@kreuzberg/node 4.0.0-rc.21 → 4.0.0-rc.24

package/dist/cli.d.mts

@@ -4,6 +4,10 @@
  *
  * This keeps `npx kreuzberg` working without shipping an additional TypeScript CLI implementation.
  */
+declare global {
+    var __filename: string | undefined;
+    var __dirname: string | undefined;
+}
 declare function main(argv: string[]): number;
 
 export { main };

package/dist/cli.d.ts

@@ -4,6 +4,10 @@
  *
  * This keeps `npx kreuzberg` working without shipping an additional TypeScript CLI implementation.
  */
+declare global {
+    var __filename: string | undefined;
+    var __dirname: string | undefined;
+}
 declare function main(argv: string[]): number;
 
 export { main };

package/dist/cli.js

@@ -37,7 +37,17 @@ var import_node_fs = require("node:fs");
 var import_node_path = require("node:path");
 var import_node_url = require("node:url");
 var import_which = __toESM(require("which"));
-const import_meta = {};
+function getDirectory() {
+  if (typeof __filename !== "undefined") {
+    return (0, import_node_path.dirname)(__filename);
+  }
+  try {
+    const url = eval("import.meta.url");
+    return (0, import_node_path.dirname)((0, import_node_url.fileURLToPath)(url));
+  } catch {
+    return process.cwd();
+  }
+}
 function main(argv) {
   const args = argv.slice(2);
   let cliPath;
@@ -46,7 +56,7 @@ function main(argv) {
   } catch {
   }
   if (!cliPath) {
-    const __dirname = typeof __filename !== "undefined" ? (0, import_node_path.dirname)(__filename) : (0, import_node_path.dirname)((0, import_node_url.fileURLToPath)(import_meta.url));
+    const __dirname = getDirectory();
     const devBinary = (0, import_node_path.join)(__dirname, "..", "..", "..", "target", "release", "kreuzberg");
     if ((0, import_node_fs.existsSync)(devBinary)) {
       cliPath = devBinary;

package/dist/cli.js.map

@@ -1 +1 @@
-{"version":3,"sources":["../typescript/cli.ts"],"sourcesContent":["#!/usr/bin/env node\n\n/**\n * Proxy entry point that forwards to the Rust-based Kreuzberg CLI.\n *\n * This keeps `npx kreuzberg` working without shipping an additional TypeScript CLI implementation.\n */\n\nimport { spawnSync } from \"node:child_process\";\nimport { existsSync } from \"node:fs\";\nimport { dirname, join } from \"node:path\";\nimport { fileURLToPath } from \"node:url\";\nimport which from \"which\";\n\nfunction main(argv: string[]): number {\n\tconst args = argv.slice(2);\n\n\tlet cliPath: string | undefined;\n\ttry {\n\t\tcliPath = which.sync(\"kreuzberg-cli\");\n\t} catch {}\n\n\tif (!cliPath) {\n\t\tconst __dirname = typeof __filename !== \"undefined\" ? dirname(__filename) : dirname(fileURLToPath(import.meta.url));\n\t\tconst devBinary = join(__dirname, \"..\", \"..\", \"..\", \"target\", \"release\", \"kreuzberg\");\n\t\tif (existsSync(devBinary)) {\n\t\t\tcliPath = devBinary;\n\t\t}\n\t}\n\n\tif (!cliPath) {\n\t\tconsole.error(\n\t\t\t\"The embedded Kreuzberg CLI binary could not be located. \" +\n\t\t\t\t\"This indicates a packaging issue; please open an issue at \" +\n\t\t\t\t\"https://github.com/kreuzberg-dev/kreuzberg/issues so we can investigate.\",\n\t\t);\n\t\treturn 1;\n\t}\n\n\tconst result = spawnSync(cliPath, args, {\n\t\tstdio: \"inherit\",\n\t\tshell: false,\n\t});\n\n\tif (result.error) {\n\t\tconsole.error(`Failed to execute kreuzberg-cli: ${result.error.message}`);\n\t\treturn 1;\n\t}\n\n\treturn result.status ?? 1;\n}\n\nif (require.main === module) {\n\tprocess.exit(main(process.argv));\n}\n\nexport { main };\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA;AAAA;AAAA;AAAA;AAAA;AAQA,gCAA0B;AAC1B,qBAA2B;AAC3B,uBAA8B;AAC9B,sBAA8B;AAC9B,mBAAkB;AAZlB;AAcA,SAAS,KAAK,MAAwB;AACrC,QAAM,OAAO,KAAK,MAAM,CAAC;AAEzB,MAAI;AACJ,MAAI;AACH,cAAU,aAAAA,QAAM,KAAK,eAAe;AAAA,EACrC,QAAQ;AAAA,EAAC;AAET,MAAI,CAAC,SAAS;AACb,UAAM,YAAY,OAAO,eAAe,kBAAc,0BAAQ,UAAU,QAAI,8BAAQ,+BAAc,YAAY,GAAG,CAAC;AAClH,UAAM,gBAAY,uBAAK,WAAW,MAAM,MAAM,MAAM,UAAU,WAAW,WAAW;AACpF,YAAI,2BAAW,SAAS,GAAG;AAC1B,gBAAU;AAAA,IACX;AAAA,EACD;AAEA,MAAI,CAAC,SAAS;AACb,YAAQ;AAAA,MACP;AAAA,IAGD;AACA,WAAO;AAAA,EACR;AAEA,QAAM,aAAS,qCAAU,SAAS,MAAM;AAAA,IACvC,OAAO;AAAA,IACP,OAAO;AAAA,EACR,CAAC;AAED,MAAI,OAAO,OAAO;AACjB,YAAQ,MAAM,oCAAoC,OAAO,MAAM,OAAO,EAAE;AACxE,WAAO;AAAA,EACR;AAEA,SAAO,OAAO,UAAU;AACzB;AAEA,IAAI,QAAQ,SAAS,QAAQ;AAC5B,UAAQ,KAAK,KAAK,QAAQ,IAAI,CAAC;AAChC;","names":["which"]}
+{"version":3,"sources":["../typescript/cli.ts"],"sourcesContent":["#!/usr/bin/env node\n\n/**\n * Proxy entry point that forwards to the Rust-based Kreuzberg CLI.\n *\n * This keeps `npx kreuzberg` working without shipping an additional TypeScript CLI implementation.\n */\n\nimport { spawnSync } from \"node:child_process\";\nimport { existsSync } from \"node:fs\";\nimport { dirname, join } from \"node:path\";\nimport { fileURLToPath } from \"node:url\";\nimport which from \"which\";\n\ndeclare global {\n\tvar __filename: string | undefined;\n\tvar __dirname: string | undefined;\n}\n\nfunction getDirectory(): string {\n\t// In CJS, __filename will be defined\n\tif (typeof __filename !== \"undefined\") {\n\t\treturn dirname(__filename);\n\t}\n\t// Fallback for ESM\n\ttry {\n\t\t// Use eval to avoid esbuild warnings about import.meta in CJS builds\n\t\t// @ts-ignore - import.meta is only available in ESM\n\t\tconst url = eval(\"import.meta.url\");\n\t\treturn dirname(fileURLToPath(url));\n\t} catch {\n\t\treturn process.cwd();\n\t}\n}\n\nfunction main(argv: string[]): number {\n\tconst args = argv.slice(2);\n\n\tlet cliPath: string | undefined;\n\ttry {\n\t\tcliPath = which.sync(\"kreuzberg-cli\");\n\t} catch {}\n\n\tif (!cliPath) {\n\t\tconst __dirname = getDirectory();\n\t\tconst devBinary = join(__dirname, \"..\", \"..\", \"..\", \"target\", \"release\", \"kreuzberg\");\n\t\tif (existsSync(devBinary)) {\n\t\t\tcliPath = devBinary;\n\t\t}\n\t}\n\n\tif (!cliPath) {\n\t\tconsole.error(\n\t\t\t\"The embedded Kreuzberg CLI binary could not be located. \" +\n\t\t\t\t\"This indicates a packaging issue; please open an issue at \" +\n\t\t\t\t\"https://github.com/kreuzberg-dev/kreuzberg/issues so we can investigate.\",\n\t\t);\n\t\treturn 1;\n\t}\n\n\tconst result = spawnSync(cliPath, args, {\n\t\tstdio: \"inherit\",\n\t\tshell: false,\n\t});\n\n\tif (result.error) {\n\t\tconsole.error(`Failed to execute kreuzberg-cli: ${result.error.message}`);\n\t\treturn 1;\n\t}\n\n\treturn result.status ?? 1;\n}\n\nif (require.main === module) {\n\tprocess.exit(main(process.argv));\n}\n\nexport { main };\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;;;;;;;;;AAAA;AAAA;AAAA;AAAA;AAAA;AAQA,gCAA0B;AAC1B,qBAA2B;AAC3B,uBAA8B;AAC9B,sBAA8B;AAC9B,mBAAkB;AAOlB,SAAS,eAAuB;AAE/B,MAAI,OAAO,eAAe,aAAa;AACtC,eAAO,0BAAQ,UAAU;AAAA,EAC1B;AAEA,MAAI;AAGH,UAAM,MAAM,KAAK,iBAAiB;AAClC,eAAO,8BAAQ,+BAAc,GAAG,CAAC;AAAA,EAClC,QAAQ;AACP,WAAO,QAAQ,IAAI;AAAA,EACpB;AACD;AAEA,SAAS,KAAK,MAAwB;AACrC,QAAM,OAAO,KAAK,MAAM,CAAC;AAEzB,MAAI;AACJ,MAAI;AACH,cAAU,aAAAA,QAAM,KAAK,eAAe;AAAA,EACrC,QAAQ;AAAA,EAAC;AAET,MAAI,CAAC,SAAS;AACb,UAAM,YAAY,aAAa;AAC/B,UAAM,gBAAY,uBAAK,WAAW,MAAM,MAAM,MAAM,UAAU,WAAW,WAAW;AACpF,YAAI,2BAAW,SAAS,GAAG;AAC1B,gBAAU;AAAA,IACX;AAAA,EACD;AAEA,MAAI,CAAC,SAAS;AACb,YAAQ;AAAA,MACP;AAAA,IAGD;AACA,WAAO;AAAA,EACR;AAEA,QAAM,aAAS,qCAAU,SAAS,MAAM;AAAA,IACvC,OAAO;AAAA,IACP,OAAO;AAAA,EACR,CAAC;AAED,MAAI,OAAO,OAAO;AACjB,YAAQ,MAAM,oCAAoC,OAAO,MAAM,OAAO,EAAE;AACxE,WAAO;AAAA,EACR;AAEA,SAAO,OAAO,UAAU;AACzB;AAEA,IAAI,QAAQ,SAAS,QAAQ;AAC5B,UAAQ,KAAK,KAAK,QAAQ,IAAI,CAAC;AAChC;","names":["which"]}

package/dist/cli.mjs

@@ -4,6 +4,17 @@ import { existsSync } from "node:fs";
 import { dirname, join } from "node:path";
 import { fileURLToPath } from "node:url";
 import which from "which";
+function getDirectory() {
+  if (typeof __filename !== "undefined") {
+    return dirname(__filename);
+  }
+  try {
+    const url = eval("import.meta.url");
+    return dirname(fileURLToPath(url));
+  } catch {
+    return process.cwd();
+  }
+}
 function main(argv) {
   const args = argv.slice(2);
   let cliPath;
@@ -12,7 +23,7 @@ function main(argv) {
   } catch {
   }
   if (!cliPath) {
-    const __dirname = typeof __filename !== "undefined" ? dirname(__filename) : dirname(fileURLToPath(import.meta.url));
+    const __dirname = getDirectory();
     const devBinary = join(__dirname, "..", "..", "..", "target", "release", "kreuzberg");
     if (existsSync(devBinary)) {
       cliPath = devBinary;

package/dist/cli.mjs.map

@@ -1 +1 @@
-{"version":3,"sources":["../typescript/cli.ts"],"sourcesContent":["#!/usr/bin/env node\n\n/**\n * Proxy entry point that forwards to the Rust-based Kreuzberg CLI.\n *\n * This keeps `npx kreuzberg` working without shipping an additional TypeScript CLI implementation.\n */\n\nimport { spawnSync } from \"node:child_process\";\nimport { existsSync } from \"node:fs\";\nimport { dirname, join } from \"node:path\";\nimport { fileURLToPath } from \"node:url\";\nimport which from \"which\";\n\nfunction main(argv: string[]): number {\n\tconst args = argv.slice(2);\n\n\tlet cliPath: string | undefined;\n\ttry {\n\t\tcliPath = which.sync(\"kreuzberg-cli\");\n\t} catch {}\n\n\tif (!cliPath) {\n\t\tconst __dirname = typeof __filename !== \"undefined\" ? dirname(__filename) : dirname(fileURLToPath(import.meta.url));\n\t\tconst devBinary = join(__dirname, \"..\", \"..\", \"..\", \"target\", \"release\", \"kreuzberg\");\n\t\tif (existsSync(devBinary)) {\n\t\t\tcliPath = devBinary;\n\t\t}\n\t}\n\n\tif (!cliPath) {\n\t\tconsole.error(\n\t\t\t\"The embedded Kreuzberg CLI binary could not be located. \" +\n\t\t\t\t\"This indicates a packaging issue; please open an issue at \" +\n\t\t\t\t\"https://github.com/kreuzberg-dev/kreuzberg/issues so we can investigate.\",\n\t\t);\n\t\treturn 1;\n\t}\n\n\tconst result = spawnSync(cliPath, args, {\n\t\tstdio: \"inherit\",\n\t\tshell: false,\n\t});\n\n\tif (result.error) {\n\t\tconsole.error(`Failed to execute kreuzberg-cli: ${result.error.message}`);\n\t\treturn 1;\n\t}\n\n\treturn result.status ?? 1;\n}\n\nif (require.main === module) {\n\tprocess.exit(main(process.argv));\n}\n\nexport { main };\n"],"mappings":";AAQA,SAAS,iBAAiB;AAC1B,SAAS,kBAAkB;AAC3B,SAAS,SAAS,YAAY;AAC9B,SAAS,qBAAqB;AAC9B,OAAO,WAAW;AAElB,SAAS,KAAK,MAAwB;AACrC,QAAM,OAAO,KAAK,MAAM,CAAC;AAEzB,MAAI;AACJ,MAAI;AACH,cAAU,MAAM,KAAK,eAAe;AAAA,EACrC,QAAQ;AAAA,EAAC;AAET,MAAI,CAAC,SAAS;AACb,UAAM,YAAY,OAAO,eAAe,cAAc,QAAQ,UAAU,IAAI,QAAQ,cAAc,YAAY,GAAG,CAAC;AAClH,UAAM,YAAY,KAAK,WAAW,MAAM,MAAM,MAAM,UAAU,WAAW,WAAW;AACpF,QAAI,WAAW,SAAS,GAAG;AAC1B,gBAAU;AAAA,IACX;AAAA,EACD;AAEA,MAAI,CAAC,SAAS;AACb,YAAQ;AAAA,MACP;AAAA,IAGD;AACA,WAAO;AAAA,EACR;AAEA,QAAM,SAAS,UAAU,SAAS,MAAM;AAAA,IACvC,OAAO;AAAA,IACP,OAAO;AAAA,EACR,CAAC;AAED,MAAI,OAAO,OAAO;AACjB,YAAQ,MAAM,oCAAoC,OAAO,MAAM,OAAO,EAAE;AACxE,WAAO;AAAA,EACR;AAEA,SAAO,OAAO,UAAU;AACzB;AAEA,IAAI,QAAQ,SAAS,QAAQ;AAC5B,UAAQ,KAAK,KAAK,QAAQ,IAAI,CAAC;AAChC;","names":[]}
+{"version":3,"sources":["../typescript/cli.ts"],"sourcesContent":["#!/usr/bin/env node\n\n/**\n * Proxy entry point that forwards to the Rust-based Kreuzberg CLI.\n *\n * This keeps `npx kreuzberg` working without shipping an additional TypeScript CLI implementation.\n */\n\nimport { spawnSync } from \"node:child_process\";\nimport { existsSync } from \"node:fs\";\nimport { dirname, join } from \"node:path\";\nimport { fileURLToPath } from \"node:url\";\nimport which from \"which\";\n\ndeclare global {\n\tvar __filename: string | undefined;\n\tvar __dirname: string | undefined;\n}\n\nfunction getDirectory(): string {\n\t// In CJS, __filename will be defined\n\tif (typeof __filename !== \"undefined\") {\n\t\treturn dirname(__filename);\n\t}\n\t// Fallback for ESM\n\ttry {\n\t\t// Use eval to avoid esbuild warnings about import.meta in CJS builds\n\t\t// @ts-ignore - import.meta is only available in ESM\n\t\tconst url = eval(\"import.meta.url\");\n\t\treturn dirname(fileURLToPath(url));\n\t} catch {\n\t\treturn process.cwd();\n\t}\n}\n\nfunction main(argv: string[]): number {\n\tconst args = argv.slice(2);\n\n\tlet cliPath: string | undefined;\n\ttry {\n\t\tcliPath = which.sync(\"kreuzberg-cli\");\n\t} catch {}\n\n\tif (!cliPath) {\n\t\tconst __dirname = getDirectory();\n\t\tconst devBinary = join(__dirname, \"..\", \"..\", \"..\", \"target\", \"release\", \"kreuzberg\");\n\t\tif (existsSync(devBinary)) {\n\t\t\tcliPath = devBinary;\n\t\t}\n\t}\n\n\tif (!cliPath) {\n\t\tconsole.error(\n\t\t\t\"The embedded Kreuzberg CLI binary could not be located. \" +\n\t\t\t\t\"This indicates a packaging issue; please open an issue at \" +\n\t\t\t\t\"https://github.com/kreuzberg-dev/kreuzberg/issues so we can investigate.\",\n\t\t);\n\t\treturn 1;\n\t}\n\n\tconst result = spawnSync(cliPath, args, {\n\t\tstdio: \"inherit\",\n\t\tshell: false,\n\t});\n\n\tif (result.error) {\n\t\tconsole.error(`Failed to execute kreuzberg-cli: ${result.error.message}`);\n\t\treturn 1;\n\t}\n\n\treturn result.status ?? 1;\n}\n\nif (require.main === module) {\n\tprocess.exit(main(process.argv));\n}\n\nexport { main };\n"],"mappings":";AAQA,SAAS,iBAAiB;AAC1B,SAAS,kBAAkB;AAC3B,SAAS,SAAS,YAAY;AAC9B,SAAS,qBAAqB;AAC9B,OAAO,WAAW;AAOlB,SAAS,eAAuB;AAE/B,MAAI,OAAO,eAAe,aAAa;AACtC,WAAO,QAAQ,UAAU;AAAA,EAC1B;AAEA,MAAI;AAGH,UAAM,MAAM,KAAK,iBAAiB;AAClC,WAAO,QAAQ,cAAc,GAAG,CAAC;AAAA,EAClC,QAAQ;AACP,WAAO,QAAQ,IAAI;AAAA,EACpB;AACD;AAEA,SAAS,KAAK,MAAwB;AACrC,QAAM,OAAO,KAAK,MAAM,CAAC;AAEzB,MAAI;AACJ,MAAI;AACH,cAAU,MAAM,KAAK,eAAe;AAAA,EACrC,QAAQ;AAAA,EAAC;AAET,MAAI,CAAC,SAAS;AACb,UAAM,YAAY,aAAa;AAC/B,UAAM,YAAY,KAAK,WAAW,MAAM,MAAM,MAAM,UAAU,WAAW,WAAW;AACpF,QAAI,WAAW,SAAS,GAAG;AAC1B,gBAAU;AAAA,IACX;AAAA,EACD;AAEA,MAAI,CAAC,SAAS;AACb,YAAQ;AAAA,MACP;AAAA,IAGD;AACA,WAAO;AAAA,EACR;AAEA,QAAM,SAAS,UAAU,SAAS,MAAM;AAAA,IACvC,OAAO;AAAA,IACP,OAAO;AAAA,EACR,CAAC;AAED,MAAI,OAAO,OAAO;AACjB,YAAQ,MAAM,oCAAoC,OAAO,MAAM,OAAO,EAAE;AACxE,WAAO;AAAA,EACR;AAEA,SAAO,OAAO,UAAU;AACzB;AAEA,IAAI,QAAQ,SAAS,QAAQ;AAC5B,UAAQ,KAAK,KAAK,QAAQ,IAAI,CAAC;AAChC;","names":[]}

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, OcrConfig, ChunkingConfig, ImageExtractionConfig, PdfConfig, KeywordConfig, LanguageDetectionConfig, ErrorClassification } from './types.mjs';
-export { ArchiveMetadata, Chunk, ChunkMetadata, EmailMetadata, ErrorMetadata, ExcelMetadata, ExtractedImage, HtmlConversionOptions, HtmlMetadata, HtmlPreprocessingOptions, ImageMetadata, ImagePreprocessingMetadata, KeywordAlgorithm, Metadata, OcrMetadata, PageBoundary, PageConfig, PageContent, PageInfo, PageStructure, PageUnitType, 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';
 
 /**
@@ -610,72 +610,12 @@ declare function unregisterDocumentExtractor(name: string): void;
  */
 declare function clearDocumentExtractors(): void;
 /**
- * Builder class for creating ExtractionConfig objects with a fluent API.
- *
- * Provides a convenient way to build extraction configurations using method chaining.
- *
- * @example
- * ```typescript
- * import { ExtractionConfig, extractFile } from '@kreuzberg/node';
- *
- * // Create with builder pattern
- * const config = ExtractionConfig.default()
- *   .withChunking({ maxChars: 2048 })
- *   .withOcr({ backend: 'tesseract', language: 'eng' })
- *   .build();
- *
- * const result = await extractFile('document.pdf', null, config);
- * ```
- */
-declare class ExtractionConfigBuilder {
-    private config;
-    /**
-     * Create a new builder with default configuration.
-     */
-    static default(): ExtractionConfigBuilder;
-    /**
-     * Set OCR configuration.
-     */
-    withOcr(ocr: OcrConfig): ExtractionConfigBuilder;
-    /**
-     * Set chunking configuration.
-     */
-    withChunking(chunking: ChunkingConfig): ExtractionConfigBuilder;
-    /**
-     * Set image extraction configuration.
-     */
-    withImageExtraction(images: ImageExtractionConfig): ExtractionConfigBuilder;
-    /**
-     * Set PDF configuration.
-     */
-    withPdf(pdf: PdfConfig): ExtractionConfigBuilder;
-    /**
-     * Set keyword extraction configuration.
-     */
-    withKeywords(keywords: KeywordConfig): ExtractionConfigBuilder;
-    /**
-     * Set language detection configuration.
-     */
-    withLanguageDetection(languageDetection: LanguageDetectionConfig): ExtractionConfigBuilder;
-    /**
-     * Set whether to enable metadata extraction.
-     */
-    withMetadataExtraction(enabled: boolean): ExtractionConfigBuilder;
-    /**
-     * Set whether to enable quality mode.
-     */
-    withQualityMode(enabled: boolean): ExtractionConfigBuilder;
-    /**
-     * Build and return the final ExtractionConfig object.
-     */
-    build(): ExtractionConfig$1;
-}
-/**
- * ExtractionConfig namespace with static methods for loading configuration from files
- * and creating new configurations with the builder pattern.
+ * ExtractionConfig namespace with static methods for loading configuration from files.
  *
  * Provides factory methods to load extraction configuration from TOML, YAML, or JSON files,
- * or to create configurations using a fluent builder API.
+ * or to discover configuration files in the current directory tree.
+ *
+ * For creating configurations programmatically, use plain TypeScript objects instead:
  *
  * @example
  * ```typescript
@@ -684,35 +624,17 @@ declare class ExtractionConfigBuilder {
  * // Load configuration from file
  * const config1 = ExtractionConfig.fromFile('config.toml');
  *
- * // Create with builder pattern
- * const config2 = ExtractionConfig.default()
- *   .withChunking({ maxChars: 2048 })
- *   .build();
+ * // 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, config2);
  * ```
  */
 declare const ExtractionConfig: {
-    /**
-     * Create a default extraction configuration using the builder pattern.
-     *
-     * Returns a builder object that allows you to configure extraction settings
-     * using method chaining.
-     *
-     * @returns ExtractionConfigBuilder for chaining configuration calls
-     *
-     * @example
-     * ```typescript
-     * import { ExtractionConfig } from '@kreuzberg/node';
-     *
-     * const config = ExtractionConfig.default()
-     *   .withChunking({ maxChars: 2048 })
-     *   .withOcr({ backend: 'tesseract', language: 'eng' })
-     *   .build();
-     * ```
-     */
-    default(): ExtractionConfigBuilder;
     /**
      * Load extraction configuration from a file.
      *
@@ -1060,6 +982,151 @@ declare function getErrorCodeDescription(code: number): string;
  * ```
  */
 declare function classifyError(errorMessage: string): ErrorClassification;
-declare const __version__ = "4.0.0-rc.21";
+/**
+ * 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-rc.24";
 
-export { ChunkingConfig, type EmbeddingPreset, ErrorClassification, ExtractionConfig, ExtractionResult, ImageExtractionConfig, KeywordConfig, LanguageDetectionConfig, OcrBackendProtocol, OcrConfig, PanicContext, PdfConfig, PostProcessorProtocol, ValidatorProtocol, __resetBindingForTests, __setBindingForTests, __version__, batchExtractBytes, batchExtractBytesSync, batchExtractFiles, batchExtractFilesSync, classifyError, clearDocumentExtractors, clearOcrBackends, clearPostProcessors, clearValidators, detectMimeType, detectMimeTypeFromPath, extractBytes, extractBytesSync, extractFile, extractFileSync, getEmbeddingPreset, getErrorCodeDescription, getErrorCodeName, 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 };