pdf-oxide 0.3.24

package/src/index.ts

@@ -0,0 +1,453 @@
+// PDF Oxide Node.js bindings - Native module loader
+
+import { platform, arch } from 'node:process';
+import { createRequire } from 'node:module';
+import { fileURLToPath } from 'node:url';
+import { dirname } from 'node:path';
+import {
+  PdfException,
+  ParseException,
+  IoException,
+  EncryptionException,
+  UnsupportedFeatureException,
+  InvalidStateException,
+  ValidationException,
+  RenderingException,
+  SearchException,
+  ComplianceException,
+  OcrException,
+  SignatureException,
+  CertificateLoadFailed,
+  SigningFailed,
+  RedactionException,
+  AccessibilityException,
+  OptimizationException,
+  UnknownError,
+  ErrorCategory,
+  ErrorSeverity,
+  wrapError,
+  wrapMethod,
+  wrapAsyncMethod,
+  mapFfiErrorCode,
+} from './errors';
+import {
+  addPdfDocumentProperties,
+  addPdfProperties,
+  addPdfPageProperties,
+} from './properties';
+import {
+  PdfBuilder,
+  ConversionOptionsBuilder,
+  MetadataBuilder,
+  AnnotationBuilder,
+  SearchOptionsBuilder,
+} from './builders/index';
+import {
+  OutlineManager,
+  MetadataManager,
+  ExtractionManager,
+  SearchManager,
+  SecurityManager,
+  AnnotationManager,
+  LayerManager,
+  RenderingManager,
+  SearchStream,
+  ExtractionStream,
+  MetadataStream,
+  createSearchStream,
+  createExtractionStream,
+  createMetadataStream,
+  BatchManager,
+  type BatchDocument,
+  type BatchProgress,
+  type BatchResult,
+  type BatchOptions,
+  type BatchStatistics,
+} from './managers/index';
+import { WorkerPool, workerPool } from './workers/index';
+import type { WorkerTask, WorkerResult } from './workers/index';
+
+// Create require function for CommonJS modules
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = dirname(__filename);
+const require = createRequire(import.meta.url);
+
+// Phase 4+ managers (compiled JavaScript - use require for dynamic import)
+// Phase 9: Now imports from canonical consolidated managers in managers/
+const {
+  OcrManager,
+  OcrManager: OCRManager,
+  OcrDetectionMode: OCRDetectionMode,
+  ComplianceManager,
+  PdfALevel,
+  PdfXLevel,
+  PdfUALevel,
+  ComplianceIssueType,
+  IssueSeverity,
+  SignatureManager,
+  SignatureAlgorithm,
+  DigestAlgorithm,
+  BarcodeManager,
+  BarcodeFormat,
+  BarcodeErrorCorrection,
+  FormFieldManager,
+  FormFieldType,
+  FieldVisibility,
+  ResultAccessorsManager,
+  SearchResultProperties,
+  FontProperties,
+  ImageProperties,
+  AnnotationProperties,
+  ThumbnailManager,
+  ThumbnailSize,
+  ImageFormat,
+  HybridMLManager,
+  PageComplexity,
+  ContentType,
+  XfaManager,
+  XfaFormType,
+  XfaFieldType,
+  CacheManager,
+  EditingManager,
+  AccessibilityManager,
+  OptimizationManager,
+  EnterpriseManager,
+} = require('../lib/managers/index.js') as any;
+// OcrLanguage re-exported from canonical OcrManager
+const { OcrLanguage: OCRLanguage } = require('../lib/managers/ocr-manager.js') as any;
+
+/**
+ * Platform-specific native module packages
+ */
+const PLATFORMS: Record<string, Record<string, string>> = {
+  'darwin': {
+    'x64': 'pdf_oxide-darwin-x64',
+    'arm64': 'pdf_oxide-darwin-arm64',
+  },
+  'linux': {
+    'x64': 'pdf_oxide-linux-x64-gnu',
+    'arm64': 'pdf_oxide-linux-arm64-gnu',
+  },
+  'win32': {
+    'x64': 'pdf_oxide-win32-x64-msvc',
+    'arm64': 'pdf_oxide-win32-arm64-msvc',
+  },
+};
+
+/**
+ * Gets the native package name for the current platform and architecture
+ * @returns Native package name
+ * @throws Error if platform or architecture is not supported
+ */
+function getNativePackageName(): string {
+  const osPackages = PLATFORMS[platform];
+  if (!osPackages) {
+    throw new Error(`Unsupported platform: ${platform}. Supported platforms: ${Object.keys(PLATFORMS).join(', ')}`);
+  }
+
+  const pkg = osPackages[arch];
+  if (!pkg) {
+    throw new Error(`Unsupported architecture: ${arch} for ${platform}. Supported architectures: ${Object.keys(osPackages).join(', ')}`);
+  }
+
+  return pkg;
+}
+
+let nativeModule: any;
+
+/**
+ * Loads the native module dynamically based on platform and architecture
+ * @returns Native module
+ * @throws Error if native module cannot be loaded
+ */
+function loadNativeModule(): any {
+  if (nativeModule) {
+    return nativeModule;
+  }
+
+  try {
+    // Try loading from platform-specific package first (CommonJS)
+    const packageName = getNativePackageName();
+    try {
+      // Use require to load the native module
+      nativeModule = require(packageName);
+    } catch (e) {
+      // Fallback to local binary if in development
+      if (process.env.NODE_ENV === 'development' || process.env.NAPI_DEV) {
+        try {
+          nativeModule = require('./pdf-oxide');
+        } catch {
+          throw e;
+        }
+      } else {
+        throw e;
+      }
+    }
+    return nativeModule;
+  } catch (error) {
+    throw new Error(`Failed to load native module: ${(error as Error).message}`);
+  }
+}
+
+// Load native module
+const native = loadNativeModule();
+
+/**
+ * Wraps native class methods to convert errors to proper JavaScript Error subclasses.
+ * This ensures that errors thrown from native code are instanceof the appropriate Error class.
+ * @param nativeClass - The native class to wrap
+ * @param asyncMethods - Names of async methods to wrap specially
+ * @returns Wrapped class with error-handling methods
+ */
+function wrapNativeClass(nativeClass: any, asyncMethods: string[] = []): any {
+  if (!nativeClass) return nativeClass;
+
+  // For static methods like PdfDocument.open()
+  for (const key of Object.getOwnPropertyNames(nativeClass)) {
+    if (key !== 'prototype' && key !== 'length' && key !== 'name' && typeof nativeClass[key] === 'function') {
+      const isAsync = asyncMethods.includes(key);
+      if (isAsync) {
+        nativeClass[key] = wrapAsyncMethod(nativeClass[key], nativeClass);
+      } else {
+        nativeClass[key] = wrapMethod(nativeClass[key], nativeClass);
+      }
+    }
+  }
+
+  // For instance methods, wrap the prototype
+  if (nativeClass.prototype) {
+    for (const key of Object.getOwnPropertyNames(nativeClass.prototype)) {
+      if (key !== 'constructor' && typeof nativeClass.prototype[key] === 'function') {
+        const isAsync = asyncMethods.includes(key);
+        const descriptor = Object.getOwnPropertyDescriptor(nativeClass.prototype, key);
+        if (descriptor && descriptor.writable) {
+          if (isAsync) {
+            nativeClass.prototype[key] = wrapAsyncMethod(nativeClass.prototype[key]);
+          } else {
+            nativeClass.prototype[key] = wrapMethod(nativeClass.prototype[key]);
+          }
+        }
+      }
+    }
+  }
+
+  return nativeClass;
+}
+
+// List of async methods for each class (Phase 2.4 implementation)
+const asyncMethodsByClass: Record<string, string[]> = {
+  PdfDocument: [
+    'extract_text_async',
+    'to_markdown_async',
+  ],
+  Pdf: [
+    'save_async',
+  ],
+  PdfBuilder: [],
+  PdfPage: [],
+  PdfElement: [],
+  PdfText: [],
+  PdfImage: [],
+  PdfPath: [],
+  PdfTable: [],
+  PdfStructure: [],
+  Annotation: [],
+  TextAnnotation: [],
+  HighlightAnnotation: [],
+  LinkAnnotation: [],
+  TextSearcher: [],
+};
+
+// Wrap native classes with error handling and property getters
+const wrappedClasses: Record<string, any> = {};
+for (const className of Object.keys(asyncMethodsByClass)) {
+  if (native[className]) {
+    wrappedClasses[className] = wrapNativeClass(native[className], asyncMethodsByClass[className]);
+  }
+}
+
+// Add property getters to enhance idiomatic JavaScript API
+if (wrappedClasses.PdfDocument) {
+  addPdfDocumentProperties(wrappedClasses.PdfDocument);
+}
+if (wrappedClasses.Pdf) {
+  addPdfProperties(wrappedClasses.Pdf);
+}
+if (wrappedClasses.PdfPage) {
+  addPdfPageProperties(wrappedClasses.PdfPage);
+}
+
+// Export as ES module
+const getVersion = native.getVersion;
+const getPdfOxideVersion = native.getPdfOxideVersion;
+const PdfDocument = wrappedClasses.PdfDocument || native.PdfDocument;
+const Pdf = wrappedClasses.Pdf || native.Pdf;
+// PdfBuilder is imported from ./builders/index - don't redeclare
+const PdfPage = wrappedClasses.PdfPage || native.PdfPage;
+const PdfElement = wrappedClasses.PdfElement || native.PdfElement;
+const PdfText = wrappedClasses.PdfText || native.PdfText;
+const PdfImage = wrappedClasses.PdfImage || native.PdfImage;
+const PdfPath = wrappedClasses.PdfPath || native.PdfPath;
+const PdfTable = wrappedClasses.PdfTable || native.PdfTable;
+const PdfStructure = wrappedClasses.PdfStructure || native.PdfStructure;
+const Annotation = wrappedClasses.Annotation || native.Annotation;
+const TextAnnotation = wrappedClasses.TextAnnotation || native.TextAnnotation;
+const HighlightAnnotation = wrappedClasses.HighlightAnnotation || native.HighlightAnnotation;
+const LinkAnnotation = wrappedClasses.LinkAnnotation || native.LinkAnnotation;
+const PdfError = PdfException;
+const PageSize = native.PageSize;
+const Rect = native.Rect;
+const Point = native.Point;
+const Color = native.Color;
+const ConversionOptions = native.ConversionOptions;
+const SearchOptions = native.SearchOptions;
+const SearchResult = native.SearchResult;
+const TextSearcher = wrappedClasses.TextSearcher || native.TextSearcher;
+
+export {
+  // Version info
+  getVersion,
+  getPdfOxideVersion,
+
+  // Main classes
+  PdfDocument,
+  Pdf,
+  PdfPage,
+
+  // Element types
+  PdfElement,
+  PdfText,
+  PdfImage,
+  PdfPath,
+  PdfTable,
+  PdfStructure,
+
+  // Annotation types
+  Annotation,
+  TextAnnotation,
+  HighlightAnnotation,
+  LinkAnnotation,
+
+  // Error types
+  PdfError,
+  PdfException,
+  ParseException,
+  IoException,
+  EncryptionException,
+  UnsupportedFeatureException,
+  InvalidStateException,
+  ValidationException,
+  RenderingException,
+  SearchException,
+  ComplianceException,
+  OcrException,
+  SignatureException,
+  CertificateLoadFailed,
+  SigningFailed,
+  RedactionException,
+  AccessibilityException,
+  OptimizationException,
+  UnknownError,
+
+  // Types
+  PageSize,
+  Rect,
+  Point,
+  Color,
+  ConversionOptions,
+  SearchOptions,
+  SearchResult,
+
+  // Utilities
+  TextSearcher,
+
+  // Error utilities
+  ErrorCategory,
+  ErrorSeverity,
+  wrapError,
+  wrapMethod,
+  wrapAsyncMethod,
+  mapFfiErrorCode,
+
+  // Builders
+  PdfBuilder,
+  ConversionOptionsBuilder,
+  MetadataBuilder,
+  AnnotationBuilder,
+  SearchOptionsBuilder,
+
+  // Managers (Phase 1-3: Core)
+  OutlineManager,
+  MetadataManager,
+  ExtractionManager,
+  SearchManager,
+  SecurityManager,
+  AnnotationManager,
+  LayerManager,
+  RenderingManager,
+
+  // Managers (Phase 4+, consolidated in Phase 9)
+  OcrManager,
+  OCRManager,
+  OCRLanguage,
+  OCRDetectionMode,
+  ComplianceManager,
+  PdfALevel,
+  PdfXLevel,
+  PdfUALevel,
+  ComplianceIssueType,
+  IssueSeverity,
+  SignatureManager,
+  SignatureAlgorithm,
+  DigestAlgorithm,
+  BarcodeManager,
+  BarcodeFormat,
+  BarcodeErrorCorrection,
+  FormFieldManager,
+  FormFieldType,
+  FieldVisibility,
+  ResultAccessorsManager,
+  SearchResultProperties,
+  FontProperties,
+  ImageProperties,
+  AnnotationProperties,
+  ThumbnailManager,
+  ThumbnailSize,
+  ImageFormat,
+  HybridMLManager,
+  PageComplexity,
+  ContentType,
+  XfaManager,
+  XfaFormType,
+  XfaFieldType,
+  CacheManager,
+  EditingManager,
+  AccessibilityManager,
+  OptimizationManager,
+  EnterpriseManager,
+
+  // Phase 2.4: Stream API
+  SearchStream,
+  ExtractionStream,
+  MetadataStream,
+  createSearchStream,
+  createExtractionStream,
+  createMetadataStream,
+
+  // Phase 2.5: Batch Processing API
+  BatchManager,
+
+  // Worker Threads API
+  WorkerPool,
+  workerPool,
+};
+
+export type {
+  WorkerTask,
+  WorkerResult,
+  BatchDocument,
+  BatchProgress,
+  BatchResult,
+  BatchOptions,
+  BatchStatistics,
+};