ledgrab/build-common.sh

#!/usr/bin/env bash
#
# Shared build functions for LedGrab distribution packaging.
# Sourced by build-dist.sh (Linux) and build-dist-windows.sh (Windows).
#
# Expected variables set by the caller before sourcing:
#   SCRIPT_DIR, BUILD_DIR, DIST_DIR, SERVER_DIR, APP_DIR

# ── Version detection ────────────────────────────────────────

detect_version() {
    # Usage: detect_version [explicit_version]
    local version="${1:-}"

    if [ -z "$version" ]; then
        version=$(git describe --tags --exact-match 2>/dev/null || true)
    fi
    if [ -z "$version" ]; then
        version="${GITEA_REF_NAME:-${GITHUB_REF_NAME:-}}"
    fi
    if [ -z "$version" ]; then
        version=$(grep -oP '^version\s*=\s*"\K[^"]+' "$SERVER_DIR/pyproject.toml" 2>/dev/null || echo "0.0.0")
    fi

    VERSION_CLEAN="${version#v}"

    # Normalize non-PEP440 version labels (e.g. "dev", "nightly", "snapshot")
    # to a valid PEP440 dev release. Without this, pip/setuptools rejects the
    # pyproject.toml with: `project.version` must be pep440.
    if ! [[ "$VERSION_CLEAN" =~ ^[0-9]+(\.[0-9]+)*((a|b|rc|\.dev|\.post)[0-9]+)*(\+[a-zA-Z0-9.]+)?$ ]]; then
        echo "  Warning: '$VERSION_CLEAN' is not PEP440-compliant, using 0.0.0.dev0"
        VERSION_CLEAN="0.0.0.dev0"
    fi

    # Stamp the resolved version into pyproject.toml so that
    # importlib.metadata reads the correct value at runtime.
    sed -i "s/^version = .*/version = \"${VERSION_CLEAN}\"/" "$SERVER_DIR/pyproject.toml"
}

# ── Clean previous build ─────────────────────────────────────

clean_dist() {
    if [ -d "$DIST_DIR" ]; then
        echo "  Cleaning previous build..."
        rm -rf "$DIST_DIR"
    fi
    mkdir -p "$DIST_DIR"
}

# ── Build frontend ───────────────────────────────────────────

build_frontend() {
    echo "  Building frontend bundle..."
    (cd "$SERVER_DIR" && npm ci --loglevel error && npm run build) 2>&1 | {
        grep -v 'RemoteException' || true
    }
}

# ── Copy application files ───────────────────────────────────

copy_app_files() {
    echo "  Copying application files..."
    mkdir -p "$APP_DIR"

    cp -r "$SERVER_DIR/src" "$APP_DIR/src"
    cp -r "$SERVER_DIR/config" "$APP_DIR/config"
    mkdir -p "$DIST_DIR/data" "$DIST_DIR/logs"

    # Clean up source maps and __pycache__
    find "$APP_DIR" -name "*.map" -delete 2>/dev/null || true
    find "$APP_DIR" -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
}

# ── Site-packages cleanup ────────────────────────────────────
#
# Strips tests, type stubs, unused submodules, and debug symbols
# from the installed site-packages directory.
#
# Args:
#   $1 — path to site-packages directory
#   $2 — native extension suffix: "pyd" (Windows) or "so" (Linux)
#   $3 — native lib suffix for OpenCV ffmpeg: "dll" or "so"

cleanup_site_packages() {
    local sp_dir="$1"
    local ext_suffix="${2:-so}"
    local lib_suffix="${3:-so}"

    echo "  Cleaning up site-packages to reduce size..."

    # ── Generic cleanup ──────────────────────────────────────
    find "$sp_dir" -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
    find "$sp_dir" -type d -name tests -exec rm -rf {} + 2>/dev/null || true
    find "$sp_dir" -type d -name test -exec rm -rf {} + 2>/dev/null || true
    find "$sp_dir" -type d -name "*.dist-info" -exec rm -rf {} + 2>/dev/null || true
    find "$sp_dir" -name "*.pyi" -delete 2>/dev/null || true

    # ── pip / setuptools (not needed at runtime) ─────────────
    rm -rf "$sp_dir"/pip "$sp_dir"/pip-* 2>/dev/null || true
    rm -rf "$sp_dir"/setuptools "$sp_dir"/setuptools-* "$sp_dir"/pkg_resources 2>/dev/null || true
    rm -rf "$sp_dir"/_distutils_hack 2>/dev/null || true

    # ── OpenCV ───────────────────────────────────────────────
    local cv2_dir="$sp_dir/cv2"
    if [ -d "$cv2_dir" ]; then
        # Remove ffmpeg (28 MB on Windows), Haar cascades, dev files
        rm -f "$cv2_dir"/opencv_videoio_ffmpeg*."$lib_suffix" 2>/dev/null || true
        rm -rf "$cv2_dir/data" "$cv2_dir/gapi" "$cv2_dir/misc" "$cv2_dir/utils" 2>/dev/null || true
        rm -rf "$cv2_dir/typing_stubs" "$cv2_dir/typing" 2>/dev/null || true
    fi

    # ── NumPy ────────────────────────────────────────────────
    # Only strip modules that are safely unused by numpy's own import chain.
    # DO NOT strip: lib, linalg, ma, matrixlib — numpy.__init__ imports them
    # transitively (e.g. matrixlib → defmatrix → linalg), so removing any of
    # these breaks `import numpy` itself, cascading into every downstream
    # module. Learned the hard way in the v0.0.0.dev0 Windows build.
    for mod in polynomial distutils f2py typing _pyinstaller; do
        rm -rf "$sp_dir/numpy/$mod" 2>/dev/null || true
    done
    rm -rf "$sp_dir/numpy/tests" "$sp_dir/numpy/*/tests" 2>/dev/null || true

    # ── Pillow (only used for system tray icon) ──────────────
    rm -rf "$sp_dir/PIL/tests" 2>/dev/null || true
    # Remove unused image format plugins (keep JPEG, PNG, ICO, BMP)
    for plugin in Eps Gif Tiff Webp Psd Pcx Xbm Xpm Dds Ftex Gbr Grib \
                  Icns Im Imt Iptc McIrdas Mpo Msp Pcd Pixar Ppm Sgi \
                  Spider Sun Tga Wal Wmf; do
        rm -f "$sp_dir/PIL/${plugin}ImagePlugin.py" 2>/dev/null || true
        rm -f "$sp_dir/PIL/${plugin}ImagePlugin.pyc" 2>/dev/null || true
    done

    # ── zeroconf ─────────────────────────────────────────────
    # DO NOT strip zeroconf/_services — the compiled Cython _listener.pyd
    # imports from it, and the import fails at runtime with:
    # ModuleNotFoundError: No module named 'zeroconf._services'
    # Same class of bug as numpy — "presumed unused" submodule is actually
    # imported internally by the package's own compiled code.

    # ── Strip debug symbols ──────────────────────────────────
    if command -v strip &>/dev/null; then
        echo "  Stripping debug symbols from .$ext_suffix files..."
        find "$sp_dir" -name "*.$ext_suffix" -exec strip --strip-debug {} \; 2>/dev/null || true
    fi

    # ── Remove ledgrab if pip-installed ───────────────
    rm -rf "$sp_dir"/ledgrab* "$sp_dir"/ledgrab*.dist-info 2>/dev/null || true

    local cleaned_size
    cleaned_size=$(du -sh "$sp_dir" | cut -f1)
    echo "  Site-packages after cleanup: $cleaned_size"
}

# ── Pre-compile .py → .pyc (keep sources) ────────────────────
#
# MUST run AFTER cleanup_site_packages. Speeds up first startup by
# producing .pyc alongside .py. We deliberately do NOT delete .py
# sources afterwards:
#
#   1. OpenCV's loader does literal file I/O on cv2/config.py (not
#      an import) — stripping it breaks `import cv2` with:
#      "OpenCV loader: missing configuration file: ['config.py']".
#   2. Other packages may do similar tricks (inspect.getsource,
#      runtime introspection, __file__-relative data loading).
#   3. The size saving (~30%) isn't worth the whack-a-mole of
#      shipping broken installers. We already hit this with
#      numpy.linalg and zeroconf._services — enough incidents.
#
# Args:
#   $1 — directory to compile (site-packages or app/src)
#   $2 — python executable to use (default: python3)

compile_and_strip_sources() {
    local target_dir="$1"
    local py_cmd="${2:-python3}"

    if [ ! -d "$target_dir" ]; then
        return 0
    fi

    echo "  Pre-compiling Python bytecode in $(basename "$target_dir")..."
    "$py_cmd" -m compileall -b -q "$target_dir" 2>/dev/null || {
        echo "  ERROR: compileall failed for $target_dir — aborting"
        return 1
    }
    # Drop __pycache__ to save the duplicated PEP-3147 copies; the
    # `-b` flag above placed legacy .pyc next to each .py, so nothing
    # of value is lost here.
    find "$target_dir" -type d -name __pycache__ -exec rm -rf {} + 2>/dev/null || true
}

# ── Import smoke test ────────────────────────────────────────
#
# Verifies that every top-level dependency that ledgrab actually
# uses can be imported from the stripped site-packages. Catches regressions
# where cleanup_site_packages removes a submodule that turns out to be
# imported internally by the package (e.g. numpy.linalg, zeroconf._services).
# Failing here is cheap; failing on a user's machine after install is not.
#
# Args:
#   $1 — path to site-packages to test against
#   $2 — python executable
#   $3 — (optional) extra PYTHONPATH entry (e.g. app/src for ledgrab)

smoke_test_imports() {
    local sp_dir="$1"
    local py_cmd="${2:-python3}"
    local extra_path="${3:-}"

    echo "  Running import smoke test..."
    local pypath="$sp_dir"
    if [ -n "$extra_path" ]; then
        pypath="$extra_path:$sp_dir"
    fi

    # Modules that MUST import cleanly IF PRESENT. We don't enforce
    # installation — Pillow for example is only a Windows dep. But if a
    # module's top-level package dir exists in site-packages and we
    # can't import it, that's a broken install and we abort.
    local smoke_script
    smoke_script=$(cat <<'PYEOF'
import importlib
import os
import sys

sp_dir = sys.argv[1]

# (module_name, site-packages path to check for presence)
candidates = [
    ('numpy', 'numpy'),
    ('numpy.linalg', 'numpy/linalg'),
    ('numpy.lib', 'numpy/lib'),
    ('numpy.matrixlib', 'numpy/matrixlib'),
    ('cv2', 'cv2'),
    ('fastapi', 'fastapi'),
    ('uvicorn', 'uvicorn'),
    ('starlette', 'starlette'),
    ('pydantic', 'pydantic'),
    ('zeroconf', 'zeroconf'),
    ('zeroconf._services', 'zeroconf/_services'),
    ('PIL', 'PIL'),
    ('PIL.Image', 'PIL'),
    ('yaml', 'yaml'),
]

tested = 0
skipped = 0
failed = []
for mod, path in candidates:
    if not os.path.exists(os.path.join(sp_dir, path)):
        skipped += 1
        continue
    try:
        importlib.import_module(mod)
        tested += 1
    except Exception as e:
        failed.append(f'{mod}: {type(e).__name__}: {e}')

if failed:
    print('SMOKE TEST FAILED:', file=sys.stderr)
    for f in failed:
        print(f'  {f}', file=sys.stderr)
    sys.exit(1)

print(f'  Smoke test passed ({tested} imported, {skipped} not installed)')
PYEOF
)

    if ! PYTHONPATH="$pypath" "$py_cmd" -c "$smoke_script" "$sp_dir"; then
        echo "  ERROR: smoke test failed — site-packages is broken, aborting build"
        return 1
    fi
}