@kreuzberg/node 4.0.0-rc.8 → 4.0.0

package/dist/index.d.mts

@@ -1,7 +1,7 @@
 import { PanicContext } from './errors.mjs';
 export { CacheError, ErrorCode, ImageProcessingError, KreuzbergError, MissingDependencyError, OcrError, ParsingError, PluginError, ValidationError } from './errors.mjs';
-import { ExtractionConfig as ExtractionConfig$1, ExtractionResult, PostProcessorProtocol, ValidatorProtocol, OcrBackendProtocol } from './types.mjs';
-export { ArchiveMetadata, Chunk, ChunkMetadata, ChunkingConfig, EmailMetadata, ErrorMetadata, ExcelMetadata, ExtractedImage, HtmlConversionOptions, HtmlMetadata, HtmlPreprocessingOptions, ImageExtractionConfig, ImageMetadata, ImagePreprocessingMetadata, KeywordAlgorithm, KeywordConfig, LanguageDetectionConfig, Metadata, OcrConfig, OcrMetadata, PageBoundary, PageConfig, PageContent, PageInfo, PageStructure, PageUnitType, PdfConfig, PdfMetadata, PostProcessorConfig, PptxMetadata, ProcessingStage, RakeParams, Table, TesseractConfig, TextMetadata, TokenReductionConfig, XmlMetadata, YakeParams } from './types.mjs';
+import { ExtractionConfig as ExtractionConfig$1, ExtractionResult, PostProcessorProtocol, ValidatorProtocol, OcrBackendProtocol, ErrorClassification, WorkerPool, WorkerPoolStats } from './types.mjs';
+export { ArchiveMetadata, Chunk, ChunkMetadata, ChunkingConfig, EmailMetadata, ErrorMetadata, ExcelMetadata, ExtractedImage, ExtractedKeyword, HeaderMetadata, HierarchyConfig, HtmlConversionOptions, HtmlImageMetadata, HtmlMetadata, HtmlPreprocessingOptions, ImageExtractionConfig, ImageMetadata, ImagePreprocessingMetadata, KeywordAlgorithm, KeywordConfig, LanguageDetectionConfig, LinkMetadata, Metadata, OcrConfig, OcrMetadata, PageBoundary, PageContent, PageExtractionConfig, PageInfo, PageStructure, PageUnitType, PdfConfig, PdfMetadata, PostProcessorConfig, PptxMetadata, ProcessingStage, RakeParams, StructuredData, Table, TesseractConfig, TextMetadata, TokenReductionConfig, XmlMetadata, YakeParams } from './types.mjs';
 export { GutenOcrBackend } from './ocr/guten-ocr.mjs';
 
 /**
@@ -65,10 +65,15 @@ declare function __resetBindingForTests(): void;
  * **Usage Note**: For processing multiple files, prefer `batchExtractFilesSync()` which
  * provides better performance and memory management.
  *
- * @param filePath - Path to the file (string)
- * @param mimeType - Optional MIME type hint (auto-detected if null)
- * @param config - Extraction configuration (uses defaults if null)
- * @returns ExtractionResult with content, metadata, and tables
+ * @param filePath - Path to the file to extract (string). Can be absolute or relative.
+ * @param mimeType - Optional MIME type hint for format detection. If null, MIME type is auto-detected from file extension or content.
+ * @param config - Extraction configuration object. If null, uses default extraction settings.
+ * @returns ExtractionResult containing extracted content, metadata, tables, and optional chunks/images
+ * @throws {Error} If file doesn't exist, cannot be accessed, or cannot be read
+ * @throws {ParsingError} When document format is invalid or corrupted
+ * @throws {OcrError} When OCR processing fails (if OCR is enabled)
+ * @throws {ValidationError} When extraction result fails validation (if validators registered)
+ * @throws {KreuzbergError} For other extraction-related failures
  *
  * @example
  * ```typescript
@@ -92,17 +97,22 @@ declare function __resetBindingForTests(): void;
  * const result2 = extractFileSync('scanned.pdf', null, config);
  * ```
  */
-declare function extractFileSync(filePath: string, mimeType?: string | null, config?: ExtractionConfig$1 | null): ExtractionResult;
+declare function extractFileSync(filePath: string, mimeTypeOrConfig?: string | null | ExtractionConfig$1, maybeConfig?: ExtractionConfig$1 | null): ExtractionResult;
 /**
  * Extract content from a single file (asynchronous).
  *
  * **Usage Note**: For processing multiple files, prefer `batchExtractFiles()` which
  * provides better performance and memory management.
  *
- * @param filePath - Path to the file (string)
- * @param mimeType - Optional MIME type hint (auto-detected if null)
- * @param config - Extraction configuration (uses defaults if null)
- * @returns Promise<ExtractionResult> with content, metadata, and tables
+ * @param filePath - Path to the file to extract (string). Can be absolute or relative.
+ * @param mimeType - Optional MIME type hint for format detection. If null, MIME type is auto-detected from file extension or content.
+ * @param config - Extraction configuration object. If null, uses default extraction settings.
+ * @returns Promise<ExtractionResult> containing extracted content, metadata, tables, and optional chunks/images
+ * @throws {Error} If file doesn't exist, cannot be accessed, or cannot be read
+ * @throws {ParsingError} When document format is invalid or corrupted
+ * @throws {OcrError} When OCR processing fails (if OCR is enabled)
+ * @throws {ValidationError} When extraction result fails validation (if validators registered)
+ * @throws {KreuzbergError} For other extraction-related failures
  *
  * @example
  * ```typescript
@@ -123,17 +133,23 @@ declare function extractFileSync(filePath: string, mimeType?: string | null, con
  * console.log(result2.chunks); // Array of text chunks
  * ```
  */
-declare function extractFile(filePath: string, mimeType?: string | null, config?: ExtractionConfig$1 | null): Promise<ExtractionResult>;
+declare function extractFile(filePath: string, mimeTypeOrConfig?: string | null | ExtractionConfig$1, maybeConfig?: ExtractionConfig$1 | null): Promise<ExtractionResult>;
 /**
  * Extract content from raw bytes (synchronous).
  *
  * **Usage Note**: For processing multiple byte arrays, prefer `batchExtractBytesSync()`
  * which provides better performance and memory management.
  *
- * @param data - File content as Uint8Array
- * @param mimeType - MIME type of the data (required for format detection)
- * @param config - Extraction configuration (uses defaults if null)
- * @returns ExtractionResult with content, metadata, and tables
+ * @param data - File content as Uint8Array (Buffer will be converted)
+ * @param mimeType - MIME type of the data (required for accurate format detection). Must be a valid MIME type string.
+ * @param config - Extraction configuration object. If null, uses default extraction settings.
+ * @returns ExtractionResult containing extracted content, metadata, tables, and optional chunks/images
+ * @throws {TypeError} When data is not a valid Uint8Array
+ * @throws {Error} When file cannot be read or parsed
+ * @throws {ParsingError} When document format is invalid or corrupted
+ * @throws {OcrError} When OCR processing fails (if OCR is enabled)
+ * @throws {ValidationError} When extraction result fails validation (if validators registered)
+ * @throws {KreuzbergError} For other extraction-related failures
  *
  * @example
  * ```typescript
@@ -145,17 +161,23 @@ declare function extractFile(filePath: string, mimeType?: string | null, config?
  * console.log(result.content);
  * ```
  */
-declare function extractBytesSync(data: Uint8Array, mimeType: string, config?: ExtractionConfig$1 | null): ExtractionResult;
+declare function extractBytesSync(dataOrPath: Uint8Array | string, mimeType: string, config?: ExtractionConfig$1 | null): ExtractionResult;
 /**
  * Extract content from raw bytes (asynchronous).
  *
  * **Usage Note**: For processing multiple byte arrays, prefer `batchExtractBytes()`
  * which provides better performance and memory management.
  *
- * @param data - File content as Uint8Array
- * @param mimeType - MIME type of the data (required for format detection)
- * @param config - Extraction configuration (uses defaults if null)
- * @returns Promise<ExtractionResult> with content, metadata, and tables
+ * @param data - File content as Uint8Array (Buffer will be converted)
+ * @param mimeType - MIME type of the data (required for accurate format detection). Must be a valid MIME type string.
+ * @param config - Extraction configuration object. If null, uses default extraction settings.
+ * @returns Promise<ExtractionResult> containing extracted content, metadata, tables, and optional chunks/images
+ * @throws {TypeError} When data is not a valid Uint8Array
+ * @throws {Error} When file cannot be read or parsed
+ * @throws {ParsingError} When document format is invalid or corrupted
+ * @throws {OcrError} When OCR processing fails (if OCR is enabled)
+ * @throws {ValidationError} When extraction result fails validation (if validators registered)
+ * @throws {KreuzbergError} For other extraction-related failures
  *
  * @example
  * ```typescript
@@ -167,7 +189,7 @@ declare function extractBytesSync(data: Uint8Array, mimeType: string, config?: E
  * console.log(result.content);
  * ```
  */
-declare function extractBytes(data: Uint8Array, mimeType: string, config?: ExtractionConfig$1 | null): Promise<ExtractionResult>;
+declare function extractBytes(dataOrPath: Uint8Array | string, mimeType: string, config?: ExtractionConfig$1 | null): Promise<ExtractionResult>;
 /**
  * Extract content from multiple files in parallel (synchronous).
  *
@@ -179,9 +201,14 @@ declare function extractBytes(data: Uint8Array, mimeType: string, config?: Extra
  * - Optimized memory usage across all extractions
  * - More reliable for batch document processing
  *
- * @param paths - List of file paths to extract
- * @param config - Extraction configuration (uses defaults if null)
+ * @param paths - List of file paths to extract (absolute or relative paths)
+ * @param config - Extraction configuration object. If null, uses default extraction settings.
  * @returns Array of ExtractionResults (one per file, in same order as input)
+ * @throws {Error} If any file cannot be read or parsed
+ * @throws {ParsingError} When any document format is invalid or corrupted
+ * @throws {OcrError} When OCR processing fails (if OCR is enabled)
+ * @throws {ValidationError} When any extraction result fails validation (if validators registered)
+ * @throws {KreuzbergError} For other extraction-related failures
  *
  * @example
  * ```typescript
@@ -207,9 +234,14 @@ declare function batchExtractFilesSync(paths: string[], config?: ExtractionConfi
  * - Optimized memory usage across all extractions
  * - More reliable for batch document processing
  *
- * @param paths - List of file paths to extract
- * @param config - Extraction configuration (uses defaults if null)
+ * @param paths - List of file paths to extract (absolute or relative paths)
+ * @param config - Extraction configuration object. If null, uses default extraction settings.
  * @returns Promise resolving to array of ExtractionResults (one per file, in same order as input)
+ * @throws {Error} If any file cannot be read or parsed
+ * @throws {ParsingError} When any document format is invalid or corrupted
+ * @throws {OcrError} When OCR processing fails (if OCR is enabled)
+ * @throws {ValidationError} When any extraction result fails validation (if validators registered)
+ * @throws {KreuzbergError} For other extraction-related failures
  *
  * @example
  * ```typescript
@@ -238,10 +270,16 @@ declare function batchExtractFiles(paths: string[], config?: ExtractionConfig$1
  * - Optimized memory usage across all extractions
  * - More reliable for batch document processing
  *
- * @param dataList - List of file contents as Uint8Arrays
- * @param mimeTypes - List of MIME types (one per data item, required for format detection)
- * @param config - Extraction configuration (uses defaults if null)
+ * @param dataList - List of file contents as Uint8Arrays (must be same length as mimeTypes)
+ * @param mimeTypes - List of MIME types (one per data item, required for accurate format detection)
+ * @param config - Extraction configuration object. If null, uses default extraction settings.
  * @returns Array of ExtractionResults (one per data item, in same order as input)
+ * @throws {TypeError} When dataList contains non-Uint8Array items or length mismatch with mimeTypes
+ * @throws {Error} If any data cannot be read or parsed
+ * @throws {ParsingError} When any document format is invalid or corrupted
+ * @throws {OcrError} When OCR processing fails (if OCR is enabled)
+ * @throws {ValidationError} When any extraction result fails validation (if validators registered)
+ * @throws {KreuzbergError} For other extraction-related failures
  *
  * @example
  * ```typescript
@@ -270,10 +308,16 @@ declare function batchExtractBytesSync(dataList: Uint8Array[], mimeTypes: string
  * - Optimized memory usage across all extractions
  * - More reliable for batch document processing
  *
- * @param dataList - List of file contents as Uint8Arrays
- * @param mimeTypes - List of MIME types (one per data item, required for format detection)
- * @param config - Extraction configuration (uses defaults if null)
+ * @param dataList - List of file contents as Uint8Arrays (must be same length as mimeTypes)
+ * @param mimeTypes - List of MIME types (one per data item, required for accurate format detection)
+ * @param config - Extraction configuration object. If null, uses default extraction settings.
  * @returns Promise resolving to array of ExtractionResults (one per data item, in same order as input)
+ * @throws {TypeError} When dataList contains non-Uint8Array items or length mismatch with mimeTypes
+ * @throws {Error} If any data cannot be read or parsed
+ * @throws {ParsingError} When any document format is invalid or corrupted
+ * @throws {OcrError} When OCR processing fails (if OCR is enabled)
+ * @throws {ValidationError} When any extraction result fails validation (if validators registered)
+ * @throws {KreuzbergError} For other extraction-related failures
  *
  * @example
  * ```typescript
@@ -306,7 +350,10 @@ declare function batchExtractBytes(dataList: Uint8Array[], mimeTypes: string[],
  * preventing JavaScript callbacks from executing. For v4.0, use async extraction
  * when you need custom processors.
  *
- * @param processor - PostProcessorProtocol implementation
+ * @param processor - PostProcessorProtocol implementation with name(), process(), and optional processingStage()
+ * @throws {Error} If processor is missing required methods (name or process)
+ * @throws {Error} If processor name is empty string
+ * @throws {Error} If a processor with the same name is already registered
  *
  * @example
  * ```typescript
@@ -339,8 +386,9 @@ declare function registerPostProcessor(processor: PostProcessorProtocol): void;
  * Unregister a postprocessor by name.
  *
  * Removes a previously registered postprocessor from the registry.
+ * If the processor doesn't exist, this is a no-op (does not throw).
  *
- * @param name - Name of the processor to unregister
+ * @param name - Name of the processor to unregister (case-sensitive)
  *
  * @example
  * ```typescript
@@ -353,7 +401,8 @@ declare function unregisterPostProcessor(name: string): void;
 /**
  * Clear all registered postprocessors.
  *
- * Removes all postprocessors from the registry.
+ * Removes all postprocessors from the registry. Useful for test cleanup or resetting state.
+ * If no postprocessors are registered, this is a no-op.
  *
  * @example
  * ```typescript
@@ -366,9 +415,9 @@ declare function clearPostProcessors(): void;
 /**
  * List all registered post-processors.
  *
- * Returns the names of all currently registered post-processors.
+ * Returns the names of all currently registered post-processors (both built-in and custom).
  *
- * @returns Array of post-processor names
+ * @returns Array of post-processor names (empty array if none registered)
  *
  * @example
  * ```typescript
@@ -386,7 +435,10 @@ declare function listPostProcessors(): string[];
  * Unlike post-processors, validator errors **fail fast** - if a validator throws an error,
  * the extraction fails immediately.
  *
- * @param validator - ValidatorProtocol implementation
+ * @param validator - ValidatorProtocol implementation with name(), validate(), and optional priority()/shouldValidate()
+ * @throws {Error} If validator is missing required methods (name or validate)
+ * @throws {Error} If validator name is empty string
+ * @throws {Error} If a validator with the same name is already registered
  *
  * @example
  * ```typescript
@@ -416,8 +468,9 @@ declare function registerValidator(validator: ValidatorProtocol): void;
  * Unregister a validator by name.
  *
  * Removes a previously registered validator from the global registry.
+ * If the validator doesn't exist, this is a no-op (does not throw).
  *
- * @param name - Validator name to unregister
+ * @param name - Validator name to unregister (case-sensitive)
  *
  * @example
  * ```typescript
@@ -444,9 +497,9 @@ declare function clearValidators(): void;
 /**
  * List all registered validators.
  *
- * Returns the names of all currently registered validators.
+ * Returns the names of all currently registered validators (both built-in and custom).
  *
- * @returns Array of validator names
+ * @returns Array of validator names (empty array if none registered)
  *
  * @example
  * ```typescript
@@ -464,7 +517,7 @@ declare function registerOcrBackend(backend: OcrBackendProtocol): void;
  * Returns an array of names of all currently registered OCR backends,
  * including built-in backends like "tesseract".
  *
- * @returns Array of OCR backend names
+ * @returns Array of OCR backend names (empty array if none registered)
  *
  * @example
  * ```typescript
@@ -497,7 +550,7 @@ declare function unregisterOcrBackend(name: string): void;
  *
  * Removes all OCR backends from the registry, including built-in backends.
  * Use with caution as this will make OCR functionality unavailable until
- * backends are re-registered.
+ * backends are re-registered. If no backends are registered, this is a no-op.
  *
  * @example
  * ```typescript
@@ -513,7 +566,7 @@ declare function clearOcrBackends(): void;
  * Returns an array of names of all currently registered document extractors,
  * including built-in extractors for PDF, Office documents, images, etc.
  *
- * @returns Array of document extractor names
+ * @returns Array of document extractor names (empty array if none registered)
  *
  * @example
  * ```typescript
@@ -559,18 +612,26 @@ declare function clearDocumentExtractors(): void;
 /**
  * ExtractionConfig namespace with static methods for loading configuration from files.
  *
- * Provides a factory method to load extraction configuration from TOML, YAML, or JSON files.
- * The file format is automatically detected based on the file extension.
+ * Provides factory methods to load extraction configuration from TOML, YAML, or JSON files,
+ * or to discover configuration files in the current directory tree.
+ *
+ * For creating configurations programmatically, use plain TypeScript objects instead:
  *
  * @example
  * ```typescript
  * import { ExtractionConfig, extractFile } from '@kreuzberg/node';
  *
  * // Load configuration from file
- * const config = ExtractionConfig.fromFile('config.toml');
+ * const config1 = ExtractionConfig.fromFile('config.toml');
+ *
+ * // Or create with plain object
+ * const config2 = {
+ *   chunking: { maxChars: 2048 },
+ *   ocr: { backend: 'tesseract', language: 'eng' }
+ * };
  *
  * // Use with extraction
- * const result = await extractFile('document.pdf', null, config);
+ * const result = await extractFile('document.pdf', null, config2);
  * ```
  */
 declare const ExtractionConfig: {
@@ -658,30 +719,30 @@ declare function detectMimeType(bytes: Buffer): string;
 /**
  * Detect MIME type from a file path.
  *
- * Uses file extension to determine MIME type. Falls back to `mime_guess` crate
- * if extension-based detection fails.
+ * Determines the MIME type based on the file extension in the provided path.
+ * By default, checks if the file exists; can be disabled with checkExists parameter.
  *
- * @param path - Path to the file (string)
- * @param checkExists - Whether to verify file existence (default: true)
- * @returns The detected MIME type string
+ * @param filePath - The file path to detect MIME type from (e.g., 'document.pdf')
+ * @param checkExists - Whether to verify the file exists (default: true)
+ * @returns The detected MIME type as a string (e.g., 'application/pdf')
  *
- * @throws {Error} If file doesn't exist (when checkExists is true)
- * @throws {Error} If MIME type cannot be determined from path/extension
- * @throws {Error} If extension is unknown
+ * @throws {Error} If MIME type cannot be determined from the file extension,
+ * or if checkExists is true and the file does not exist
  *
  * @example
  * ```typescript
  * import { detectMimeTypeFromPath } from '@kreuzberg/node';
  *
- * // Detect from existing file
- * const mimeType = detectMimeTypeFromPath('document.pdf');
+ * // Detect MIME type from existing file
+ * const mimeType = detectMimeTypeFromPath('/path/to/document.pdf');
  * console.log(mimeType); // 'application/pdf'
  *
- * const mimeType2 = detectMimeTypeFromPath('document.docx');
+ * // Detect without checking file existence
+ * const mimeType2 = detectMimeTypeFromPath('document.docx', false);
  * console.log(mimeType2); // 'application/vnd.openxmlformats-officedocument.wordprocessingml.document'
  * ```
  */
-declare function detectMimeTypeFromPath(path: string, checkExists?: boolean): string;
+declare function detectMimeTypeFromPath(filePath: string, checkExists?: boolean): string;
 /**
  * Validate that a MIME type is supported by Kreuzberg.
  *
@@ -852,6 +913,220 @@ declare function getLastErrorCode(): number;
  * ```
  */
 declare function getLastPanicContext(): PanicContext | null;
-declare const __version__ = "4.0.0-rc.8";
+/**
+ * Returns the human-readable name for an error code.
+ *
+ * Maps numeric error codes to their string names, providing a consistent way
+ * to get error code names across all platforms.
+ *
+ * @param code - The numeric error code (0-7)
+ * @returns The error code name as a string (e.g., "validation", "ocr", "unknown")
+ *
+ * @example
+ * ```typescript
+ * import { getErrorCodeName } from '@kreuzberg/node';
+ *
+ * const name = getErrorCodeName(0);  // returns "validation"
+ * const name = getErrorCodeName(2);  // returns "ocr"
+ * const name = getErrorCodeName(99); // returns "unknown"
+ * ```
+ */
+declare function getErrorCodeName(code: number): string;
+/**
+ * Returns the description for an error code.
+ *
+ * Retrieves user-friendly descriptions of error types from the FFI layer.
+ *
+ * @param code - The numeric error code (0-7)
+ * @returns A brief description of the error type
+ *
+ * @example
+ * ```typescript
+ * import { getErrorCodeDescription } from '@kreuzberg/node';
+ *
+ * const desc = getErrorCodeDescription(0);  // returns "Input validation error"
+ * const desc = getErrorCodeDescription(4);  // returns "File system I/O error"
+ * const desc = getErrorCodeDescription(99); // returns "Unknown error code"
+ * ```
+ */
+declare function getErrorCodeDescription(code: number): string;
+/**
+ * Classifies an error message string into an error code category.
+ *
+ * This function analyzes the error message content and returns the most likely
+ * error code (0-7) based on keyword patterns. Used to programmatically classify
+ * errors for handling purposes.
+ *
+ * The classification is based on keyword matching:
+ * - **Validation (0)**: Keywords like "invalid", "validation", "schema", "required"
+ * - **Parsing (1)**: Keywords like "parsing", "corrupted", "malformed"
+ * - **Ocr (2)**: Keywords like "ocr", "tesseract", "language", "model"
+ * - **MissingDependency (3)**: Keywords like "not found", "missing", "dependency"
+ * - **Io (4)**: Keywords like "file", "disk", "read", "write", "permission"
+ * - **Plugin (5)**: Keywords like "plugin", "register", "extension"
+ * - **UnsupportedFormat (6)**: Keywords like "unsupported", "format", "mime"
+ * - **Internal (7)**: Keywords like "internal", "bug", "panic"
+ *
+ * @param errorMessage - The error message string to classify
+ * @returns An object with the classification details
+ *
+ * @example
+ * ```typescript
+ * import { classifyError } from '@kreuzberg/node';
+ *
+ * const result = classifyError("PDF file is corrupted");
+ * // Returns: { code: 1, name: "parsing", confidence: 0.95 }
+ *
+ * const result = classifyError("Tesseract not found");
+ * // Returns: { code: 3, name: "missing_dependency", confidence: 0.9 }
+ * ```
+ */
+declare function classifyError(errorMessage: string): ErrorClassification;
+/**
+ * Create a worker pool for concurrent file extraction.
+ *
+ * The worker pool manages a set of background worker threads that can process
+ * extraction requests concurrently, improving throughput when handling multiple files.
+ *
+ * @param size - Optional number of worker threads (defaults to CPU count). Must be > 0
+ * @returns A WorkerPool instance to use with extraction functions
+ *
+ * @throws {Error} If size is invalid or pool creation fails
+ *
+ * @example
+ * ```typescript
+ * import { createWorkerPool, extractFileInWorker, closeWorkerPool } from '@kreuzberg/node';
+ *
+ * // Create pool with 4 workers
+ * const pool = createWorkerPool(4);
+ *
+ * try {
+ *   const result = await extractFileInWorker(pool, 'document.pdf');
+ *   console.log(result.content);
+ * } finally {
+ *   // Always close the pool when done
+ *   await closeWorkerPool(pool);
+ * }
+ * ```
+ */
+declare function createWorkerPool(size?: number): WorkerPool;
+/**
+ * Get statistics about a worker pool.
+ *
+ * Returns information about the pool's current state, including the number of active workers,
+ * queued tasks, and total processed tasks.
+ *
+ * @param pool - The worker pool instance
+ * @returns WorkerPoolStats with pool information
+ *
+ * @example
+ * ```typescript
+ * import { createWorkerPool, getWorkerPoolStats } from '@kreuzberg/node';
+ *
+ * const pool = createWorkerPool(4);
+ * const stats = getWorkerPoolStats(pool);
+ *
+ * console.log(`Pool size: ${stats.size}`);
+ * console.log(`Active workers: ${stats.activeWorkers}`);
+ * console.log(`Queued tasks: ${stats.queuedTasks}`);
+ * ```
+ */
+declare function getWorkerPoolStats(pool: WorkerPool): WorkerPoolStats;
+/**
+ * Extract content from a single file using a worker pool (asynchronous).
+ *
+ * Submits an extraction task to the worker pool. The task is executed by one of the
+ * available workers in the background, allowing other tasks to be processed concurrently.
+ *
+ * @param pool - The worker pool instance
+ * @param filePath - Path to the file to extract
+ * @param mimeTypeOrConfig - Optional MIME type or extraction configuration
+ * @param maybeConfig - Optional extraction configuration (if second param is MIME type)
+ * @returns Promise<ExtractionResult> containing extracted content and metadata
+ *
+ * @throws {Error} If the file cannot be read or extraction fails
+ *
+ * @example
+ * ```typescript
+ * import { createWorkerPool, extractFileInWorker, closeWorkerPool } from '@kreuzberg/node';
+ *
+ * const pool = createWorkerPool(4);
+ *
+ * try {
+ *   const files = ['doc1.pdf', 'doc2.docx', 'doc3.xlsx'];
+ *   const results = await Promise.all(
+ *     files.map(f => extractFileInWorker(pool, f))
+ *   );
+ *
+ *   results.forEach((r, i) => {
+ *     console.log(`${files[i]}: ${r.content.substring(0, 100)}...`);
+ *   });
+ * } finally {
+ *   await closeWorkerPool(pool);
+ * }
+ * ```
+ */
+declare function extractFileInWorker(pool: WorkerPool, filePath: string, mimeTypeOrConfig?: string | null | ExtractionConfig$1, maybeConfig?: ExtractionConfig$1 | null): Promise<ExtractionResult>;
+/**
+ * Extract content from multiple files in parallel using a worker pool (asynchronous).
+ *
+ * Submits multiple extraction tasks to the worker pool for concurrent processing.
+ * This is more efficient than using `extractFileInWorker` multiple times sequentially.
+ *
+ * @param pool - The worker pool instance
+ * @param paths - Array of file paths to extract
+ * @param config - Extraction configuration object (applies to all files)
+ * @returns Promise<ExtractionResult[]> array of results (one per file, in same order)
+ *
+ * @throws {Error} If any file cannot be read or extraction fails
+ *
+ * @example
+ * ```typescript
+ * import { createWorkerPool, batchExtractFilesInWorker, closeWorkerPool } from '@kreuzberg/node';
+ *
+ * const pool = createWorkerPool(4);
+ *
+ * try {
+ *   const files = ['invoice1.pdf', 'invoice2.pdf', 'invoice3.pdf'];
+ *   const results = await batchExtractFilesInWorker(pool, files, {
+ *     ocr: { backend: 'tesseract', language: 'eng' }
+ *   });
+ *
+ *   const total = results.reduce((sum, r) => sum + extractAmount(r.content), 0);
+ *   console.log(`Total: $${total}`);
+ * } finally {
+ *   await closeWorkerPool(pool);
+ * }
+ * ```
+ */
+declare function batchExtractFilesInWorker(pool: WorkerPool, paths: string[], config?: ExtractionConfig$1 | null): Promise<ExtractionResult[]>;
+/**
+ * Close a worker pool and shut down all worker threads.
+ *
+ * Should be called when the pool is no longer needed to clean up resources
+ * and gracefully shut down worker threads. Any pending tasks will be cancelled.
+ *
+ * @param pool - The worker pool instance to close
+ * @returns Promise that resolves when the pool is fully closed
+ *
+ * @throws {Error} If pool shutdown fails
+ *
+ * @example
+ * ```typescript
+ * import { createWorkerPool, extractFileInWorker, closeWorkerPool } from '@kreuzberg/node';
+ *
+ * const pool = createWorkerPool(4);
+ *
+ * try {
+ *   const result = await extractFileInWorker(pool, 'document.pdf');
+ *   console.log(result.content);
+ * } finally {
+ *   // Clean up the pool
+ *   await closeWorkerPool(pool);
+ * }
+ * ```
+ */
+declare function closeWorkerPool(pool: WorkerPool): Promise<void>;
+declare const __version__ = "4.0.0";
 
-export { type EmbeddingPreset, ExtractionConfig, ExtractionResult, OcrBackendProtocol, PanicContext, PostProcessorProtocol, ValidatorProtocol, __resetBindingForTests, __setBindingForTests, __version__, batchExtractBytes, batchExtractBytesSync, batchExtractFiles, batchExtractFilesSync, clearDocumentExtractors, clearOcrBackends, clearPostProcessors, clearValidators, detectMimeType, detectMimeTypeFromPath, extractBytes, extractBytesSync, extractFile, extractFileSync, getEmbeddingPreset, getExtensionsForMime, getLastErrorCode, getLastPanicContext, listDocumentExtractors, listEmbeddingPresets, listOcrBackends, listPostProcessors, listValidators, registerOcrBackend, registerPostProcessor, registerValidator, unregisterDocumentExtractor, unregisterOcrBackend, unregisterPostProcessor, unregisterValidator, validateMimeType };
+export { type EmbeddingPreset, ErrorClassification, ExtractionConfig, ExtractionResult, OcrBackendProtocol, PanicContext, PostProcessorProtocol, ValidatorProtocol, WorkerPool, WorkerPoolStats, __resetBindingForTests, __setBindingForTests, __version__, batchExtractBytes, batchExtractBytesSync, batchExtractFiles, batchExtractFilesInWorker, batchExtractFilesSync, classifyError, clearDocumentExtractors, clearOcrBackends, clearPostProcessors, clearValidators, closeWorkerPool, createWorkerPool, detectMimeType, detectMimeTypeFromPath, extractBytes, extractBytesSync, extractFile, extractFileInWorker, extractFileSync, getEmbeddingPreset, getErrorCodeDescription, getErrorCodeName, getExtensionsForMime, getLastErrorCode, getLastPanicContext, getWorkerPoolStats, listDocumentExtractors, listEmbeddingPresets, listOcrBackends, listPostProcessors, listValidators, registerOcrBackend, registerPostProcessor, registerValidator, unregisterDocumentExtractor, unregisterOcrBackend, unregisterPostProcessor, unregisterValidator, validateMimeType };