-7 — build knob, scheduler shim, test successor suite Toolchain: emsdk 6.0.6 (versions.sh; cache-hash keys on it). Build knob PCBJAM_ASYNC_BACKEND=jspi|asyncify: build-kicad-target.sh links editors with -sJSPI + -sJSPI_EXPORTS=@scripts/common/jspi-exports.txt + --pre-js jspi-scheduler.js (no DYNCALLS, no post-link asyncify pipeline); wx build stamps the backend and forces clean on flip or unknown provenance; docker/build.sh passes the knob, seeds the emscripten ports cache from the volume every launch, jspi postprocess = patch-env-shim only. scripts/common/shims/jspi-scheduler.js: the JSPI successor scheduler — token-wait registry, resume turnstile (one armed resume between engine re-entries, SP swaps only at microtask boundaries), green-region spill stacks (16-aligned tops), S1 embind mutator FIFO lane + parker wraps, S6 shutdown, libctx integration hooks (suspend/end/quarantine + g_current arm/clear), SuspendError attributor, lost-wake + stuck-window watchdogs, __wxWaitDump observability. Embind: PARKER registrations get emscripten::async() under PCBJAM_JSPI (wasm/bindings/pcbjam_async_policy.h). nanosleep yields route via the shim. Tests: tests/asyncify -> tests/jspi successor suite (jspi-stack red/green shadow-stack battery, jspi-coroutine MiniCoro harness, suspend-races semantic scenarios + __wxWaitDump books coherence); projects jspi-firefox/ jspi-chrome (asyncify-webkit retired — no JSPI in WebKit); unconditional Firefox JSPI pref; guard-beacons -> wait-beacons (+wxScheduler/libctxJspi families); Makefile.wasm links test apps against JSPI with the shim as a tracked link prerequisite. Web: WasmTool setRo await + __wxWaitDump forensics, open-flow contained promise, scheduler-shim.test.ts retargeted (8 green). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NDeBaKKhQztd8KiVtHuyXr
2026-08-13 07:06:24 +02:00
# Async suspension backend (JSPI migration). Default asyncify; PCBJAM_ASYNC_BACKEND=jspi
# links the browser apps with -sJSPI (native stack switching) instead of
# -sASYNCIFY/-sDYNCALLS, defines PCBJAM_JSPI=1 for every TU (selects the JSPI
# libcontext coroutine backend + the wx port's JSPI dispatch paths), and skips
# the whole post-link asyncify pipeline (stub dance + host wasm-opt). The wx
# build this app links against must be built with the SAME knob
# (build-wx-wasm.sh reads it too). The headless CLIs (kicad_tools, occ_service)
# stay asyncify-shaped regardless: their targets pin -sASYNCIFY=0 and run in
# node/worker where no suspension backend is wanted.
PCBJAM_ASYNC_BACKEND="${PCBJAM_ASYNC_BACKEND:-asyncify}"
case "${APP_NAME}" in
kicad_tools|occ_service) PCBJAM_ASYNC_BACKEND="asyncify" ;;
esac
if [ "${PCBJAM_ASYNC_BACKEND}" = "jspi" ]; then
EXTRA_FLAGS="${EXTRA_FLAGS} -DPCBJAM_JSPI=1"
# PCBJAM_JSPI is ABI-affecting (coroutine.h's callerStub grows a finish
# hook — a template, so every instantiating TU must agree): route it into
# KICAD_TU_ABI_FLAGS via EMBIND_CONFIG_DEFINES like the Debug -DDEBUG.
EMBIND_CONFIG_DEFINES="${EMBIND_CONFIG_DEFINES} -DPCBJAM_JSPI=1"
log_info "Async backend: JSPI (-sJSPI, -DPCBJAM_JSPI=1)"
else
log_info "Async backend: asyncify"
fi
# Step 6: Create build directory
mkdir -p "${KICAD_BUILD}"
cd "${KICAD_BUILD}"
# Step 6.1: Build stub libraries for missing symbols
# Generic stubs (libgit2, curl, nng) are shared across apps and built in BUILD_ROOT/stubs.
# App-specific stubs (e.g. pcbnew_scripting_stub) build into the same directory but
# are only linked in when the corresponding source exists.
STUBS_DIR="${PROJECT_ROOT}/wasm/stubs"
STUBS_BUILD="${BUILD_ROOT}/stubs"
mkdir -p "${STUBS_BUILD}"
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
# ABI-affecting flags shared by EVERY C++ TU compiled OUTSIDE CMake (the embind + the app stubs below).
# The core CMake TUs get all of these (DEBUG via Config=Debug -> kicad/CMakeLists.txt:351;
# KICAD_USE_PLATFORM_WASM; the char16_t char_traits force-include). A TU that misses any of them can
# diverge in vtable layout / ABI from the core — task #54: the embind missing -DDEBUG shifted its vtable
# slot offsets by one and hung the collab apply (call_indirect signature-mismatch). Keep them in ONE
# place so no out-of-CMake C++ TU can skew again. (EMBIND_CONFIG_DEFINES holds the build-config -DDEBUG,
# set in the BUILD_TYPE block above; empty in Release where neither side defines DEBUG.)
KICAD_TU_ABI_FLAGS="${EMBIND_CONFIG_DEFINES} -DKICAD_USE_PLATFORM_WASM=1 -include ${STUBS_DIR}/char_traits_uint16_workaround.h"
2026-06-01 17:31:03 +02:00
kw_stage kicad-stubs
log_info "Building stub libraries..."
# Compile libgit2 stub
emcc -c "${STUBS_DIR}/libgit2_stub.c" -o "${STUBS_BUILD}/libgit2_stub.o"
emar rcs "${STUBS_BUILD}/libgit2_stub.a" "${STUBS_BUILD}/libgit2_stub.o"
# Compile curl stub
emcc -c "${STUBS_DIR}/curl_stub.c" -o "${STUBS_BUILD}/curl_stub.o"
emar rcs "${STUBS_BUILD}/libcurl_stub.a" "${STUBS_BUILD}/curl_stub.o"
2026-06-05 13:18:24 +02:00
# Note: the GLU tesselator is provided by kicad/libs/kimath/glu_tess/glu_tess_impl.cpp,
# compiled as part of the GAL library (requires KiCad headers).
# Compile NNG stub (IPC API requires NNG but sockets don't work in WASM)
emcc -c -I"${STUBS_DIR}" "${STUBS_DIR}/nng_stub.c" -o "${STUBS_BUILD}/nng_stub.o"
emar rcs "${STUBS_BUILD}/libnng_stub.a" "${STUBS_BUILD}/nng_stub.o"
# wx flags for any C++ stubs that include wx headers
WX_CXXFLAGS=$("${WX_BUILD}/wx-config" --cxxflags 2>/dev/null || echo "-I${WX_BUILD}/lib/wx/include/emscripten-unicode-static-3.2 -I${PROJECT_ROOT}/wxwidgets/include")
# App-specific stubs:
# - pcbnew: pcbnew_scripting_stub.cpp (action-plugin scripting placeholders)
# - eeschema: eeschema_frame_stub.cpp (placeholder; grows as linker dictates)
APP_STUB_LINK=""
APP_SCRIPTING_STUB_SRC="${STUBS_DIR}/${STUB_APP}_scripting_stub.cpp"
if [ -f "${APP_SCRIPTING_STUB_SRC}" ]; then
log_info "Building app scripting stub: ${STUB_APP}_scripting_stub.cpp"
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
em++ -c ${KICAD_TU_ABI_FLAGS} ${WX_CXXFLAGS} "${APP_SCRIPTING_STUB_SRC}" -o "${STUBS_BUILD}/${STUB_APP}_scripting_stub.o"
emar rcs "${STUBS_BUILD}/lib${STUB_APP}_scripting_stub.a" "${STUBS_BUILD}/${STUB_APP}_scripting_stub.o"
APP_STUB_LINK="${APP_STUB_LINK} ${STUBS_BUILD}/lib${STUB_APP}_scripting_stub.a"
fi
APP_FRAME_STUB_SRC="${STUBS_DIR}/${STUB_APP}_frame_stub.cpp"
if [ -f "${APP_FRAME_STUB_SRC}" ] && [ -s "${APP_FRAME_STUB_SRC}" ]; then
log_info "Building app frame stub: ${STUB_APP}_frame_stub.cpp"
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
em++ -c ${KICAD_TU_ABI_FLAGS} ${WX_CXXFLAGS} "${APP_FRAME_STUB_SRC}" -o "${STUBS_BUILD}/${STUB_APP}_frame_stub.o"
emar rcs "${STUBS_BUILD}/lib${STUB_APP}_frame_stub.a" "${STUBS_BUILD}/${STUB_APP}_frame_stub.o"
APP_STUB_LINK="${APP_STUB_LINK} ${STUBS_BUILD}/lib${STUB_APP}_frame_stub.a"
fi
log_info "Stub libraries built"
# Step 6.2/6.3: wasm-opt + wasm-emscripten-finalize handling.
# For the editor apps these tools OOM on the huge debug wasm, so we stub them in
# the container and run them on the host (docker/build.sh phase 2). The small,
# debug-stripped (-g0) CLI finalizes fine in-container, so for kicad_tools
# we restore/keep the real tools and skip host post-processing entirely.
if [ -z "${EMSDK}" ]; then
log_error "EMSDK environment variable is not set."
exit 1
fi
EMSDK_WASM_OPT="${EMSDK}/upstream/bin/wasm-opt"
EMSDK_FINALIZE="${EMSDK}/upstream/bin/wasm-emscripten-finalize"
jspi: migration phases 0-7 — build knob, scheduler shim, test successor suite Toolchain: emsdk 6.0.6 (versions.sh; cache-hash keys on it). Build knob PCBJAM_ASYNC_BACKEND=jspi|asyncify: build-kicad-target.sh links editors with -sJSPI + -sJSPI_EXPORTS=@scripts/common/jspi-exports.txt + --pre-js jspi-scheduler.js (no DYNCALLS, no post-link asyncify pipeline); wx build stamps the backend and forces clean on flip or unknown provenance; docker/build.sh passes the knob, seeds the emscripten ports cache from the volume every launch, jspi postprocess = patch-env-shim only. scripts/common/shims/jspi-scheduler.js: the JSPI successor scheduler — token-wait registry, resume turnstile (one armed resume between engine re-entries, SP swaps only at microtask boundaries), green-region spill stacks (16-aligned tops), S1 embind mutator FIFO lane + parker wraps, S6 shutdown, libctx integration hooks (suspend/end/quarantine + g_current arm/clear), SuspendError attributor, lost-wake + stuck-window watchdogs, __wxWaitDump observability. Embind: PARKER registrations get emscripten::async() under PCBJAM_JSPI (wasm/bindings/pcbjam_async_policy.h). nanosleep yields route via the shim. Tests: tests/asyncify -> tests/jspi successor suite (jspi-stack red/green shadow-stack battery, jspi-coroutine MiniCoro harness, suspend-races semantic scenarios + __wxWaitDump books coherence); projects jspi-firefox/ jspi-chrome (asyncify-webkit retired — no JSPI in WebKit); unconditional Firefox JSPI pref; guard-beacons -> wait-beacons (+wxScheduler/libctxJspi families); Makefile.wasm links test apps against JSPI with the shim as a tracked link prerequisite. Web: WasmTool setRo await + __wxWaitDump forensics, open-flow contained promise, scheduler-shim.test.ts retargeted (8 green). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NDeBaKKhQztd8KiVtHuyXr
2026-08-13 07:06:24 +02:00
if [ "${APP_NAME}" = "kicad_tools" ] || [ "${APP_NAME}" = "occ_service" ] || [ "${PCBJAM_ASYNC_BACKEND}" = "jspi" ]; then
# Use the real tools so the module is fully finalized inside the container
# (no host post-processing / asyncify for these targets). JSPI mode has no
# asyncify pass at all, so every app finalizes in-container.
[ -f "${EMSDK_WASM_OPT}.real" ] && cp "${EMSDK_WASM_OPT}.real" "${EMSDK_WASM_OPT}"
[ -f "${EMSDK_FINALIZE}.real" ] && cp "${EMSDK_FINALIZE}.real" "${EMSDK_FINALIZE}"
feat(wasm): occ-split — lazy occ_service worker; kicad_editor drops OCC (−31%) Move OpenCASCADE out of the merged editor image into occ_service: a separate emscripten module (-sASYNCIFY=0, MODULARIZE, in-container -Oz finalize, 2N+8 pre-warmed pthread pool) booted lazily in a dedicated Web Worker on the first STEP export or STEP/IGES model parse. kicad_editor.wasm ~190 MB -> 130 MB; sessions that never touch OCC never fetch its 57 MB. STEP export works in the browser for the first time: the unchanged desktop dialog runs EXPORTER_STEP, whose wasm shadow suspends into globalThis.occService and the export bytes go straight to a browser download (never entering the editor heap). STEP/IGES 3D models parse in the worker via the oce shadow (S3D WriteCache/ReadCache wire). - wasm/occ-service/: service CMake target (hooked from the kicad fork's top-level CMakeLists, wasm/editor pattern), embind entry (occExport/occLoadModel), wxConfig pre-js. - wasm/stubs/{exporter_step,oce_plugin}_stub.cpp: EM_ASYNC_JS worker bridges (callee-shadowing; no caller #ifdefs). - web/standalone: provider installed whenever the kicad_editor bundle boots (cross-face safe); ONE shared worker-boot source occ-worker.js (vite ?raw; the e2e stub reads the same file) — blob worker with locateFile absolutized against the glue URL; export download-name guard. - deps: OCC builds with RapidJSON so its glTF/GLB writer exists — pinned to the vcpkg master snapshot 2025-02-26 (24b5e7a8b27f), the same code official KiCad consumes via vcpkg.json's opencascade[rapidjson]; rapidjson's latest tag (v1.1.0, 2016) is ill-formed under modern clang. - tests: occ-export dialog e2e (lazy-fetch boundary + STEP download bytes), occ-probe incl. a 9-format matrix (step/stpz/brep/xao/ply/stl/glb/u3d/pdf), 3d-viewer-models hard-asserts the worker parse; occ provider stub installed ambiently by the kicad fixtures. Validated against desktop kicad-cli 10.0.4: geometric exact equality (bbox delta 0 um, volume delta 0.0000%) for STEP/GLB/STL/BREP/STPZ across three boards and option sweeps — with desktop OCC 7.9 vs wasm OCC 7.8; PLY/XAO/PDF structurally equal; U3D same-size (quantizer float LSBs differ). Full kicad e2e green on Firefox and Chromium; standalone verified end to end (lazy fetch only on the Export click; export.step 60,628 B ISO-10303-21; loadModel 700 KB STEP -> 569 KB scenegraph cache). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:39:58 +02:00
log_info "Using real wasm-opt/finalize for ${APP_NAME} (finalize in-container)"
else
if [ -f "${EMSDK_WASM_OPT}" ] && [ ! -f "${EMSDK_WASM_OPT}.real" ]; then
log_info "Backing up real wasm-opt..."
mv "${EMSDK_WASM_OPT}" "${EMSDK_WASM_OPT}.real"
fi
# Always copy the latest stub (in case it was updated)
cp "${STUBS_DIR}/wasm-opt-stub.sh" "${EMSDK_WASM_OPT}"
chmod +x "${EMSDK_WASM_OPT}"
log_info "wasm-opt stub installed (asyncify will run on host)"
if [ -f "${EMSDK_FINALIZE}" ] && [ ! -f "${EMSDK_FINALIZE}.real" ]; then
log_info "Backing up real wasm-emscripten-finalize..."
mv "${EMSDK_FINALIZE}" "${EMSDK_FINALIZE}.real"
fi
# Always copy the latest stub (in case it was updated)
cp "${STUBS_DIR}/wasm-emscripten-finalize-stub.sh" "${EMSDK_FINALIZE}"
chmod +x "${EMSDK_FINALIZE}"
log_info "wasm-emscripten-finalize stub installed (finalize will run on host)"
fi
# Step 6.5: Verify WASM support is in KiCad fork
# The kicad submodule should already have WASM port detection and kiplatform support
KICAD_CMAKE="${KICAD_DIR}/CMakeLists.txt"
if ! grep -q "msw|qt|gtk|osx|wasm" "${KICAD_CMAKE}"; then
log_error "KiCad fork is missing WASM port detection support."
log_error "Please ensure the kicad submodule has WASM modifications."
exit 1
fi
KIPLATFORM_CMAKE="${KICAD_DIR}/libs/kiplatform/CMakeLists.txt"
if ! grep -q "KICAD_WX_PORT STREQUAL wasm" "${KIPLATFORM_CMAKE}"; then
log_error "KiCad fork is missing kiplatform WASM support."
log_error "Please ensure the kicad submodule has WASM modifications."
exit 1
fi
log_info "KiCad WASM support verified"
# Embind object — built after CMake configure runs (so config.h exists). The
# linker line below references "${STUBS_BUILD}/${APP_NAME}_embind.o" so we
# create an empty placeholder when the source is missing, to keep the link
# line stable across apps.
feat(wasm): kicad_editor — merge the pcbnew+eeschema kifaces into ONE bundle (Part 2) All four editors (PCB / Footprint / Schematic / Symbol) are now runtime --frame choices of a single kicad_editor.wasm (178 MB at -O1 vs 147+82 separate; shared wx/common/boost linked once). One editor per page load, as before; frames pcb / fpedit / sch / symedit. - wasm/editor/: the merged executable target (single_top + both kiface library sets, whole-archive pcbcommon) + the safety-net focus-walk Kiface() dispatch TU. Gated by KICAD_WASM_MERGED_EDITOR (kicad submodule bump carries the fork side: per-engine Kiface/getter binding + ODR renames + dual-kiface launcher). - wasm/bindings/: per-editor collab entries renamed pcbCollab*/schCollab* (JS names unchanged); duplicate kicadOpenFile/kicadCollabOnSave + shared-name registrations guarded behind KICAD_MERGED_EMBIND; new kicad_editor_embind.cpp registers each shared JS name once, dispatching on the live frame. - Build: kicad_editor app (build wrapper, target case arms, 3-object embind compile with the ABI-critical flags, STUB_APP=pcbnew); docker/build.sh "all" = kicad_editor calculator pl_editor gerbview (pcbnew/eeschema stay as explicit debug apps); scripts/kicad/audit-merged-symbols.sh = repeatable ODR-collision audit (run on kicad bumps). - Frontend: Bundle type (bundle ≠ tool); TOOL_BUNDLE maps all four editors to kicad_editor; explicit --frame tokens for pcbnew (pcb) and eeschema (sch); publish list = the 4 real bundles. - Tests/CI: five harnesses load kicad_editor.js with explicit frame tokens; PCBNEW_FAMILY_SPECS renamed BIG_MODULE_SPECS + the 8 eeschema-family specs (they now boot the merged module — SpiderMonkey x86 CI OOM routing); frame-runtime spec covers all four frames from the one bundle. Validated so far: frame-runtime 4/4 (each frame boots with the right title, no aborts, no duplicate embind registration); 24-spec merged-module regression green; 3D raytracer renders. Known pre-existing failure: 3d-viewer title-bar drag deadlock, fixed on main by 7630c7e (2N+8 pthread pre-warm) — picked up by the follow-up rebase. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 14:48:11 +02:00
# kicad_editor (merged image) links THREE embind objects: both per-editor TUs
# (compiled with -DKICAD_MERGED_EMBIND, which compiles out their duplicate
# definitions and shared-name registrations) plus the dispatcher TU that registers
# the shared JS names once — see wasm/bindings/kicad_editor_embind.cpp.
if [ "${APP_NAME}" = "kicad_editor" ]; then
EMBIND_OBJ="${STUBS_BUILD}/pcbnew_embind.o ${STUBS_BUILD}/eeschema_embind.o ${STUBS_BUILD}/kicad_editor_embind.o"
else
EMBIND_OBJ="${STUBS_BUILD}/${EMBIND_APP}_embind.o"
fi
EMBIND_SRC="${PROJECT_ROOT}/wasm/bindings/${EMBIND_APP}_embind.cpp"
# Step 7: Configure KiCad with CMake
# We use CMAKE_MODULE_PATH to inject our compatibility layer
2026-06-01 17:31:03 +02:00
kw_stage kicad-configure
log_info "Configuring KiCad with CMake..."
jspi: migration phases 0-7 — build knob, scheduler shim, test successor suite Toolchain: emsdk 6.0.6 (versions.sh; cache-hash keys on it). Build knob PCBJAM_ASYNC_BACKEND=jspi|asyncify: build-kicad-target.sh links editors with -sJSPI + -sJSPI_EXPORTS=@scripts/common/jspi-exports.txt + --pre-js jspi-scheduler.js (no DYNCALLS, no post-link asyncify pipeline); wx build stamps the backend and forces clean on flip or unknown provenance; docker/build.sh passes the knob, seeds the emscripten ports cache from the volume every launch, jspi postprocess = patch-env-shim only. scripts/common/shims/jspi-scheduler.js: the JSPI successor scheduler — token-wait registry, resume turnstile (one armed resume between engine re-entries, SP swaps only at microtask boundaries), green-region spill stacks (16-aligned tops), S1 embind mutator FIFO lane + parker wraps, S6 shutdown, libctx integration hooks (suspend/end/quarantine + g_current arm/clear), SuspendError attributor, lost-wake + stuck-window watchdogs, __wxWaitDump observability. Embind: PARKER registrations get emscripten::async() under PCBJAM_JSPI (wasm/bindings/pcbjam_async_policy.h). nanosleep yields route via the shim. Tests: tests/asyncify -> tests/jspi successor suite (jspi-stack red/green shadow-stack battery, jspi-coroutine MiniCoro harness, suspend-races semantic scenarios + __wxWaitDump books coherence); projects jspi-firefox/ jspi-chrome (asyncify-webkit retired — no JSPI in WebKit); unconditional Firefox JSPI pref; guard-beacons -> wait-beacons (+wxScheduler/libctxJspi families); Makefile.wasm links test apps against JSPI with the shim as a tracked link prerequisite. Web: WasmTool setRo await + __wxWaitDump forensics, open-flow contained promise, scheduler-shim.test.ts retargeted (8 green). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NDeBaKKhQztd8KiVtHuyXr
2026-08-13 07:06:24 +02:00
# emcc 6: -sUSE_ZLIB is gone; prebuild the zlib port (regular + PIC) so the
# explicit ZLIB_LIBRARY below always exists (PIC needed by the .so links).
"${EMSDK}/upstream/emscripten/embuilder" build zlib >/dev/null 2>&1 || true
"${EMSDK}/upstream/emscripten/embuilder" build zlib --pic >/dev/null 2>&1 || true
# CMAKE_SHARED/MODULE_LINKER_FLAGS below: set EXPLICITLY (a stale CMakeCache
# once resurrected experimental -L hacks), and carry the SAME ODR policy as
# the exe links (wasm/editor/CMakeLists.txt): KiCad deliberately ships
# wx-compat copies (wxGetContentRect in grid_checkbox.cpp vs wx grid.cpp)
# that emcc 6 kiface/.so links reject without --allow-multiple-definition.
# Use ccache if available (CMAKE_*_COMPILER_LAUNCHER is the proper CMake way)
CCACHE_OPTS=""
if command -v ccache &> /dev/null; then
CCACHE_OPTS="-DCMAKE_C_COMPILER_LAUNCHER=ccache -DCMAKE_CXX_COMPILER_LAUNCHER=ccache"
log_info "Using ccache for compilation"
fi
# The merged headless CLI configures with BOTH diet options (each trims its
# own kiface for this tree; the eeschema one also trims the SCH_IO factory to
# the two KiCad plugins) plus the wasm/tools/ subdir gate.
KICAD_TOOLS_CMAKE_FLAG=""
if [ "${APP_NAME}" = "kicad_tools" ]; then
KICAD_TOOLS_CMAKE_FLAG="-DKICAD_TOOLS_WASM=ON -DKICAD_SYM_CONVERTER_WASM=ON -DKICAD_PCB_CONVERTER_WASM=ON"
fi
feat(wasm): kicad_editor — merge the pcbnew+eeschema kifaces into ONE bundle (Part 2) All four editors (PCB / Footprint / Schematic / Symbol) are now runtime --frame choices of a single kicad_editor.wasm (178 MB at -O1 vs 147+82 separate; shared wx/common/boost linked once). One editor per page load, as before; frames pcb / fpedit / sch / symedit. - wasm/editor/: the merged executable target (single_top + both kiface library sets, whole-archive pcbcommon) + the safety-net focus-walk Kiface() dispatch TU. Gated by KICAD_WASM_MERGED_EDITOR (kicad submodule bump carries the fork side: per-engine Kiface/getter binding + ODR renames + dual-kiface launcher). - wasm/bindings/: per-editor collab entries renamed pcbCollab*/schCollab* (JS names unchanged); duplicate kicadOpenFile/kicadCollabOnSave + shared-name registrations guarded behind KICAD_MERGED_EMBIND; new kicad_editor_embind.cpp registers each shared JS name once, dispatching on the live frame. - Build: kicad_editor app (build wrapper, target case arms, 3-object embind compile with the ABI-critical flags, STUB_APP=pcbnew); docker/build.sh "all" = kicad_editor calculator pl_editor gerbview (pcbnew/eeschema stay as explicit debug apps); scripts/kicad/audit-merged-symbols.sh = repeatable ODR-collision audit (run on kicad bumps). - Frontend: Bundle type (bundle ≠ tool); TOOL_BUNDLE maps all four editors to kicad_editor; explicit --frame tokens for pcbnew (pcb) and eeschema (sch); publish list = the 4 real bundles. - Tests/CI: five harnesses load kicad_editor.js with explicit frame tokens; PCBNEW_FAMILY_SPECS renamed BIG_MODULE_SPECS + the 8 eeschema-family specs (they now boot the merged module — SpiderMonkey x86 CI OOM routing); frame-runtime spec covers all four frames from the one bundle. Validated so far: frame-runtime 4/4 (each frame boots with the right title, no aborts, no duplicate embind registration); 24-spec merged-module regression green; 3D raytracer renders. Known pre-existing failure: 3d-viewer title-bar drag deadlock, fixed on main by 7630c7e (2N+8 pthread pre-warm) — picked up by the follow-up rebase. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 14:48:11 +02:00
# Merged pcbnew+eeschema editor: gates the wasm/editor/ subdir, the per-engine
# Kiface()/KIFACE_GETTER renames, and the PCB-side ODR symbol renames in the fork's
# CMake (see KICAD_WASM_MERGED_EDITOR in kicad/CMakeLists.txt). Only this tree
# (build-wasm/kicad-kicad_editor) configures with it ON.
MERGED_EDITOR_CMAKE_FLAG=""
if [ "${APP_NAME}" = "kicad_editor" ]; then
MERGED_EDITOR_CMAKE_FLAG="-DKICAD_WASM_MERGED_EDITOR=ON"
fi
feat(wasm): occ-split — lazy occ_service worker; kicad_editor drops OCC (−31%) Move OpenCASCADE out of the merged editor image into occ_service: a separate emscripten module (-sASYNCIFY=0, MODULARIZE, in-container -Oz finalize, 2N+8 pre-warmed pthread pool) booted lazily in a dedicated Web Worker on the first STEP export or STEP/IGES model parse. kicad_editor.wasm ~190 MB -> 130 MB; sessions that never touch OCC never fetch its 57 MB. STEP export works in the browser for the first time: the unchanged desktop dialog runs EXPORTER_STEP, whose wasm shadow suspends into globalThis.occService and the export bytes go straight to a browser download (never entering the editor heap). STEP/IGES 3D models parse in the worker via the oce shadow (S3D WriteCache/ReadCache wire). - wasm/occ-service/: service CMake target (hooked from the kicad fork's top-level CMakeLists, wasm/editor pattern), embind entry (occExport/occLoadModel), wxConfig pre-js. - wasm/stubs/{exporter_step,oce_plugin}_stub.cpp: EM_ASYNC_JS worker bridges (callee-shadowing; no caller #ifdefs). - web/standalone: provider installed whenever the kicad_editor bundle boots (cross-face safe); ONE shared worker-boot source occ-worker.js (vite ?raw; the e2e stub reads the same file) — blob worker with locateFile absolutized against the glue URL; export download-name guard. - deps: OCC builds with RapidJSON so its glTF/GLB writer exists — pinned to the vcpkg master snapshot 2025-02-26 (24b5e7a8b27f), the same code official KiCad consumes via vcpkg.json's opencascade[rapidjson]; rapidjson's latest tag (v1.1.0, 2016) is ill-formed under modern clang. - tests: occ-export dialog e2e (lazy-fetch boundary + STEP download bytes), occ-probe incl. a 9-format matrix (step/stpz/brep/xao/ply/stl/glb/u3d/pdf), 3d-viewer-models hard-asserts the worker parse; occ provider stub installed ambiently by the kicad fixtures. Validated against desktop kicad-cli 10.0.4: geometric exact equality (bbox delta 0 um, volume delta 0.0000%) for STEP/GLB/STL/BREP/STPZ across three boards and option sweeps — with desktop OCC 7.9 vs wasm OCC 7.8; PLY/XAO/PDF structurally equal; U3D same-size (quantizer float LSBs differ). Full kicad e2e green on Firefox and Chromium; standalone verified end to end (lazy fetch only on the Export click; export.step 60,628 B ISO-10303-21; loadModel 700 KB STEP -> 569 KB scenegraph cache). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:39:58 +02:00
# The standalone OCC 3D service (worker embind module) — gates the
# wasm/occ-service/ subdir in the fork's top-level CMakeLists.
OCC_SERVICE_CMAKE_FLAG=""
if [ "${APP_NAME}" = "occ_service" ]; then
OCC_SERVICE_CMAKE_FLAG="-DKICAD_OCC_SERVICE_WASM=ON"
fi
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
# 3D viewer: built by DEFAULT (BUILD_3D_VIEWER=ON). Opt out with BUILD_3D_VIEWER=OFF, which links the
# 3D stubs instead. The default renderer is still the GL-free CPU raytracer
# (RENDER_3D_RAYTRACE_RAM) blitted through a WebGL2 textured quad, but KiCad's fixed-function
# OpenGL renderer (RENDER_3D_OPENGL) now links against the REAL GL1->WebGL2 emulation layer
# (wasm/gl1, no -sLEGACY_GL_EMULATION): the shim implements the FFP-only entry points and
# --wrap-intercepts the Emscripten-owned names it must observe (wasm/gl1/wrapped_symbols.txt).
# Regression gate: tests/3d-regression (47 golden scenarios). The KiCad CMake option
# KICAD_BUILD_3D_VIEWER_WASM stays OFF upstream; our build passes it explicitly.
# See wasm/gl1/README.md and docs/features/fork-cleanup/10-3d-viewer.md.
feat(wasm-eh): migrate the WASM build to native wasm exceptions (+ 3D viewer default-on) Replace the legacy Emscripten JS-exceptions model with native wasm-EH (legacy encoding) across the whole build, keeping Asyncify coroutines working via a from-source Binaryen --hoist-cpp-catches pre-pass. Net result: native-EH is the only build mode, the 3D viewer is on by default, and pcbnew shrinks substantially. Highlights: - Binaryen submodule everywhere + --hoist-cpp-catches integration in apply-asyncify; post-link Asyncify covers every app wasm (not just standalone test wasm). - Build deps (incl. OpenCASCADE without OCC_CONVERT_SIGNALS) and all KiCad apps with -fwasm-exceptions; emscripten_sleep added to the post-link asyncify-imports. - libcontext fiber entry wired under native exceptions; while-loop main loop + currData shim injected into all wx apps. - Native-EH collab apply fixed: DEBUG-define the embind TU + match all out-of-CMake C++ TUs' ABI flags to the core, fixing the vtable-layout skew / mis-dispatch. - 3D viewer enabled by default (real raytracer linked, not the stub). - Retire the EH-spike scaffolding; flip the asyncify-races ablation pins to shim-redundancy pins (native-EH stays clean with the legacy shims ablated). - Fix the asyncify-races quiescence check to not require Asyncify.currData==0: under the native-EH per-frame-yield top loop the main stack is asyncify-suspended every frame, so currData legitimately churns (a freed-but-not-yet-nulled buffer, not a leak). Refresh the pcbnew toolbar screenshot baseline for the new kicad. - CI: drop the obsolete binaryen_version input/env (the build uses the binaryen submodule fork's wasm-opt, not a version download); key the wasm-output cache on the binaryen submodule SHA instead. Bumps the wxwidgets + binaryen submodules to their squashed feature commits. Validated green: all 7 apps native-EH (real 3D in pcbnew); KiCad e2e 63/63 Firefox + Chromium (3D viewer renders); wx 336; coroutine 34/34 both engines; asyncify 7/7 both engines. Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-29 19:50:18 +02:00
BUILD_3D_VIEWER="${BUILD_3D_VIEWER:-ON}"
GL3D_LINK_FLAGS=""
if [ "${BUILD_3D_VIEWER}" = "ON" ]; then
log_info "3D viewer ENABLED for WASM (BUILD_3D_VIEWER=ON) — compiling wasm/gl1 FFP shim"
GL1_DIR="${PROJECT_ROOT}/wasm/gl1"
while IFS= read -r gl1_src; do
[ -n "${gl1_src}" ] || continue
em++ -c ${EXTRA_FLAGS} -pthread -std=c++20 \
-I"${GL1_DIR}/include" -I"${STUBS_DIR}" -I"${SYSROOT}/include" \
"${GL1_DIR}/src/${gl1_src}" -o "${STUBS_BUILD}/${gl1_src%.cpp}.o"
GL3D_LINK_FLAGS="${GL3D_LINK_FLAGS} ${STUBS_BUILD}/${gl1_src%.cpp}.o"
done < "${GL1_DIR}/sources.txt"
while IFS= read -r gl1_sym; do
[ -n "${gl1_sym}" ] || continue
GL3D_LINK_FLAGS="${GL3D_LINK_FLAGS} -Wl,--wrap=${gl1_sym}"
done < "${GL1_DIR}/wrapped_symbols.txt"
fi
# Multi-threaded CPU raytracer (mainline threading restored): link the main-thread
# nanosleep->Asyncify-yield shim so the raw-thread raytracer joins (sleep_for busy-wait)
# yield to the JS event loop instead of deadlocking on-demand pthread-Worker creation.
# Mirrors the wasm/gl1 pattern (compile to .o, add to the link). Shim:
# wasm/shims/nanosleep_yield.c; its EM_ASYNC_JS yield is covered by env.__asyncjs__* in
# scripts/common/asyncify-imports.txt.
# The synchronous node CLIs (ASYNCIFY=0) must NOT link it: their Asyncify JS
# runtime doesn't exist, so the shim's yield throws "Asyncify is not defined"
# on the first main-thread sleep (e.g. DRC copper-clearance's worker-poll).
# A blocking CLI wants libc's real blocking nanosleep anyway — node permits
# Atomics.wait on its main thread.
if [ "${APP_NAME}" = "kicad_tools" ] || [ "${APP_NAME}" = "occ_service" ]; then
NANOSLEEP_YIELD_LINK=""
else
emcc -c -pthread "${PROJECT_ROOT}/wasm/shims/nanosleep_yield.c" -o "${STUBS_BUILD}/nanosleep_yield.o"
jspi: migration phases 0-7 — build knob, scheduler shim, test successor suite Toolchain: emsdk 6.0.6 (versions.sh; cache-hash keys on it). Build knob PCBJAM_ASYNC_BACKEND=jspi|asyncify: build-kicad-target.sh links editors with -sJSPI + -sJSPI_EXPORTS=@scripts/common/jspi-exports.txt + --pre-js jspi-scheduler.js (no DYNCALLS, no post-link asyncify pipeline); wx build stamps the backend and forces clean on flip or unknown provenance; docker/build.sh passes the knob, seeds the emscripten ports cache from the volume every launch, jspi postprocess = patch-env-shim only. scripts/common/shims/jspi-scheduler.js: the JSPI successor scheduler — token-wait registry, resume turnstile (one armed resume between engine re-entries, SP swaps only at microtask boundaries), green-region spill stacks (16-aligned tops), S1 embind mutator FIFO lane + parker wraps, S6 shutdown, libctx integration hooks (suspend/end/quarantine + g_current arm/clear), SuspendError attributor, lost-wake + stuck-window watchdogs, __wxWaitDump observability. Embind: PARKER registrations get emscripten::async() under PCBJAM_JSPI (wasm/bindings/pcbjam_async_policy.h). nanosleep yields route via the shim. Tests: tests/asyncify -> tests/jspi successor suite (jspi-stack red/green shadow-stack battery, jspi-coroutine MiniCoro harness, suspend-races semantic scenarios + __wxWaitDump books coherence); projects jspi-firefox/ jspi-chrome (asyncify-webkit retired — no JSPI in WebKit); unconditional Firefox JSPI pref; guard-beacons -> wait-beacons (+wxScheduler/libctxJspi families); Makefile.wasm links test apps against JSPI with the shim as a tracked link prerequisite. Web: WasmTool setRo await + __wxWaitDump forensics, open-flow contained promise, scheduler-shim.test.ts retargeted (8 green). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NDeBaKKhQztd8KiVtHuyXr
2026-08-13 07:06:24 +02:00
if [ "${PCBJAM_ASYNC_BACKEND}" = "jspi" ]; then
# JSPI: no scheduler-context lane exists — the weak
# pcbjam_context_sleep_ms stays null and every main-thread sleep
# suspends in place (legal on a promising activation; EM_ASYNC_JS
# auto-wraps as WebAssembly.Suspending under -sJSPI).
# context_sleep.cpp is fiber-only machinery and must not link.
NANOSLEEP_YIELD_LINK="${STUBS_BUILD}/nanosleep_yield.o"
else
# Its scheduler-aware half (docs/features/async/22 Phase B): a main-thread
# sleep on a scheduler context parks THAT CONTEXT instead of suspending the
# stack in place. C++ because the registry is a header-only C++ layer in
# wx's port, hence WX_CXXFLAGS for the include path (the same reason
# thirdparty/libcontext needed it at Phase A).
em++ -c -std=c++17 -pthread ${WX_CXXFLAGS} \
"${PROJECT_ROOT}/wasm/shims/context_sleep.cpp" -o "${STUBS_BUILD}/context_sleep.o"
NANOSLEEP_YIELD_LINK="${STUBS_BUILD}/nanosleep_yield.o ${STUBS_BUILD}/context_sleep.o"
fi
fi
# mallinfo() stub for the mimalloc build: -sMALLOC=mimalloc doesn't export the
# glibc mallinfo() that OpenCASCADE's OSD_MemInfo.cxx (libTKernel, pcbnew's 3D)
# references — without this the pcbnew link fails `undefined symbol: mallinfo`.
# Zeroed no-op (memory reporting only); harmless for apps that don't reference it.
emcc -c "${PROJECT_ROOT}/wasm/shims/mallinfo_stub.c" -o "${STUBS_BUILD}/mallinfo_stub.o"
MALLINFO_STUB_LINK="${STUBS_BUILD}/mallinfo_stub.o"
# Pre-warmed Web Worker pool size (emscripten pthreads). The 3D-viewer CPU raytracer runs
# KiCad's shared thread pool (hardware_concurrency long-lived threads, created at startup)
# AND, for camera-move preview + post-process passes, spawns its OWN set of raw std::thread
# workers — so a 3D-viewer session needs ~2x hardware_concurrency Workers *simultaneously*.
# With only one set pre-warmed, the raytracer's threads fall back to on-demand `new Worker()`,
# whose loaded→run handshake needs the main thread back in the JS event loop — which it can't
# reach while blocked in the render's busy-wait join (no PROXY_TO_PTHREAD; the join runs on the
# browser main thread). That circular wait is the 3D-viewer deadlock: moving the model, dragging
# or resizing the viewer all freeze the tab. Pre-warm enough Workers that on-demand creation
# never happens. Only 3D-viewer builds pay the extra startup Workers; other apps keep one set.
if [ "${BUILD_3D_VIEWER:-OFF}" = "ON" ]; then
PTHREAD_POOL_EXPR='navigator.hardwareConcurrency*2+8'
else
PTHREAD_POOL_EXPR='navigator.hardwareConcurrency'
fi
jspi: migration phases 0-7 — build knob, scheduler shim, test successor suite Toolchain: emsdk 6.0.6 (versions.sh; cache-hash keys on it). Build knob PCBJAM_ASYNC_BACKEND=jspi|asyncify: build-kicad-target.sh links editors with -sJSPI + -sJSPI_EXPORTS=@scripts/common/jspi-exports.txt + --pre-js jspi-scheduler.js (no DYNCALLS, no post-link asyncify pipeline); wx build stamps the backend and forces clean on flip or unknown provenance; docker/build.sh passes the knob, seeds the emscripten ports cache from the volume every launch, jspi postprocess = patch-env-shim only. scripts/common/shims/jspi-scheduler.js: the JSPI successor scheduler — token-wait registry, resume turnstile (one armed resume between engine re-entries, SP swaps only at microtask boundaries), green-region spill stacks (16-aligned tops), S1 embind mutator FIFO lane + parker wraps, S6 shutdown, libctx integration hooks (suspend/end/quarantine + g_current arm/clear), SuspendError attributor, lost-wake + stuck-window watchdogs, __wxWaitDump observability. Embind: PARKER registrations get emscripten::async() under PCBJAM_JSPI (wasm/bindings/pcbjam_async_policy.h). nanosleep yields route via the shim. Tests: tests/asyncify -> tests/jspi successor suite (jspi-stack red/green shadow-stack battery, jspi-coroutine MiniCoro harness, suspend-races semantic scenarios + __wxWaitDump books coherence); projects jspi-firefox/ jspi-chrome (asyncify-webkit retired — no JSPI in WebKit); unconditional Firefox JSPI pref; guard-beacons -> wait-beacons (+wxScheduler/libctxJspi families); Makefile.wasm links test apps against JSPI with the shim as a tracked link prerequisite. Web: WasmTool setRo await + __wxWaitDump forensics, open-flow contained promise, scheduler-shim.test.ts retargeted (8 green). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NDeBaKKhQztd8KiVtHuyXr
2026-08-13 07:06:24 +02:00
# Suspension-backend link surface. Asyncify: binaryen instrumentation +
# DYNCALLS (+ the dynCall runtime export used by the post-link dyncall shims).
# JSPI: native suspension — -sJSPI + the promising-export census
# (scripts/common/jspi-exports.txt: the wx KEEPALIVE entries that can park +
# pcbjam_libctx_entry; regenerate by grep, not memory), the jspi-scheduler
# pre-js (successor of the post-link-injected asyncify-scheduler.js), and the
# runtime methods its green-copy stack discipline needs (stackSave/
# stackRestore/HEAPU8). -sDYNCALLS is a fatal link error under JSPI.
if [ "${PCBJAM_ASYNC_BACKEND}" = "jspi" ]; then
ASYNC_LINK_FLAGS="-sJSPI -sJSPI_EXPORTS=@${PROJECT_ROOT}/scripts/common/jspi-exports.txt --pre-js ${PROJECT_ROOT}/scripts/common/shims/jspi-scheduler.js"
ASYNC_RUNTIME_METHODS="-sEXPORTED_RUNTIME_METHODS=['ccall','cwrap','UTF8ToString','stringToUTF8','lengthBytesUTF8','stackSave','stackRestore','HEAPU8','HEAP8','HEAP32']"
else
ASYNC_LINK_FLAGS="-sASYNCIFY=1 -sDYNCALLS=1 -sASYNCIFY_STACK_SIZE=65536"
ASYNC_RUNTIME_METHODS="-sEXPORTED_RUNTIME_METHODS=['ccall','cwrap','UTF8ToString','stringToUTF8','lengthBytesUTF8','dynCall'] -sDEFAULT_LIBRARY_FUNCS_TO_INCLUDE=['\$dynCall']"
fi
emcmake cmake "${KICAD_DIR}" \
${CCACHE_OPTS} \
${KICAD_TOOLS_CMAKE_FLAG} \
feat(wasm): kicad_editor — merge the pcbnew+eeschema kifaces into ONE bundle (Part 2) All four editors (PCB / Footprint / Schematic / Symbol) are now runtime --frame choices of a single kicad_editor.wasm (178 MB at -O1 vs 147+82 separate; shared wx/common/boost linked once). One editor per page load, as before; frames pcb / fpedit / sch / symedit. - wasm/editor/: the merged executable target (single_top + both kiface library sets, whole-archive pcbcommon) + the safety-net focus-walk Kiface() dispatch TU. Gated by KICAD_WASM_MERGED_EDITOR (kicad submodule bump carries the fork side: per-engine Kiface/getter binding + ODR renames + dual-kiface launcher). - wasm/bindings/: per-editor collab entries renamed pcbCollab*/schCollab* (JS names unchanged); duplicate kicadOpenFile/kicadCollabOnSave + shared-name registrations guarded behind KICAD_MERGED_EMBIND; new kicad_editor_embind.cpp registers each shared JS name once, dispatching on the live frame. - Build: kicad_editor app (build wrapper, target case arms, 3-object embind compile with the ABI-critical flags, STUB_APP=pcbnew); docker/build.sh "all" = kicad_editor calculator pl_editor gerbview (pcbnew/eeschema stay as explicit debug apps); scripts/kicad/audit-merged-symbols.sh = repeatable ODR-collision audit (run on kicad bumps). - Frontend: Bundle type (bundle ≠ tool); TOOL_BUNDLE maps all four editors to kicad_editor; explicit --frame tokens for pcbnew (pcb) and eeschema (sch); publish list = the 4 real bundles. - Tests/CI: five harnesses load kicad_editor.js with explicit frame tokens; PCBNEW_FAMILY_SPECS renamed BIG_MODULE_SPECS + the 8 eeschema-family specs (they now boot the merged module — SpiderMonkey x86 CI OOM routing); frame-runtime spec covers all four frames from the one bundle. Validated so far: frame-runtime 4/4 (each frame boots with the right title, no aborts, no duplicate embind registration); 24-spec merged-module regression green; 3D raytracer renders. Known pre-existing failure: 3d-viewer title-bar drag deadlock, fixed on main by 7630c7e (2N+8 pthread pre-warm) — picked up by the follow-up rebase. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 14:48:11 +02:00
${MERGED_EDITOR_CMAKE_FLAG} \
feat(wasm): occ-split — lazy occ_service worker; kicad_editor drops OCC (−31%) Move OpenCASCADE out of the merged editor image into occ_service: a separate emscripten module (-sASYNCIFY=0, MODULARIZE, in-container -Oz finalize, 2N+8 pre-warmed pthread pool) booted lazily in a dedicated Web Worker on the first STEP export or STEP/IGES model parse. kicad_editor.wasm ~190 MB -> 130 MB; sessions that never touch OCC never fetch its 57 MB. STEP export works in the browser for the first time: the unchanged desktop dialog runs EXPORTER_STEP, whose wasm shadow suspends into globalThis.occService and the export bytes go straight to a browser download (never entering the editor heap). STEP/IGES 3D models parse in the worker via the oce shadow (S3D WriteCache/ReadCache wire). - wasm/occ-service/: service CMake target (hooked from the kicad fork's top-level CMakeLists, wasm/editor pattern), embind entry (occExport/occLoadModel), wxConfig pre-js. - wasm/stubs/{exporter_step,oce_plugin}_stub.cpp: EM_ASYNC_JS worker bridges (callee-shadowing; no caller #ifdefs). - web/standalone: provider installed whenever the kicad_editor bundle boots (cross-face safe); ONE shared worker-boot source occ-worker.js (vite ?raw; the e2e stub reads the same file) — blob worker with locateFile absolutized against the glue URL; export download-name guard. - deps: OCC builds with RapidJSON so its glTF/GLB writer exists — pinned to the vcpkg master snapshot 2025-02-26 (24b5e7a8b27f), the same code official KiCad consumes via vcpkg.json's opencascade[rapidjson]; rapidjson's latest tag (v1.1.0, 2016) is ill-formed under modern clang. - tests: occ-export dialog e2e (lazy-fetch boundary + STEP download bytes), occ-probe incl. a 9-format matrix (step/stpz/brep/xao/ply/stl/glb/u3d/pdf), 3d-viewer-models hard-asserts the worker parse; occ provider stub installed ambiently by the kicad fixtures. Validated against desktop kicad-cli 10.0.4: geometric exact equality (bbox delta 0 um, volume delta 0.0000%) for STEP/GLB/STL/BREP/STPZ across three boards and option sweeps — with desktop OCC 7.9 vs wasm OCC 7.8; PLY/XAO/PDF structurally equal; U3D same-size (quantizer float LSBs differ). Full kicad e2e green on Firefox and Chromium; standalone verified end to end (lazy fetch only on the Export click; export.step 60,628 B ISO-10303-21; loadModel 700 KB STEP -> 569 KB scenegraph cache). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:39:58 +02:00
${OCC_SERVICE_CMAKE_FLAG} \
-DCMAKE_BUILD_TYPE=${BUILD_TYPE} \
-DCMAKE_INSTALL_PREFIX="${SYSROOT}" \
-DCMAKE_MODULE_PATH="${WASM_LAYER}/cmake" \
-DSYSROOT="${SYSROOT}" \
-DCMAKE_POLICY_VERSION_MINIMUM=3.5 \
jspi: migration phases 0-7 — build knob, scheduler shim, test successor suite Toolchain: emsdk 6.0.6 (versions.sh; cache-hash keys on it). Build knob PCBJAM_ASYNC_BACKEND=jspi|asyncify: build-kicad-target.sh links editors with -sJSPI + -sJSPI_EXPORTS=@scripts/common/jspi-exports.txt + --pre-js jspi-scheduler.js (no DYNCALLS, no post-link asyncify pipeline); wx build stamps the backend and forces clean on flip or unknown provenance; docker/build.sh passes the knob, seeds the emscripten ports cache from the volume every launch, jspi postprocess = patch-env-shim only. scripts/common/shims/jspi-scheduler.js: the JSPI successor scheduler — token-wait registry, resume turnstile (one armed resume between engine re-entries, SP swaps only at microtask boundaries), green-region spill stacks (16-aligned tops), S1 embind mutator FIFO lane + parker wraps, S6 shutdown, libctx integration hooks (suspend/end/quarantine + g_current arm/clear), SuspendError attributor, lost-wake + stuck-window watchdogs, __wxWaitDump observability. Embind: PARKER registrations get emscripten::async() under PCBJAM_JSPI (wasm/bindings/pcbjam_async_policy.h). nanosleep yields route via the shim. Tests: tests/asyncify -> tests/jspi successor suite (jspi-stack red/green shadow-stack battery, jspi-coroutine MiniCoro harness, suspend-races semantic scenarios + __wxWaitDump books coherence); projects jspi-firefox/ jspi-chrome (asyncify-webkit retired — no JSPI in WebKit); unconditional Firefox JSPI pref; guard-beacons -> wait-beacons (+wxScheduler/libctxJspi families); Makefile.wasm links test apps against JSPI with the shim as a tracked link prerequisite. Web: WasmTool setRo await + __wxWaitDump forensics, open-flow contained promise, scheduler-shim.test.ts retargeted (8 green). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NDeBaKKhQztd8KiVtHuyXr
2026-08-13 07:06:24 +02:00
-DCMAKE_CXX_FLAGS="${EXTRA_FLAGS} -Xclang -fno-pch-timestamp -pthread --use-port=zlib -DKICAD_USE_PLATFORM_WASM=1${DIAG_DEFINES} -I${SYSROOT}/include -I${STUBS_DIR} -include ${STUBS_DIR}/char_traits_uint16_workaround.h" \
-DCMAKE_C_FLAGS="${EXTRA_FLAGS} -pthread --use-port=zlib -I${SYSROOT}/include -I${STUBS_DIR}" \
-DCMAKE_EXE_LINKER_FLAGS="${LINKER_DEBUG_FLAGS} -pthread ${ASYNC_LINK_FLAGS} -sUSE_PTHREADS=1 -sMALLOC=mimalloc -sPTHREAD_POOL_SIZE='${PTHREAD_POOL_EXPR}' -sPTHREAD_POOL_SIZE_STRICT=0 -sALLOW_MEMORY_GROWTH=1 -sINITIAL_MEMORY=256MB -sMAXIMUM_MEMORY=4GB -sMAX_WEBGL_VERSION=2 ${GL3D_LINK_FLAGS} ${NANOSLEEP_YIELD_LINK} ${MALLINFO_STUB_LINK} ${ASYNC_RUNTIME_METHODS} ${EMBIND_LINK_FLAG} -L${SYSROOT}/lib -L${KICAD_BUILD}/common -L${KICAD_BUILD}/common/gal ${STUBS_BUILD}/libgit2_stub.a ${STUBS_BUILD}/libcurl_stub.a${APP_STUB_LINK} ${STUBS_BUILD}/libnng_stub.a ${EMBIND_OBJ}" \
-DCMAKE_SHARED_LINKER_FLAGS="-Wl,--allow-multiple-definition" \
-DCMAKE_MODULE_LINKER_FLAGS="-Wl,--allow-multiple-definition" \
-DZLIB_LIBRARY="${EMSDK}/upstream/emscripten/cache/sysroot/lib/wasm32-emscripten/pic/libz.a" \
-DZLIB_INCLUDE_DIR="${EMSDK}/upstream/emscripten/cache/sysroot/include" \
-DCMAKE_PREFIX_PATH="${SYSROOT};${WX_BUILD}" \
-DwxWidgets_CONFIG_EXECUTABLE="${WX_BUILD}/wx-config" \
\
-DKICAD_BUILD_QA_TESTS=OFF \
-DKICAD_USE_EGL=OFF \
-DKICAD_USE_BUNDLED_GLEW=ON \
-DKICAD_BUILD_3D_VIEWER_WASM=${BUILD_3D_VIEWER} \
-DKICAD_IPC_API=ON \
-DKICAD_USE_PCH=ON \
\
-DZSTD_ROOT="${SYSROOT}" \
-DZSTD_INCLUDE_DIR="${SYSROOT}/include" \
-DZSTD_LIBRARY="${SYSROOT}/lib/libzstd.a" \
-DGLM_INCLUDE_DIR="${SYSROOT}/include" \
-DGLM_VERSION="0.9.9.8" \
-DBOOST_ROOT="${SYSROOT}" \
-DBoost_INCLUDE_DIR="${SYSROOT}/include" \
-DBoost_LIBRARY_DIR="${SYSROOT}/lib" \
-DBoost_NO_SYSTEM_PATHS=ON \
-DBoost_NO_BOOST_CMAKE=ON \
-DFREETYPE_INCLUDE_DIR_ft2build="${SYSROOT}/include/freetype2" \
-DFREETYPE_INCLUDE_DIR_freetype2="${SYSROOT}/include/freetype2" \
-DFREETYPE_LIBRARY="${SYSROOT}/lib/libfreetype.a" \
-DHarfBuzz_INCLUDE_DIR="${SYSROOT}/include/harfbuzz" \
-DHarfBuzz_LIBRARY="${SYSROOT}/lib/libharfbuzz.a" \
-DOCC_INCLUDE_DIR="${SYSROOT}/include/opencascade" \
-DOCC_LIBRARY_DIR="${SYSROOT}/lib" \
-DProtobuf_INCLUDE_DIR="${SYSROOT}/include" \
-DProtobuf_LIBRARY="${SYSROOT}/lib/libprotobuf.a" \
-DProtobuf_LITE_LIBRARY="${SYSROOT}/lib/libprotobuf-lite.a" \
-DProtobuf_PROTOC_EXECUTABLE="${SYSROOT}/bin/protoc" \
-DODBC_CONFIG:STRING="stub-for-wasm" \
-DODBCLIB:STRING="" \
-DODBC_CFLAGS:STRING="" \
-DODBC_LINK_FLAGS:STRING="" \
-DODBC_LIBRARIES:STRING="" \
\
-DHAVE_STRCASECMP=1 \
-DHAVE_STRNCASECMP=1
# Step 7.1: Compile Embind bindings (after CMake so config.h exists)
# Exposes KiCad objects to JavaScript for future Pyodide integration.
# When no app-specific source exists, build an empty object so the linker line
# referencing ${APP_NAME}_embind.o doesn't break.
feat(wasm): kicad_editor — merge the pcbnew+eeschema kifaces into ONE bundle (Part 2) All four editors (PCB / Footprint / Schematic / Symbol) are now runtime --frame choices of a single kicad_editor.wasm (178 MB at -O1 vs 147+82 separate; shared wx/common/boost linked once). One editor per page load, as before; frames pcb / fpedit / sch / symedit. - wasm/editor/: the merged executable target (single_top + both kiface library sets, whole-archive pcbcommon) + the safety-net focus-walk Kiface() dispatch TU. Gated by KICAD_WASM_MERGED_EDITOR (kicad submodule bump carries the fork side: per-engine Kiface/getter binding + ODR renames + dual-kiface launcher). - wasm/bindings/: per-editor collab entries renamed pcbCollab*/schCollab* (JS names unchanged); duplicate kicadOpenFile/kicadCollabOnSave + shared-name registrations guarded behind KICAD_MERGED_EMBIND; new kicad_editor_embind.cpp registers each shared JS name once, dispatching on the live frame. - Build: kicad_editor app (build wrapper, target case arms, 3-object embind compile with the ABI-critical flags, STUB_APP=pcbnew); docker/build.sh "all" = kicad_editor calculator pl_editor gerbview (pcbnew/eeschema stay as explicit debug apps); scripts/kicad/audit-merged-symbols.sh = repeatable ODR-collision audit (run on kicad bumps). - Frontend: Bundle type (bundle ≠ tool); TOOL_BUNDLE maps all four editors to kicad_editor; explicit --frame tokens for pcbnew (pcb) and eeschema (sch); publish list = the 4 real bundles. - Tests/CI: five harnesses load kicad_editor.js with explicit frame tokens; PCBNEW_FAMILY_SPECS renamed BIG_MODULE_SPECS + the 8 eeschema-family specs (they now boot the merged module — SpiderMonkey x86 CI OOM routing); frame-runtime spec covers all four frames from the one bundle. Validated so far: frame-runtime 4/4 (each frame boots with the right title, no aborts, no duplicate embind registration); 24-spec merged-module regression green; 3D raytracer renders. Known pre-existing failure: 3d-viewer title-bar drag deadlock, fixed on main by 7630c7e (2N+8 pthread pre-warm) — picked up by the follow-up rebase. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 14:48:11 +02:00
# Compile one embind TU with the same includes and ABI-critical flags KiCad uses
# (KICAD_TU_ABI_FLAGS must match the core's -DDEBUG state or vtable dispatch skews —
# task #54). Args: <src> <obj> <app-include-subdir> [extra-defines]
compile_embind_tu() {
local _src="$1" _obj="$2" _subdir="$3" _defines="${4:-}"
local _includes="-I${KICAD_BUILD} -I${KICAD_DIR}/include -I${KICAD_DIR}/${_subdir} -I${KICAD_DIR}/common"
# Generated DSN-lexer headers (e.g. pcb_lexer.h, used transitively via kicad_clipboard.h →
# pcb_io_kicad_sexpr_parser.h) are emitted into the common build subdir by make_lexer.
_includes+=" -I${KICAD_BUILD}/common"
_includes+=" -I${KICAD_DIR}/libs/core/include -I${KICAD_DIR}/libs/kimath/include -I${KICAD_DIR}/libs/kiplatform/include"
_includes+=" -I${KICAD_DIR}/thirdparty/clipper2/Clipper2Lib/include"
_includes+=" -I${KICAD_DIR}/thirdparty/nlohmann_json"
_includes+=" -I${KICAD_DIR}/thirdparty/expected/include"
_includes+=" -I${KICAD_DIR}/thirdparty/rtree"
_includes+=" -I${KICAD_DIR}/thirdparty/fmt"
_includes+=" -I${KICAD_DIR}/thirdparty/dynamic_bitset"
_includes+=" -I${KICAD_DIR}/thirdparty/nanodbc"
_includes+=" -I${KICAD_DIR}/thirdparty/picosha2"
_includes+=" -I${KICAD_DIR}/thirdparty"
# libcontext.h lives one level deeper; tool/coroutine.h does #include <libcontext.h>
_includes+=" -I${KICAD_DIR}/thirdparty/libcontext"
_includes+=" -I${SYSROOT}/include"
# KiCad requires C++20 for concepts
em++ -std=c++20 -c ${EXTRA_FLAGS} ${KICAD_TU_ABI_FLAGS} ${WX_CXXFLAGS} ${_defines} ${_includes} "${_src}" -o "${_obj}"
}
if [ "${APP_NAME}" = "kicad_editor" ]; then
# Merged image: both per-editor TUs (with -DKICAD_MERGED_EMBIND compiling out
# their duplicate definitions / shared-name registrations) + the dispatcher TU
# registering the shared JS names once (wasm/bindings/kicad_editor_embind.cpp).
# pcbnew's TU needs the generated lexer headers — pre-build pcbcommon (no wasted
# work: kicad_editor depends on pcbcommon anyway; incremental no-op).
log_info "Pre-building pcbcommon so generated lexer headers exist for the embind compile..."
emmake make -j${JOBS} pcbcommon
log_info "Compiling merged Embind bindings (pcbnew + eeschema + dispatcher)..."
compile_embind_tu "${PROJECT_ROOT}/wasm/bindings/pcbnew_embind.cpp" \
"${STUBS_BUILD}/pcbnew_embind.o" pcbnew "-DKICAD_MERGED_EMBIND"
compile_embind_tu "${PROJECT_ROOT}/wasm/bindings/eeschema_embind.cpp" \
"${STUBS_BUILD}/eeschema_embind.o" eeschema "-DKICAD_MERGED_EMBIND"
compile_embind_tu "${PROJECT_ROOT}/wasm/bindings/kicad_editor_embind.cpp" \
"${STUBS_BUILD}/kicad_editor_embind.o" common
elif [ -f "${EMBIND_SRC}" ]; then
# pcbnew's embind TU transitively includes generated lexer headers
# (kicad_clipboard.h → pcb_io_kicad_sexpr_parser.h → pcb_lexer.h, emitted into
# ${KICAD_BUILD}/common by make_lexer custom commands on the pcbcommon target).
# On a fresh build dir they don't exist until make runs — build pcbcommon first.
# No wasted work: the app target depends on pcbcommon anyway; incremental no-op.
# Guard on EMBIND_APP (not APP_NAME) so any app whose embind IS pcbnew's also
# pre-builds pcbcommon.
if [ "${EMBIND_APP}" = "pcbnew" ]; then
log_info "Pre-building pcbcommon so generated lexer headers exist for the embind compile..."
emmake make -j${JOBS} pcbcommon
fi
log_info "Compiling Embind bindings (${APP_NAME})..."
feat(wasm): occ-split — lazy occ_service worker; kicad_editor drops OCC (−31%) Move OpenCASCADE out of the merged editor image into occ_service: a separate emscripten module (-sASYNCIFY=0, MODULARIZE, in-container -Oz finalize, 2N+8 pre-warmed pthread pool) booted lazily in a dedicated Web Worker on the first STEP export or STEP/IGES model parse. kicad_editor.wasm ~190 MB -> 130 MB; sessions that never touch OCC never fetch its 57 MB. STEP export works in the browser for the first time: the unchanged desktop dialog runs EXPORTER_STEP, whose wasm shadow suspends into globalThis.occService and the export bytes go straight to a browser download (never entering the editor heap). STEP/IGES 3D models parse in the worker via the oce shadow (S3D WriteCache/ReadCache wire). - wasm/occ-service/: service CMake target (hooked from the kicad fork's top-level CMakeLists, wasm/editor pattern), embind entry (occExport/occLoadModel), wxConfig pre-js. - wasm/stubs/{exporter_step,oce_plugin}_stub.cpp: EM_ASYNC_JS worker bridges (callee-shadowing; no caller #ifdefs). - web/standalone: provider installed whenever the kicad_editor bundle boots (cross-face safe); ONE shared worker-boot source occ-worker.js (vite ?raw; the e2e stub reads the same file) — blob worker with locateFile absolutized against the glue URL; export download-name guard. - deps: OCC builds with RapidJSON so its glTF/GLB writer exists — pinned to the vcpkg master snapshot 2025-02-26 (24b5e7a8b27f), the same code official KiCad consumes via vcpkg.json's opencascade[rapidjson]; rapidjson's latest tag (v1.1.0, 2016) is ill-formed under modern clang. - tests: occ-export dialog e2e (lazy-fetch boundary + STEP download bytes), occ-probe incl. a 9-format matrix (step/stpz/brep/xao/ply/stl/glb/u3d/pdf), 3d-viewer-models hard-asserts the worker parse; occ provider stub installed ambiently by the kicad fixtures. Validated against desktop kicad-cli 10.0.4: geometric exact equality (bbox delta 0 um, volume delta 0.0000%) for STEP/GLB/STL/BREP/STPZ across three boards and option sweeps — with desktop OCC 7.9 vs wasm OCC 7.8; PLY/XAO/PDF structurally equal; U3D same-size (quantizer float LSBs differ). Full kicad e2e green on Firefox and Chromium; standalone verified end to end (lazy fetch only on the Export click; export.step 60,628 B ISO-10303-21; loadModel 700 KB STEP -> 569 KB scenegraph cache). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:39:58 +02:00
# The include home is the app the BINDINGS belong to, not the artifact
# subdir — for occ_service the two differ (bindings = pcbnew's; artifacts
# land in the wasm/occ-service target's own occ_service/ binary dir, which
# has no KiCad sources).
case "${EMBIND_APP}" in
pcbnew) EMBIND_INC_SUBDIR="pcbnew" ;;
eeschema) EMBIND_INC_SUBDIR="eeschema" ;;
*) EMBIND_INC_SUBDIR="${KICAD_SUBDIR}" ;;
esac
compile_embind_tu "${EMBIND_SRC}" "${EMBIND_OBJ}" "${EMBIND_INC_SUBDIR}"
else
log_info "No embind source for ${APP_NAME} (expected at ${EMBIND_SRC}); using empty placeholder"
EMPTY_C="${STUBS_BUILD}/${APP_NAME}_embind_empty.c"
: > "${EMPTY_C}"
emcc -c "${EMPTY_C}" -o "${EMBIND_OBJ}"
fi
# Step 8: Build the app target
2026-06-01 17:31:03 +02:00
kw_stage kicad-compile
refactor(build): fold calculator into unified docker/build.sh dispatch Convert the calculator port (commit 0464470) from the parallel/copy-paste pattern to the dispatch pattern used by pcbnew + eeschema. - docker/build.sh: <app> is now required (no pcbnew default); missing, unknown, and --help paths all print "pcbnew | eeschema | calculator | all". Adds calculator to build_app() via kicad_subdir_for() (pcb_calculator inner subdir, calculator.* output basename). Also fixes the EMSDK=/emsdkkicad-wasm-builder typo that snuck in during the calc merge. - scripts/kicad/build-kicad-target.sh: accepts calculator; introduces KICAD_TARGET (pcb_calculator for calc, == APP_NAME otherwise) used for the make target, embind include path, and final-log line. - scripts/kicad/build-calculator.sh: 305-line copy of build-pcbnew.sh collapsed to a 7-line wrapper around build-kicad-target.sh. - tests/scripts/setup-kicad-wasm.sh: copy_app calculator added with the same pcb_calculator subdir mapping for the docker-volume fallback path. - tests/package.json: test:calculator* routed through the shared playwright-kicad.config.ts kicad/calculator.spec.ts (mirrors eeschema). - Delete: docker/build-calculator.sh, tests/scripts/setup-calculator-wasm.sh, tests/playwright-calculator.config.ts. Bumps wxwidgets d1d1627 -> 6fb2eac (origin/wasm-port). The new sha includes "unic/combobox: add GetCurrentSelection() inline default" which calc's kicad/pcb_calculator/widgets/unit_selector.cpp needs to compile. Without this bump the unified dispatch would expose calc as a buildable target but the build itself would fail. Verified: build.sh all completes clean across all three apps; pcbnew/eeschema e2e pass; calc compiles and launches (test stability separate from this refactor). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-29 12:01:44 +02:00
log_info "Building ${APP_NAME} (CMake target: ${KICAD_TARGET})..."
2026-06-03 17:49:11 +02:00
# The embind object (step 7.1) is injected via linker flags, NOT tracked as a CMake/
# make dependency — so when ONLY <app>_embind.cpp changes, make sees no changed source,
# skips the link, and the new bindings silently vanish from the binary. Force a relink
# whenever the freshly-compiled embind object is newer than the linked output.
LINK_OUT_JS="${KICAD_BUILD}/${KICAD_SUBDIR}/${APP_NAME}.js"
LINK_OUT_WASM="${KICAD_BUILD}/${KICAD_SUBDIR}/${APP_NAME}.wasm"
feat(wasm): kicad_editor — merge the pcbnew+eeschema kifaces into ONE bundle (Part 2) All four editors (PCB / Footprint / Schematic / Symbol) are now runtime --frame choices of a single kicad_editor.wasm (178 MB at -O1 vs 147+82 separate; shared wx/common/boost linked once). One editor per page load, as before; frames pcb / fpedit / sch / symedit. - wasm/editor/: the merged executable target (single_top + both kiface library sets, whole-archive pcbcommon) + the safety-net focus-walk Kiface() dispatch TU. Gated by KICAD_WASM_MERGED_EDITOR (kicad submodule bump carries the fork side: per-engine Kiface/getter binding + ODR renames + dual-kiface launcher). - wasm/bindings/: per-editor collab entries renamed pcbCollab*/schCollab* (JS names unchanged); duplicate kicadOpenFile/kicadCollabOnSave + shared-name registrations guarded behind KICAD_MERGED_EMBIND; new kicad_editor_embind.cpp registers each shared JS name once, dispatching on the live frame. - Build: kicad_editor app (build wrapper, target case arms, 3-object embind compile with the ABI-critical flags, STUB_APP=pcbnew); docker/build.sh "all" = kicad_editor calculator pl_editor gerbview (pcbnew/eeschema stay as explicit debug apps); scripts/kicad/audit-merged-symbols.sh = repeatable ODR-collision audit (run on kicad bumps). - Frontend: Bundle type (bundle ≠ tool); TOOL_BUNDLE maps all four editors to kicad_editor; explicit --frame tokens for pcbnew (pcb) and eeschema (sch); publish list = the 4 real bundles. - Tests/CI: five harnesses load kicad_editor.js with explicit frame tokens; PCBNEW_FAMILY_SPECS renamed BIG_MODULE_SPECS + the 8 eeschema-family specs (they now boot the merged module — SpiderMonkey x86 CI OOM routing); frame-runtime spec covers all four frames from the one bundle. Validated so far: frame-runtime 4/4 (each frame boots with the right title, no aborts, no duplicate embind registration); 24-spec merged-module regression green; 3D raytracer renders. Known pre-existing failure: 3d-viewer title-bar drag deadlock, fixed on main by 7630c7e (2N+8 pthread pre-warm) — picked up by the follow-up rebase. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-02 14:48:11 +02:00
# EMBIND_OBJ may hold several objects (kicad_editor) — check each.
for _embind_obj in ${EMBIND_OBJ}; do
if [ -f "${_embind_obj}" ] && [ -f "${LINK_OUT_JS}" ] && [ "${_embind_obj}" -nt "${LINK_OUT_JS}" ]; then
log_info "Embind object newer than ${APP_NAME}.js — forcing relink to pick up new bindings"
rm -f "${LINK_OUT_JS}" "${LINK_OUT_WASM}"
break
fi
done
2026-06-03 17:49:11 +02:00
refactor(build): fold calculator into unified docker/build.sh dispatch Convert the calculator port (commit 0464470) from the parallel/copy-paste pattern to the dispatch pattern used by pcbnew + eeschema. - docker/build.sh: <app> is now required (no pcbnew default); missing, unknown, and --help paths all print "pcbnew | eeschema | calculator | all". Adds calculator to build_app() via kicad_subdir_for() (pcb_calculator inner subdir, calculator.* output basename). Also fixes the EMSDK=/emsdkkicad-wasm-builder typo that snuck in during the calc merge. - scripts/kicad/build-kicad-target.sh: accepts calculator; introduces KICAD_TARGET (pcb_calculator for calc, == APP_NAME otherwise) used for the make target, embind include path, and final-log line. - scripts/kicad/build-calculator.sh: 305-line copy of build-pcbnew.sh collapsed to a 7-line wrapper around build-kicad-target.sh. - tests/scripts/setup-kicad-wasm.sh: copy_app calculator added with the same pcb_calculator subdir mapping for the docker-volume fallback path. - tests/package.json: test:calculator* routed through the shared playwright-kicad.config.ts kicad/calculator.spec.ts (mirrors eeschema). - Delete: docker/build-calculator.sh, tests/scripts/setup-calculator-wasm.sh, tests/playwright-calculator.config.ts. Bumps wxwidgets d1d1627 -> 6fb2eac (origin/wasm-port). The new sha includes "unic/combobox: add GetCurrentSelection() inline default" which calc's kicad/pcb_calculator/widgets/unit_selector.cpp needs to compile. Without this bump the unified dispatch would expose calc as a buildable target but the build itself would fail. Verified: build.sh all completes clean across all three apps; pcbnew/eeschema e2e pass; calc compiles and launches (test stability separate from this refactor). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-29 12:01:44 +02:00
emmake make -j${JOBS} "${KICAD_TARGET}"
# Step 8.1: Build bitmap resources (images.tar.gz)
# This creates the icon archive that KiCad loads at runtime. The headless
feat(wasm): occ-split — lazy occ_service worker; kicad_editor drops OCC (−31%) Move OpenCASCADE out of the merged editor image into occ_service: a separate emscripten module (-sASYNCIFY=0, MODULARIZE, in-container -Oz finalize, 2N+8 pre-warmed pthread pool) booted lazily in a dedicated Web Worker on the first STEP export or STEP/IGES model parse. kicad_editor.wasm ~190 MB -> 130 MB; sessions that never touch OCC never fetch its 57 MB. STEP export works in the browser for the first time: the unchanged desktop dialog runs EXPORTER_STEP, whose wasm shadow suspends into globalThis.occService and the export bytes go straight to a browser download (never entering the editor heap). STEP/IGES 3D models parse in the worker via the oce shadow (S3D WriteCache/ReadCache wire). - wasm/occ-service/: service CMake target (hooked from the kicad fork's top-level CMakeLists, wasm/editor pattern), embind entry (occExport/occLoadModel), wxConfig pre-js. - wasm/stubs/{exporter_step,oce_plugin}_stub.cpp: EM_ASYNC_JS worker bridges (callee-shadowing; no caller #ifdefs). - web/standalone: provider installed whenever the kicad_editor bundle boots (cross-face safe); ONE shared worker-boot source occ-worker.js (vite ?raw; the e2e stub reads the same file) — blob worker with locateFile absolutized against the glue URL; export download-name guard. - deps: OCC builds with RapidJSON so its glTF/GLB writer exists — pinned to the vcpkg master snapshot 2025-02-26 (24b5e7a8b27f), the same code official KiCad consumes via vcpkg.json's opencascade[rapidjson]; rapidjson's latest tag (v1.1.0, 2016) is ill-formed under modern clang. - tests: occ-export dialog e2e (lazy-fetch boundary + STEP download bytes), occ-probe incl. a 9-format matrix (step/stpz/brep/xao/ply/stl/glb/u3d/pdf), 3d-viewer-models hard-asserts the worker parse; occ provider stub installed ambiently by the kicad fixtures. Validated against desktop kicad-cli 10.0.4: geometric exact equality (bbox delta 0 um, volume delta 0.0000%) for STEP/GLB/STL/BREP/STPZ across three boards and option sweeps — with desktop OCC 7.9 vs wasm OCC 7.8; PLY/XAO/PDF structurally equal; U3D same-size (quantizer float LSBs differ). Full kicad e2e green on Firefox and Chromium; standalone verified end to end (lazy fetch only on the Export click; export.step 60,628 B ISO-10303-21; loadModel 700 KB STEP -> 569 KB scenegraph cache). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-03 12:39:58 +02:00
# converter/service targets have no GUI/icons, so skip it.
if [ "${APP_NAME}" != "kicad_tools" ] && [ "${APP_NAME}" != "occ_service" ]; then
kw_stage kicad-bitmaps
log_info "Building bitmap resources..."
emmake make bitmap_archive_build
fi
# Step 9: Create stamp file
create_stamp "${KICAD_STAMP}"
log_info "KiCad ${APP_NAME} build complete!"
2026-06-02 13:35:37 +02:00
log_info "Output: ${KICAD_BUILD}/${KICAD_SUBDIR}/${APP_NAME}.js"