@kreuzberg/node 4.2.3 → 4.2.5

package/README.md

@@ -22,7 +22,7 @@
     <img src="https://img.shields.io/maven-central/v/dev.kreuzberg/kreuzberg?label=Java&color=007ec6" alt="Java">
   </a>
   <a href="https://github.com/kreuzberg-dev/kreuzberg/releases">
-    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.2.3" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.2.5" alt="Go">
   </a>
   <a href="https://www.nuget.org/packages/Kreuzberg/">
     <img src="https://img.shields.io/nuget/v/Kreuzberg?label=C%23&color=007ec6" alt="C#">

package/dist/index.d.mts

@@ -1200,6 +1200,6 @@ declare function __resetBindingForTests(): void;
  * @module @kreuzberg/node
  */
 
-declare const __version__ = "4.2.3";
+declare const __version__ = "4.2.5";
 
 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, loadConfigFile, loadConfigFromPath, registerOcrBackend, registerPostProcessor, registerValidator, unregisterDocumentExtractor, unregisterOcrBackend, unregisterPostProcessor, unregisterValidator, validateMimeType };

package/dist/index.d.ts

@@ -1200,6 +1200,6 @@ declare function __resetBindingForTests(): void;
  * @module @kreuzberg/node
  */
 
-declare const __version__ = "4.2.3";
+declare const __version__ = "4.2.5";
 
 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, loadConfigFile, loadConfigFromPath, registerOcrBackend, registerPostProcessor, registerValidator, unregisterDocumentExtractor, unregisterOcrBackend, unregisterPostProcessor, unregisterValidator, validateMimeType };

package/dist/index.js

@@ -91,6 +91,12 @@ module.exports = __toCommonJS(index_exports);
 // typescript/core/binding.ts
 var import_node_module = require("module");
 var import_meta = {};
+function isNapiRuntime() {
+  if (typeof process === "undefined") return false;
+  if (process.versions?.["bun"]) return true;
+  if (process.versions?.node) return true;
+  return false;
+}
 var binding = null;
 var bindingInitialized = false;
 function createNativeBindingError(error) {
@@ -167,7 +173,7 @@ function getBinding() {
     return binding;
   }
   try {
-    if (typeof process !== "undefined" && process.versions && process.versions.node) {
+    if (isNapiRuntime()) {
       binding = loadNativeBinding();
       bindingInitialized = true;
       return binding;
@@ -528,6 +534,8 @@ function normalizeExtractionConfig(config) {
   setIfDefined(normalized, "pages", pages);
   const htmlOptions = normalizeHtmlOptions(config.htmlOptions);
   setIfDefined(normalized, "htmlOptions", htmlOptions);
+  setIfDefined(normalized, "outputFormat", config.outputFormat);
+  setIfDefined(normalized, "resultFormat", config.resultFormat);
   return normalized;
 }
 
@@ -594,6 +602,38 @@ function convertChunk(rawChunk) {
     }
   };
 }
+function convertElement(rawElement) {
+  if (!rawElement || typeof rawElement !== "object") {
+    return {
+      elementId: "",
+      elementType: "narrative_text",
+      text: "",
+      metadata: {}
+    };
+  }
+  const element = rawElement;
+  const elementMetadata = element["metadata"] ?? {};
+  return {
+    // biome-ignore lint/complexity/useLiteralKeys: required for strict TypeScript noPropertyAccessFromIndexSignature
+    elementId: element["element_id"] ?? element["elementId"] ?? "",
+    // biome-ignore lint/complexity/useLiteralKeys: required for strict TypeScript noPropertyAccessFromIndexSignature
+    elementType: element["element_type"] ?? element["elementType"] ?? "narrative_text",
+    // biome-ignore lint/complexity/useLiteralKeys: required for strict TypeScript noPropertyAccessFromIndexSignature
+    text: element["text"] ?? "",
+    metadata: {
+      // biome-ignore lint/complexity/useLiteralKeys: required for strict TypeScript noPropertyAccessFromIndexSignature
+      pageNumber: elementMetadata["page_number"] ?? elementMetadata["pageNumber"] ?? null,
+      // biome-ignore lint/complexity/useLiteralKeys: required for strict TypeScript noPropertyAccessFromIndexSignature
+      filename: elementMetadata["filename"] ?? null,
+      // biome-ignore lint/complexity/useLiteralKeys: required for strict TypeScript noPropertyAccessFromIndexSignature
+      coordinates: elementMetadata["coordinates"] ? elementMetadata["coordinates"] : null,
+      // biome-ignore lint/complexity/useLiteralKeys: required for strict TypeScript noPropertyAccessFromIndexSignature
+      elementIndex: elementMetadata["element_index"] ?? elementMetadata["elementIndex"] ?? null,
+      // biome-ignore lint/complexity/useLiteralKeys: required for strict TypeScript noPropertyAccessFromIndexSignature
+      additional: elementMetadata["additional"] ?? {}
+    }
+  };
+}
 function convertImage(rawImage) {
   if (!rawImage || typeof rawImage !== "object") {
     return {
@@ -667,6 +707,7 @@ function convertResult(rawResult) {
       detectedLanguages: null,
       chunks: null,
       images: null,
+      elements: null,
       pages: null
     };
   }
@@ -685,6 +726,7 @@ function convertResult(rawResult) {
     detectedLanguages: Array.isArray(result["detectedLanguages"]) ? result["detectedLanguages"] : null,
     chunks: null,
     images: null,
+    elements: null,
     pages: null
   };
   const chunksData = result["chunks"];
@@ -695,6 +737,10 @@ function convertResult(rawResult) {
   if (Array.isArray(imagesData)) {
     returnObj.images = imagesData.map((image) => convertImage(image));
   }
+  const elementsData = result["elements"];
+  if (Array.isArray(elementsData)) {
+    returnObj.elements = elementsData.map((element) => convertElement(element));
+  }
   const pagesData = result["pages"];
   if (Array.isArray(pagesData)) {
     returnObj.pages = pagesData.map((page) => convertPageContent(page));
@@ -1326,7 +1372,7 @@ function getEmbeddingPreset(name) {
 }
 
 // typescript/index.ts
-var __version__ = "4.2.3";
+var __version__ = "4.2.5";
 // Annotate the CommonJS export names for ESM import in node:
 0 && (module.exports = {
   CacheError,