Nomad changes

scripts/ci/install-system-deps/detect-tesseract-linux.sh

@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+version="$(
+  apt-cache policy tesseract-ocr 2>/dev/null |
+    grep 'Candidate:' |
+    grep -Eo '[0-9]+\.[0-9]+' |
+    head -1 ||
+    true
+)"
+
+if [[ -z "${version}" ]]; then
+  version="unknown"
+fi
+
+echo "version=${version}" >>"${GITHUB_OUTPUT}"
+echo "::notice title=Tesseract Version::Detected version: ${version}"

scripts/ci/install-system-deps/detect-tesseract-macos.sh

@@ -0,0 +1,25 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+version=""
+
+json="$(brew info --json=v2 tesseract 2>/dev/null || true)"
+if [[ -n "${json}" ]]; then
+  version="$(
+    python3 -c 'import json, re, sys; data = json.loads(sys.argv[1]); stable = (((data.get("formulae") or [{}])[0].get("versions") or {}).get("stable") or ""); m = re.match(r"^(\d+\.\d+)", stable); print(m.group(1) if m else "")' "${json}" || true
+  )"
+fi
+
+if [[ -z "${version}" ]]; then
+  first_line="$(brew info tesseract 2>/dev/null | head -1 || true)"
+  if [[ "${first_line}" =~ ([0-9]+\.[0-9]+) ]]; then
+    version="${BASH_REMATCH[1]}"
+  fi
+fi
+
+if [[ -z "${version}" ]]; then
+  version="unknown"
+fi
+
+echo "version=${version}" >>"${GITHUB_OUTPUT}"
+echo "::notice title=Tesseract Version::Detected version: ${version}"

scripts/ci/install-system-deps/install-linux.sh

@@ -0,0 +1,136 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+REPO_ROOT="${REPO_ROOT:-$(cd "$SCRIPT_DIR/../../.." && pwd)}"
+
+source "$REPO_ROOT/scripts/lib/retry.sh"
+
+echo "::group::Installing Linux dependencies"
+
+echo "Updating package index..."
+if ! retry_with_backoff sudo apt-get update; then
+  echo "::warning::apt-get update failed after retries, continuing anyway..."
+fi
+
+packages=(
+  tesseract-ocr
+  tesseract-ocr-eng
+  tesseract-ocr-tur
+  tesseract-ocr-deu
+  fonts-liberation
+  fonts-dejavu-core
+  fonts-noto-core
+  libssl-dev
+  pkg-config
+  build-essential
+  cmake
+  libmagic-dev
+  libuv1-dev
+  php-cli
+  php-dev
+)
+
+echo "Installing dependencies..."
+if retry_with_backoff_timeout 900 sudo apt-get install -y "${packages[@]}"; then
+  echo "✓ All packages installed successfully"
+else
+  exit_code=$?
+  if [ $exit_code -eq 124 ]; then
+    echo "::error::Package installation timed out after 15 minutes"
+  else
+    echo "::warning::Some packages failed to install, attempting individual installs..."
+    for pkg in tesseract-ocr libssl-dev pkg-config cmake; do
+      echo "Installing $pkg..."
+      if retry_with_backoff_timeout 300 sudo apt-get install -y "$pkg" 2>&1; then
+        echo "  ✓ $pkg installed"
+      else
+        echo "  ⚠ Failed to install $pkg"
+      fi
+    done
+  fi
+fi
+
+echo "::endgroup::"
+
+echo "::group::Verifying Linux installations"
+
+echo "CMake:"
+if command -v cmake >/dev/null 2>&1; then
+  cmake --version | head -1
+  echo "✓ CMake available"
+  # Export CMAKE environment variable for immediate availability in build scripts
+  CMAKE_FULL_PATH="$(command -v cmake)"
+  if [[ -n "$GITHUB_ENV" ]]; then
+    echo "CMAKE=$CMAKE_FULL_PATH" >>"$GITHUB_ENV"
+    echo "✓ Set CMAKE=$CMAKE_FULL_PATH in GITHUB_ENV"
+  fi
+  # Also add cmake binary directory to GITHUB_PATH for subsequent steps
+  CMAKE_BIN="$(dirname "$CMAKE_FULL_PATH")"
+  if [[ -n "$GITHUB_PATH" && -d "$CMAKE_BIN" ]]; then
+    echo "$CMAKE_BIN" >>"$GITHUB_PATH"
+    echo "✓ Added cmake directory to GITHUB_PATH: $CMAKE_BIN"
+  fi
+else
+  echo "::error::CMake not found after installation"
+  exit 1
+fi
+
+echo ""
+echo "Tesseract:"
+if command -v tesseract >/dev/null 2>&1; then
+  if tesseract --version 2>/dev/null | head -1; then
+    echo "✓ Tesseract CLI available"
+  else
+    echo "::warning::Tesseract CLI present but failed to run"
+  fi
+else
+  echo "::warning::Tesseract CLI not found; continuing (OCR will rely on bundled Tesseract)"
+fi
+
+echo ""
+echo "Available Tesseract languages:"
+if command -v tesseract >/dev/null 2>&1; then
+  tesseract --list-langs | head -10 || true
+else
+  echo "(tesseract CLI not available)"
+fi
+
+echo ""
+echo "PHP:"
+if command -v php >/dev/null 2>&1; then
+  php --version | head -1
+  echo "✓ PHP available"
+else
+  echo "::error::PHP not found after installation"
+  exit 1
+fi
+
+echo ""
+echo "Checking Tesseract data path..."
+
+tessdata_found=0
+for tessdata_path in "/usr/share/tesseract-ocr/5/tessdata" "/usr/share/tesseract-ocr/tessdata"; do
+  if [ -d "$tessdata_path" ]; then
+    echo "Found tessdata at: $tessdata_path"
+
+    echo "Required language files:"
+    for lang in eng tur deu; do
+      if [ -f "$tessdata_path/${lang}.traineddata" ]; then
+        size=$(stat -c%s "$tessdata_path/${lang}.traineddata" 2>/dev/null || echo "unknown")
+        echo "  ✓ ${lang}.traineddata ($size bytes)"
+      else
+        echo "  ⚠ ${lang}.traineddata (missing)"
+      fi
+    done
+    tessdata_found=1
+    break
+  fi
+done
+
+if [ $tessdata_found -eq 0 ]; then
+  echo "::error::Tessdata directory not found in standard locations"
+  exit 1
+fi
+
+echo "::endgroup::"

scripts/ci/install-system-deps/install-macos.sh

@@ -0,0 +1,136 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+REPO_ROOT="${REPO_ROOT:-$(cd "$SCRIPT_DIR/../../.." && pwd)}"
+
+source "$REPO_ROOT/scripts/lib/retry.sh"
+
+echo "::group::Installing macOS dependencies"
+
+if [[ -d "/opt/homebrew/bin" ]]; then
+  export PATH="/opt/homebrew/bin:/opt/homebrew/sbin:${PATH}"
+  echo "/opt/homebrew/bin" >>"$GITHUB_PATH"
+  echo "/opt/homebrew/sbin" >>"$GITHUB_PATH"
+fi
+if [[ -d "/usr/local/bin" ]]; then
+  export PATH="/usr/local/bin:/usr/local/sbin:${PATH}"
+  echo "/usr/local/bin" >>"$GITHUB_PATH"
+  echo "/usr/local/sbin" >>"$GITHUB_PATH"
+fi
+
+if ! brew list cmake &>/dev/null; then
+  echo "Installing CMake..."
+  retry_with_backoff brew install cmake || {
+    echo "::error::Failed to install CMake after retries"
+    exit 1
+  }
+else
+  echo "✓ CMake already installed"
+fi
+
+if ! command -v cmake >/dev/null 2>&1; then
+  echo "CMake not on PATH after install; attempting brew link..."
+  brew link --overwrite cmake >/dev/null 2>&1 || true
+fi
+
+if ! brew list tesseract &>/dev/null; then
+  echo "Installing Tesseract..."
+  retry_with_backoff brew install tesseract || {
+    echo "::error::Failed to install Tesseract after retries"
+    exit 1
+  }
+else
+  echo "✓ Tesseract already installed"
+fi
+
+if ! command -v tesseract >/dev/null 2>&1; then
+  echo "Tesseract not on PATH after install; attempting brew link..."
+  brew link --overwrite tesseract >/dev/null 2>&1 || true
+fi
+
+if ! brew list tesseract-lang &>/dev/null; then
+  echo "Installing Tesseract language packs..."
+  retry_with_backoff brew install tesseract-lang || {
+    echo "::warning::Failed to install tesseract-lang, some languages may be unavailable"
+  }
+else
+  echo "✓ Tesseract language packs already installed"
+fi
+
+if ! brew list libmagic &>/dev/null; then
+  echo "Installing libmagic..."
+  retry_with_backoff brew install libmagic || {
+    echo "::warning::Failed to install libmagic after retries"
+  }
+else
+  echo "✓ libmagic already installed"
+fi
+
+if ! brew list php &>/dev/null; then
+  echo "Installing PHP..."
+  retry_with_backoff brew install php || {
+    echo "::error::Failed to install PHP after retries"
+    exit 1
+  }
+else
+  echo "✓ PHP already installed"
+fi
+
+if ! command -v php >/dev/null 2>&1; then
+  echo "PHP not on PATH after install; attempting brew link..."
+  brew link --overwrite php >/dev/null 2>&1 || true
+fi
+
+echo "::endgroup::"
+
+echo "::group::Verifying macOS installations"
+
+echo "CMake:"
+if command -v cmake >/dev/null 2>&1; then
+  cmake --version | head -1
+  # Export CMAKE environment variable for immediate availability in build scripts
+  CMAKE_FULL_PATH="$(command -v cmake)"
+  if [[ -n "$GITHUB_ENV" ]]; then
+    echo "CMAKE=$CMAKE_FULL_PATH" >>"$GITHUB_ENV"
+    echo "✓ Set CMAKE=$CMAKE_FULL_PATH in GITHUB_ENV"
+  fi
+  # Also add cmake binary directory to GITHUB_PATH for subsequent steps
+  CMAKE_BIN="$(dirname "$CMAKE_FULL_PATH")"
+  if [[ -n "$GITHUB_PATH" && -d "$CMAKE_BIN" ]]; then
+    echo "$CMAKE_BIN" >>"$GITHUB_PATH"
+    echo "✓ Added cmake directory to GITHUB_PATH: $CMAKE_BIN"
+  fi
+else
+  echo "::error::CMake not found on PATH after installation"
+  echo "PATH=$PATH"
+  brew --prefix cmake 2>/dev/null || true
+  exit 1
+fi
+
+echo ""
+echo "Tesseract:"
+if command -v tesseract >/dev/null 2>&1; then
+  tesseract --version | head -1
+else
+  echo "::error::Tesseract not found on PATH after installation"
+  echo "PATH=$PATH"
+  brew --prefix tesseract 2>/dev/null || true
+  exit 1
+fi
+
+echo ""
+echo "Available languages:"
+tesseract --list-langs | head -5
+
+echo ""
+echo "PHP:"
+if command -v php >/dev/null 2>&1; then
+  php --version | head -1
+else
+  echo "::error::PHP not found on PATH after installation"
+  echo "PATH=$PATH"
+  exit 1
+fi
+
+echo "::endgroup::"

scripts/ci/install-system-deps/install-windows.ps1

@@ -0,0 +1,301 @@
+#!/usr/bin/env pwsh
+
+Set-StrictMode -Version Latest
+$ErrorActionPreference = 'Stop'
+
+Write-Host "::group::Installing Windows dependencies"
+
+function Retry-Command {
+  param(
+    [scriptblock]$Command,
+    [int]$MaxAttempts = 3,
+    [int]$DelaySeconds = 5
+  )
+
+  $attempt = 1
+  while ($attempt -le $MaxAttempts) {
+    try {
+      Write-Host "Attempt $attempt of $MaxAttempts..."
+      & $Command
+      return $true
+    }
+    catch {
+      $attempt++
+      if ($attempt -le $MaxAttempts) {
+        $backoffDelay = $DelaySeconds * [Math]::Pow(2, $attempt - 1)
+        Write-Host "⚠ Attempt failed, retrying in ${backoffDelay}s..." -ForegroundColor Yellow
+        Start-Sleep -Seconds $backoffDelay
+      }
+      else {
+        return $false
+      }
+    }
+  }
+}
+
+$tesseractCacheHit = $env:TESSERACT_CACHE_HIT -eq "true"
+$llvmCacheHit = $env:LLVM_CACHE_HIT -eq "true"
+$cmakeCacheHit = $env:CMAKE_CACHE_HIT -eq "true"
+$cmakeInstalled = $false
+
+Write-Host "Cache status:"
+Write-Host "  TESSERACT_CACHE_HIT: $env:TESSERACT_CACHE_HIT (evaluated: $tesseractCacheHit)"
+Write-Host "  LLVM_CACHE_HIT: $env:LLVM_CACHE_HIT (evaluated: $llvmCacheHit)"
+Write-Host "  CMAKE_CACHE_HIT: $env:CMAKE_CACHE_HIT (evaluated: $cmakeCacheHit)"
+Write-Host ""
+try {
+  & cmake --version 2>$null
+  Write-Host "✓ CMake already installed"
+  $cmakeInstalled = $true
+}
+catch {
+  Write-Host "CMake not found, will attempt to install"
+}
+
+if (-not $tesseractCacheHit) {
+  Write-Host "Tesseract cache miss, installing (optional for build - needed for tests only)..."
+  if (-not (Retry-Command { choco install -y tesseract --no-progress } -MaxAttempts 3)) {
+    Write-Host "::warning::Failed to install Tesseract (optional dependency - gem build does not require it)"
+  }
+  else {
+    Write-Host "✓ Tesseract installed"
+    # Ensure tessdata directory exists and is accessible
+    $tesseractPath = "C:\Program Files\Tesseract-OCR"
+    if (Test-Path $tesseractPath) {
+      Write-Host "  Configuring Tesseract data paths..."
+
+      # Create tessdata directory if it doesn't exist
+      $tessdataPath = "$tesseractPath\tessdata"
+      if (-not (Test-Path $tessdataPath)) {
+        Write-Host "  Creating tessdata directory at: $tessdataPath"
+        New-Item -ItemType Directory -Path $tessdataPath -Force | Out-Null
+      }
+
+      # Download English language data if not present
+      if (-not (Test-Path "$tessdataPath\eng.traineddata")) {
+        Write-Host "  Downloading English language data..."
+        try {
+          $engUrl = "https://github.com/tesseract-ocr/tessdata_fast/raw/main/eng.traineddata"
+          Invoke-WebRequest -Uri $engUrl -OutFile "$tessdataPath\eng.traineddata" -ErrorAction Stop
+          Write-Host "  ✓ Downloaded eng.traineddata"
+        }
+        catch {
+          Write-Host "  ::warning::Failed to download eng.traineddata: $($_.Exception.Message)"
+        }
+      }
+
+      # Download OSD data if not present (needed for orientation detection)
+      if (-not (Test-Path "$tessdataPath\osd.traineddata")) {
+        Write-Host "  Downloading OSD data..."
+        try {
+          $osdUrl = "https://github.com/tesseract-ocr/tessdata_fast/raw/main/osd.traineddata"
+          Invoke-WebRequest -Uri $osdUrl -OutFile "$tessdataPath\osd.traineddata" -ErrorAction Stop
+          Write-Host "  ✓ Downloaded osd.traineddata"
+        }
+        catch {
+          Write-Host "  ::warning::Failed to download osd.traineddata: $($_.Exception.Message)"
+        }
+      }
+    }
+  }
+}
+else {
+  Write-Host "✓ Tesseract found in cache"
+}
+
+if (-not $llvmCacheHit) {
+  Write-Host "LLVM cache miss, installing LLVM/Clang (required for bindgen)..."
+  if (-not (Retry-Command { choco install -y llvm --no-progress } -MaxAttempts 3)) {
+    Write-Host "::warning::Failed to install LLVM/Clang via Chocolatey"
+  }
+  else {
+    Write-Host "✓ LLVM/Clang installed"
+  }
+}
+else {
+  Write-Host "✓ LLVM/Clang found in cache"
+}
+
+Write-Host "Installing PHP..."
+$phpInstalled = $false
+try {
+  & php --version 2>$null
+  Write-Host "✓ PHP already installed"
+  $phpInstalled = $true
+}
+catch {
+  Write-Host "PHP not found, installing via Chocolatey..."
+  if (-not (Retry-Command { choco install -y php --no-progress } -MaxAttempts 3)) {
+    Write-Host "::warning::Failed to install PHP via Chocolatey, will rely on shivammathur/setup-php action"
+  }
+  else {
+    Write-Host "✓ PHP installed via Chocolatey"
+    $phpInstalled = $true
+  }
+}
+
+Write-Host "Installing CMake..."
+if (-not $cmakeCacheHit) {
+  Write-Host "CMake cache miss, installing..."
+  if (-not (Retry-Command { choco install -y cmake --no-progress } -MaxAttempts 3)) {
+    throw "Failed to install CMake after 3 attempts"
+  }
+  Write-Host "✓ CMake installed"
+}
+else {
+  Write-Host "✓ CMake found in cache"
+}
+
+Write-Host "Configuring PATH and environment variables..."
+$paths = @(
+  "C:\Program Files\CMake\bin",
+  "C:\Program Files\Tesseract-OCR",
+  "C:\Program Files\LLVM\bin",
+  "C:\tools\php",
+  "C:\Program Files\PHP"
+)
+
+foreach ($path in $paths) {
+  if (Test-Path $path) {
+    Write-Host "  Adding to PATH: $path"
+    Write-Output $path | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
+    $env:PATH = "$path;$env:PATH"
+  }
+  else {
+    Write-Host "  Path not found (skipping): $path"
+  }
+}
+
+# Ensure TESSDATA_PREFIX is set for Windows OCR tests
+$tesseractPath = "C:\Program Files\Tesseract-OCR"
+if (Test-Path $tesseractPath) {
+  $tessdataPath = "$tesseractPath\tessdata"
+  if (Test-Path $tessdataPath) {
+    Write-Host "  Setting TESSDATA_PREFIX for tests: $tessdataPath"
+    Add-Content -Path $env:GITHUB_ENV -Value "TESSDATA_PREFIX=$tessdataPath"
+    $env:TESSDATA_PREFIX = $tessdataPath
+  }
+}
+
+Write-Host "::endgroup::"
+
+Write-Host "::group::Verifying Windows installations"
+
+Write-Host "Tesseract (optional for build):"
+try {
+  $tesseractCmd = Get-Command tesseract -ErrorAction Stop
+  $tesseractPath = $tesseractCmd.Path
+  Write-Host "  Found at: $tesseractPath"
+  Write-Host "  Command type: $($tesseractCmd.CommandType)"
+
+  # Get installation directory
+  $tesseractDir = Split-Path -Parent $tesseractPath
+  Write-Host "  Installation directory: $tesseractDir"
+
+  # Check for tessdata
+  $tessdataPath = Join-Path $tesseractDir "tessdata"
+  if (Test-Path $tessdataPath) {
+    Write-Host "  tessdata directory: $tessdataPath"
+    Write-Host "  Available language files:"
+    Get-ChildItem "$tessdataPath\*.traineddata" -ErrorAction SilentlyContinue | ForEach-Object {
+      Write-Host "    - $($_.Name)"
+    }
+  }
+  else {
+    Write-Host "  tessdata directory not found at: $tessdataPath"
+  }
+
+  try {
+    $version = & tesseract --version 2>&1
+    Write-Host "  Version output: $version"
+    Write-Host "✓ Tesseract available and working"
+
+    Write-Host ""
+    Write-Host "Available Tesseract languages:"
+    & tesseract --list-langs 2>&1 | ForEach-Object { Write-Host "  $_" }
+  }
+  catch {
+    Write-Host "⚠ Warning: Tesseract found but failed to run: $($_.Exception.Message)"
+  }
+
+  # Set TESSDATA_PREFIX environment variable for tests
+  if (Test-Path $tessdataPath) {
+    Write-Host ""
+    Write-Host "Setting TESSDATA_PREFIX environment variable..."
+    Add-Content -Path $env:GITHUB_ENV -Value "TESSDATA_PREFIX=$tessdataPath"
+    Write-Host "✓ Set TESSDATA_PREFIX=$tessdataPath in GITHUB_ENV"
+    $env:TESSDATA_PREFIX = $tessdataPath
+  }
+}
+catch {
+  Write-Host "⚠ Tesseract not found on PATH (not required for build)"
+  Write-Host "  Error details: $($_.Exception.Message)"
+  Write-Host "  Searching common installation locations..."
+
+  $commonPaths = @(
+    "C:\Program Files\Tesseract-OCR\tesseract.exe",
+    "C:\Program Files (x86)\Tesseract-OCR\tesseract.exe",
+    "${env:ProgramFiles}\Tesseract-OCR\tesseract.exe",
+    "${env:ProgramFiles(x86)}\Tesseract-OCR\tesseract.exe"
+  )
+
+  $found = $false
+  foreach ($path in $commonPaths) {
+    if (Test-Path $path) {
+      Write-Host "  Found Tesseract at: $path (not on PATH)"
+      $tesseractDir = Split-Path -Parent $path
+      $tessdataPath = Join-Path $tesseractDir "tessdata"
+      if (Test-Path $tessdataPath) {
+        Write-Host "  Found tessdata at: $tessdataPath"
+        Add-Content -Path $env:GITHUB_ENV -Value "TESSDATA_PREFIX=$tessdataPath"
+        Write-Host "✓ Set TESSDATA_PREFIX=$tessdataPath in GITHUB_ENV"
+        $env:TESSDATA_PREFIX = $tessdataPath
+      }
+      $found = $true
+      break
+    }
+  }
+
+  if (-not $found) {
+    Write-Host "  Tesseract not found in common locations"
+  }
+}
+
+Write-Host ""
+Write-Host "CMake:"
+try {
+  & cmake --version
+  Write-Host "✓ CMake available"
+  # Export CMAKE environment variable for immediate availability in build scripts
+  $cmakePath = (Get-Command cmake -ErrorAction Stop).Source
+  if ($cmakePath) {
+    Add-Content -Path $env:GITHUB_ENV -Value "CMAKE=$cmakePath"
+    Write-Host "✓ Set CMAKE=$cmakePath in GITHUB_ENV"
+  }
+}
+catch {
+  Write-Host "::error::CMake not found after installation"
+  throw "CMake verification failed"
+}
+
+Write-Host ""
+Write-Host "Clang:"
+try {
+  & clang --version
+  Write-Host "✓ Clang available"
+}
+catch {
+  Write-Host "⚠ Warning: Clang not currently available on PATH"
+}
+
+Write-Host ""
+Write-Host "PHP:"
+try {
+  & php --version
+  Write-Host "✓ PHP available"
+}
+catch {
+  Write-Host "⚠ Warning: PHP not currently available on PATH (will be set up by shivammathur/setup-php action)"
+}
+
+Write-Host "::endgroup::"