pcbjam/scripts/build-wx-wasm.sh

258 lines
11 KiB
Shell
Raw Normal View History

#!/bin/bash
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
# Build wxWidgets for WebAssembly (the DOM port: widgets are real HTML
# elements; owner-drawn widgets render into per-window canvas islands)
# Redirect all output to a log file (re-execs script with redirection)
source "$(dirname "$0")/common/logging.sh"
# Source common environment (sets up local emsdk)
source "$(dirname "$0")/common/env.sh"
2026-06-01 17:31:03 +02:00
# Build-progress markers (parsed by scripts/build-monitor.sh).
source "$(dirname "$0")/common/stages.sh"
# This builds the GUI-enabled wxWidgets needed for KiCad
#
# Prerequisites:
# - Emscripten SDK (auto-installed by env.sh via scripts/setup-emsdk.sh)
# - autoconf (for regenerating configure from configure.in)
#
# To regenerate Makefile.in from bakefiles (after modifying files.bkl):
# cd wxwidgets/build/bakefiles
# docker run --rm -v "$(pwd)/../..":"$(pwd)/../.." -w "$(pwd)" \
# ghcr.io/vslavik/bakefile:0.2 bakefile_gen
#
# Usage:
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
# ./build-wx-wasm.sh # Incremental build (default)
# ./build-wx-wasm.sh --clean # Clean build from scratch
set -e
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
PROJECT_ROOT="$(dirname "$SCRIPT_DIR")"
WX_SOURCE="$PROJECT_ROOT/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
# Parse arguments: --clean or --no-clean (default; kept for the kicad
# pipeline's explicit call), in any order.
CLEAN_BUILD=0
for arg in "$@"; do
case "$arg" in
--clean) CLEAN_BUILD=1 ;;
--no-clean) CLEAN_BUILD=0 ;;
*) echo "Unknown argument: $arg"; exit 1 ;;
esac
done
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
BUILD_DIR="$PROJECT_ROOT/build-wasm/wxwidgets"
WXLIB_PREFIX="libwx_wasmu"
# Use our config.sub wrapper for autoconf projects
# CONFIG_SHELL is critical: nested configures (pcre, etc.) do SHELL=${CONFIG_SHELL-/bin/sh}
# Without CONFIG_SHELL, nested configures would reset SHELL to /bin/sh and bypass our wrapper
export SHELL="$SCRIPT_DIR/config/config-sub-wrapper.sh"
export CONFIG_SHELL="$SCRIPT_DIR/config/config-sub-wrapper.sh"
# Disable autom4te cache to keep submodules clean
export AUTOM4TE="$SCRIPT_DIR/config/autom4te-wrapper.sh"
# Use JOBS from env.sh if set, otherwise use all available cores
JOBS="${JOBS:-$(nproc 2>/dev/null || sysctl -n hw.ncpu 2>/dev/null || echo 4)}"
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
echo "=== Building wxWidgets for WASM ==="
echo "Project root: $PROJECT_ROOT"
echo "Build dir: $BUILD_DIR"
echo "wxWidgets source: $WX_SOURCE"
# Verify we're in the right place
if [ ! -f "$WX_SOURCE/configure.in" ]; then
echo "ERROR: wxWidgets source not found at $WX_SOURCE"
echo "Make sure the wxwidgets submodule is initialized"
exit 1
fi
# Regenerate configure if configure.in or autoconf_inc.m4 is newer.
# (configure.in sincludes autoconf_inc.m4, which bakefile regenerates from
# build/bakefiles/files.bkl — new build conditions live there.)
if [ "$WX_SOURCE/configure.in" -nt "$WX_SOURCE/configure" ] || \
[ "$WX_SOURCE/autoconf_inc.m4" -nt "$WX_SOURCE/configure" ]; then
echo "configure inputs changed, regenerating configure..."
(cd "$WX_SOURCE" && autoconf)
fi
# Incremental build by default, use --clean for full rebuild
if [ "$CLEAN_BUILD" = "1" ]; then
echo "Cleaning build directory..."
rm -rf "$BUILD_DIR"
fi
# Create build directory
mkdir -p "$BUILD_DIR"
cd "$BUILD_DIR"
# Determine if we need to run configure
# Skip configure if:
# 1. Makefile exists (already configured)
# 2. configure.in hasn't changed since last configure
NEEDS_CONFIGURE=0
if [ ! -f "$BUILD_DIR/Makefile" ]; then
echo "Not configured yet, will run configure..."
NEEDS_CONFIGURE=1
elif [ "$WX_SOURCE/configure.in" -nt "$BUILD_DIR/Makefile" ]; then
echo "configure.in changed since last configure, will reconfigure..."
NEEDS_CONFIGURE=1
elif [ "$WX_SOURCE/configure" -nt "$BUILD_DIR/Makefile" ]; then
echo "configure script changed, will reconfigure..."
NEEDS_CONFIGURE=1
elif [ "$WX_SOURCE/Makefile.in" -nt "$BUILD_DIR/Makefile" ]; then
# Makefile.in is regenerated from build/bakefiles/files.bkl (see header);
# the build Makefile must be re-derived or new source files are ignored.
echo "Makefile.in changed since last configure, will reconfigure..."
NEEDS_CONFIGURE=1
else
echo "Already configured, skipping configure (use clean build to reconfigure)"
fi
if [ $NEEDS_CONFIGURE -eq 1 ]; then
# Configure with emconfigure
# Key flags based on wxWidgets-wasm:
# --host=emscripten Host system (detected via config.sub)
# --disable-shared Build static libraries
# --with-opengl Enable OpenGL/WebGL support
# --enable-exceptions Enable C++ exceptions (needed for KiCad debug builds)
# --disable-richtext Not needed for KiCad, simplifies build
# --without-libtiff Avoid external dependencies
# --disable-xlocale Browser environment handles locale
2026-06-01 17:31:03 +02:00
kw_stage wxwidgets-configure
echo ""
echo "=== Configuring ==="
# Regenerate autotools files in bundled PCRE to fix version mismatch
# The bundled PCRE was generated with automake 1.16.1 but build systems
# may have different versions. Running autoreconf ensures compatibility.
if command -v autoreconf &> /dev/null; then
echo "Regenerating autotools files for bundled PCRE..."
(cd "$WX_SOURCE/3rdparty/pcre" && autoreconf -fi 2>/dev/null || true)
fi
# Ensure Emscripten's zlib port is built (works in Docker and on host)
# This populates the cache sysroot with zlib.h and libz.a
echo "Building Emscripten zlib port..."
embuilder build zlib
# Get Emscripten cache sysroot path (portable across environments)
EM_CACHE_SYSROOT="$(em-config CACHE)/sysroot"
echo "Emscripten cache sysroot: $EM_CACHE_SYSROOT"
# Set flags for Emscripten compatibility
# Z_HAVE_UNISTD_H ensures zlib includes <unistd.h> for read/write/lseek
# Include pcre2 headers from the build directory (generated during configure)
PCRE2_INCLUDE="$BUILD_DIR/3rdparty/pcre/src"
# Configure debug/release flags based on DEBUG_BUILD environment variable
if [ "${DEBUG_BUILD:-1}" = "1" ]; then
WX_DEBUG_FLAGS="-g -O1"
WX_CONFIGURE_DEBUG="--enable-debug"
echo "Building wxWidgets in DEBUG mode"
else
WX_DEBUG_FLAGS="-O2"
WX_CONFIGURE_DEBUG=""
echo "Building wxWidgets in RELEASE mode"
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
# Exception model: native WebAssembly exceptions (legacy binary encoding) + wasm setjmp/longjmp,
# single-sourced from scripts/common/env.sh. The catch-arm-hoisting pass (run post-link, see
# build-wasm-test.sh) lets Asyncify suspend from inside C++ catch blocks. See docs/features/wasm-exceptions/.
WX_EH_FLAGS="$DEPS_EH_FLAGS"
echo "wx EH model flags: ${WX_EH_FLAGS}"
# Include emscripten cache sysroot for zlib headers
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
export CFLAGS="-DZ_HAVE_UNISTD_H=1 -I$EM_CACHE_SYSROOT/include ${WX_DEBUG_FLAGS} ${WX_EH_FLAGS} -pthread -matomics -mbulk-memory"
export CXXFLAGS="-DZ_HAVE_UNISTD_H=1 -I$EM_CACHE_SYSROOT/include -I$PCRE2_INCLUDE ${WX_DEBUG_FLAGS} ${WX_EH_FLAGS} -pthread -matomics -mbulk-memory"
export LDFLAGS="-L$EM_CACHE_SYSROOT/lib/wasm32-emscripten"
emconfigure "$WX_SOURCE/configure" \
--host=emscripten \
--without-subdirs \
--disable-shared \
--with-opengl \
--enable-exceptions \
--disable-richtext \
--without-libtiff \
--disable-xlocale \
--with-cxx=17 \
--enable-utf8 \
--with-zlib=sys \
${WX_CONFIGURE_DEBUG}
# Build PCRE first to avoid race condition with parallel builds
# PCRE headers (pcre2.h) must be generated before regex.cpp compiles
echo ""
echo "=== Building PCRE first (dependency) ==="
emmake make -C 3rdparty/pcre
fi
# Ensure the Emscripten zlib port exists before compiling. We configure with
# --with-zlib=sys, which resolves zlib.h/libz.a to the emsdk cache sysroot —
# populated only by `embuilder build zlib` in the configure branch above. When
# the build dir arrives pre-configured (e.g. CI restores build-wasm/wxwidgets
# from cache onto a runner with a freshly installed emsdk), configure is
# skipped and any recompile of a zlib-using TU (zipstrm.cpp, zstream.cpp, ...)
# fails with "'zlib.h' file not found" — deterministically, so the serial
# retry below can't clear it.
EM_CACHE_SYSROOT="$(em-config CACHE)/sysroot"
if [ ! -f "$EM_CACHE_SYSROOT/include/zlib.h" ]; then
echo "Emscripten zlib port missing from $EM_CACHE_SYSROOT; building it..."
# --force: embuilder stamps the port as built on libz.a alone, so a
# half-populated cache (lib present, headers gone) would no-op without it.
embuilder build zlib --force
fi
# Build wxWidgets
2026-06-01 17:31:03 +02:00
kw_stage wxwidgets-compile
echo ""
echo "=== Building wxWidgets (using ${JOBS} parallel jobs) ==="
# A clean -jN build occasionally fails non-deterministically: a burst of GUI
# translation units (toplevel.cpp, dirdlgg.cpp, the generic colour/dir dialogs,
# ...) all abort at once with bogus "incomplete type 'wxBitmap'" / "wxIcon has
# no member IsOk" / "unknown type 'wxTranslations'" errors, while neighbouring
# files compile fine. It never happens at -j1 (the host default) — it's a
# parallel-build race over a generated/regenerated header (config.status can
# re-emit wx/setup.h, and emscripten warms its cache on first use), so some
# compiles read a file mid-rewrite. The same source builds cleanly on a retry.
# Rather than serialize the whole (slow) build, fall back to a serial pass only
# when the parallel one trips: it resumes from the already-built objects, so it
# just recompiles the few that lost the race. A genuine error still fails the
# -j1 pass and surfaces. (Same spirit as the serial PCRE pre-build above.)
if ! emmake make -j${JOBS}; then
echo ""
echo "=== Parallel build failed; retrying serially to clear the clean-build race ==="
emmake make -j1
fi
# Create library symlinks (remove -emscripten suffix for CMake compatibility)
echo ""
echo "=== Creating library symlinks ==="
cd "$BUILD_DIR/lib"
for lib in *-emscripten.a; do
# Skip symlinks - only process real files (stub symlinks are handled below)
if [ -f "$lib" ] && [ ! -L "$lib" ]; then
newname="${lib/-emscripten/}"
rm -f "$newname" # Remove existing symlink/file to avoid conflicts
ln -sf "$lib" "$newname"
fi
done
# Create stub libraries for components wx-config reports but we didn't build
# KiCad doesn't use these directly
echo "Creating stub libraries..."
for stub in richtext webview; do
# Remove any existing symlinks first to avoid "same file" errors
rm -f "${WXLIB_PREFIX}_${stub}-3.2.a" "${WXLIB_PREFIX}_${stub}-3.2-emscripten.a"
emar rcs "${WXLIB_PREFIX}_${stub}-3.2.a"
ln -sf "${WXLIB_PREFIX}_${stub}-3.2.a" "${WXLIB_PREFIX}_${stub}-3.2-emscripten.a"
done
cd "$BUILD_DIR"
echo ""
echo "=== Build complete ==="
ls -lh "$BUILD_DIR"/lib/*.a 2>/dev/null || echo "Libraries built in $BUILD_DIR/lib"