@kreuzberg/wasm 4.4.5 → 4.5.0

package/dist/pkg/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=v4.4.5" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.5.0" 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#">
@@ -33,6 +33,9 @@
   <a href="https://rubygems.org/gems/kreuzberg">
     <img src="https://img.shields.io/gem/v/kreuzberg?label=Ruby&color=007ec6" alt="Ruby">
   </a>
+  <a href="https://kreuzberg-dev.r-universe.dev/kreuzberg">
+    <img src="https://img.shields.io/badge/R-kreuzberg-007ec6" alt="R">
+  </a>
   <a href="https://github.com/kreuzberg-dev/kreuzberg/pkgs/container/kreuzberg">
     <img src="https://img.shields.io/badge/Docker-007ec6?logo=docker&logoColor=white" alt="Docker">
   </a>
@@ -44,6 +47,9 @@
   <a href="https://docs.kreuzberg.dev">
     <img src="https://img.shields.io/badge/docs-kreuzberg.dev-blue" alt="Documentation">
   </a>
+  <a href="https://huggingface.co/Kreuzberg">
+    <img src="https://img.shields.io/badge/%F0%9F%A4%97%20Hugging%20Face-Models-yellow" alt="Hugging Face">
+  </a>
 </div>
 
 <img width="1128" height="191" alt="Banner2" src="https://github.com/user-attachments/assets/419fc06c-8313-4324-b159-4b4d3cfce5c0" />
@@ -55,9 +61,7 @@
 </div>
 
 
-Extract text, tables, images, and metadata from 75+ file formats including PDF, Office documents, and images. WebAssembly bindings for browsers, Deno, and Cloudflare Workers with portable deployment and multi-threading support.
-
-> **Full Feature Parity** — The WASM package supports all extraction capabilities at full parity with native bindings: PDF (via PDFium), Excel/spreadsheets (via Calamine), archives (ZIP, TAR, 7z, GZIP), and OCR (via built-in Tesseract-WASM). No external dependencies required.
+Extract text, tables, images, and metadata from 88+ file formats including PDF, Office documents, and images. WebAssembly bindings for browsers, Deno, and Cloudflare Workers with portable deployment and multi-threading support.
 
 
 ## Installation
@@ -97,7 +101,7 @@ yarn add @kreuzberg/wasm
 ### System Requirements
 
 - Modern browser with WebAssembly support, or Deno 1.0+, or Cloudflare Workers
-- OCR is built-in via Tesseract-WASM (enable at runtime with `enableOcr()`)
+- Optional: [Tesseract WASM](https://github.com/naptha/tesseract.js) for OCR functionality
 
 
 
@@ -176,40 +180,6 @@ extractWithOcr().catch(console.error);
 See [Table Extraction Guide](https://kreuzberg.dev/features/table-extraction/) for detailed examples.
 
 
-#### Excel/Spreadsheet Extraction
-
-Extract structured data from Excel files directly in the browser or server-side runtimes:
-
-```ts
-import { extractBytes, initWasm } from "@kreuzberg/wasm";
-
-async function extractSpreadsheet() {
-	await initWasm();
-
-	const bytes = new Uint8Array(
-		await fetch("report.xlsx").then((r) => r.arrayBuffer()),
-	);
-
-	const result = await extractBytes(
-		bytes,
-		"application/vnd.openxmlformats-officedocument.spreadsheetml.sheet",
-	);
-
-	console.log("Spreadsheet content:");
-	console.log(result.content);
-
-	if (result.tables && result.tables.length > 0) {
-		result.tables.forEach((table, index) => {
-			console.log(`\nSheet ${index + 1}:`);
-			console.log(table.markdown);
-		});
-	}
-}
-
-extractSpreadsheet().catch(console.error);
-```
-
-
 
 #### Processing Multiple Files
 
@@ -301,19 +271,21 @@ extractDocuments(fileBytes, mimes)
 
 ## Features
 
-### Supported File Formats (75+)
+### Supported File Formats (88+)
 
-75+ file formats across 8 major categories with intelligent format detection and comprehensive metadata extraction.
+88+ file formats across 8 major categories with intelligent format detection and comprehensive metadata extraction.
 
 #### Office Documents
 
 | Category | Formats | Capabilities |
 |----------|---------|--------------|
-| **Word Processing** | `.docx`, `.odt` | Full text, tables, images, metadata, styles |
-| **Spreadsheets** | `.xlsx`, `.xlsm`, `.xlsb`, `.xls`, `.xla`, `.xlam`, `.xltm`, `.ods` | Sheet data, formulas, cell metadata, charts |
-| **Presentations** | `.pptx`, `.ppt`, `.ppsx` | Slides, speaker notes, images, metadata |
+| **Word Processing** | `.docx`, `.docm`, `.dotx`, `.dotm`, `.dot`, `.odt` | Full text, tables, images, metadata, styles |
+| **Spreadsheets** | `.xlsx`, `.xlsm`, `.xlsb`, `.xls`, `.xla`, `.xlam`, `.xltm`, `.xltx`, `.xlt`, `.ods` | Sheet data, formulas, cell metadata, charts |
+| **Presentations** | `.pptx`, `.pptm`, `.ppsx`, `.potx`, `.potm`, `.pot`, `.ppt` | Slides, speaker notes, images, metadata |
 | **PDF** | `.pdf` | Text, tables, images, metadata, OCR support |
 | **eBooks** | `.epub`, `.fb2` | Chapters, metadata, embedded resources |
+| **Database** | `.dbf` | Table data extraction, field type support |
+| **Hangul** | `.hwp`, `.hwpx` | Korean document format, text extraction |
 
 #### Images (OCR-Enabled)
 
@@ -354,10 +326,14 @@ extractDocuments(fileBytes, mimes)
 - **Metadata Extraction** - Retrieve document properties, creation date, author, etc.
 - **Table Extraction** - Parse tables with structure and cell content preservation
 - **Image Extraction** - Extract embedded images and render page previews
-- **OCR Support** - Built-in Tesseract-WASM for scanned documents and images
-- **Full Feature Parity** - All extraction capabilities at parity with native bindings: PDF, Excel, archives, OCR, and 75+ formats
+- **OCR Support** - Integrate multiple OCR backends for scanned documents
+
 - **Async/Await** - Non-blocking document processing with concurrent operations
+
+
 - **Plugin System** - Extensible post-processing for custom text transformation
+
+
 - **Batch Processing** - Efficiently process multiple documents in parallel
 - **Memory Efficient** - Stream large files without loading entirely into memory
 - **Language Detection** - Detect and support multiple languages in documents

package/dist/pkg/kreuzberg_wasm.d.ts

@@ -29,6 +29,7 @@ export class ModuleInfo {
  * * `dataList: Uint8Array[]` - Array of document bytes
  * * `mimeTypes: string[]` - Array of MIME types (must match dataList length)
  * * `config?: object` - Optional extraction configuration (applied to all)
+ * * `fileConfigs?: (object | null)[]` - Optional per-file config overrides (must match dataList length if provided)
  *
  * # Returns
  *
@@ -36,7 +37,8 @@ export class ModuleInfo {
  *
  * # Throws
  *
- * Rejects if dataList and mimeTypes lengths don't match.
+ * Rejects if dataList and mimeTypes lengths don't match, or if fileConfigs
+ * is provided and its length doesn't match dataList.
  *
  * # Example
  *
@@ -57,21 +59,26 @@ export class ModuleInfo {
  *   ['application/pdf', 'application/vnd.openxmlformats-officedocument.wordprocessingml.document'],
  *   null
  * );
+ *
+ * // With per-file configs:
+ * const fileConfigs = [{ ocrConfig: { language: 'eng' } }, null];
+ * const results2 = await batchExtractBytes(buffers, mimeTypes, null, fileConfigs);
  * ```
  */
-export function batchExtractBytes(data_list: Uint8Array[], mime_types: string[], config?: any | null): Promise<any>;
+export function batchExtractBytes(data_list: Uint8Array[], mime_types: string[], config?: any | null, file_configs?: any[] | null): Promise<any>;
 
 /**
  * Batch extract from multiple byte arrays (synchronous).
  *
  * Processes multiple document byte arrays in parallel. All documents use the
- * same extraction configuration.
+ * same extraction configuration unless per-file configs are provided.
  *
  * # JavaScript Parameters
  *
  * * `dataList: Uint8Array[]` - Array of document bytes
  * * `mimeTypes: string[]` - Array of MIME types (must match dataList length)
  * * `config?: object` - Optional extraction configuration (applied to all)
+ * * `fileConfigs?: (object | null)[]` - Optional per-file config overrides (must match dataList length if provided)
  *
  * # Returns
  *
@@ -79,7 +86,8 @@ export function batchExtractBytes(data_list: Uint8Array[], mime_types: string[],
  *
  * # Throws
  *
- * Throws if dataList and mimeTypes lengths don't match.
+ * Throws if dataList and mimeTypes lengths don't match, or if fileConfigs
+ * is provided and its length doesn't match dataList.
  *
  * # Example
  *
@@ -93,9 +101,13 @@ export function batchExtractBytes(data_list: Uint8Array[], mime_types: string[],
  * results.forEach((result, i) => {
  *   console.log(`Document ${i}: ${result.content.substring(0, 50)}...`);
  * });
+ *
+ * // With per-file configs:
+ * const fileConfigs = [{ ocrConfig: { language: 'eng' } }, null, null];
+ * const results2 = batchExtractBytesSync(buffers, mimeTypes, null, fileConfigs);
  * ```
  */
-export function batchExtractBytesSync(data_list: Uint8Array[], mime_types: string[], config?: any | null): any;
+export function batchExtractBytesSync(data_list: Uint8Array[], mime_types: string[], config?: any | null, file_configs?: any[] | null): any;
 
 /**
  * Batch extract from multiple Files or Blobs (asynchronous).

package/dist/pkg/kreuzberg_wasm.js

@@ -55,6 +55,7 @@ if (Symbol.dispose) ModuleInfo.prototype[Symbol.dispose] = ModuleInfo.prototype.
  * * `dataList: Uint8Array[]` - Array of document bytes
  * * `mimeTypes: string[]` - Array of MIME types (must match dataList length)
  * * `config?: object` - Optional extraction configuration (applied to all)
+ * * `fileConfigs?: (object | null)[]` - Optional per-file config overrides (must match dataList length if provided)
  *
  * # Returns
  *
@@ -62,7 +63,8 @@ if (Symbol.dispose) ModuleInfo.prototype[Symbol.dispose] = ModuleInfo.prototype.
  *
  * # Throws
  *
- * Rejects if dataList and mimeTypes lengths don't match.
+ * Rejects if dataList and mimeTypes lengths don't match, or if fileConfigs
+ * is provided and its length doesn't match dataList.
  *
  * # Example
  *
@@ -83,18 +85,25 @@ if (Symbol.dispose) ModuleInfo.prototype[Symbol.dispose] = ModuleInfo.prototype.
  *   ['application/pdf', 'application/vnd.openxmlformats-officedocument.wordprocessingml.document'],
  *   null
  * );
+ *
+ * // With per-file configs:
+ * const fileConfigs = [{ ocrConfig: { language: 'eng' } }, null];
+ * const results2 = await batchExtractBytes(buffers, mimeTypes, null, fileConfigs);
  * ```
  * @param {Uint8Array[]} data_list
  * @param {string[]} mime_types
  * @param {any | null} [config]
+ * @param {any[] | null} [file_configs]
  * @returns {Promise<any>}
  */
-export function batchExtractBytes(data_list, mime_types, config) {
+export function batchExtractBytes(data_list, mime_types, config, file_configs) {
     const ptr0 = passArrayJsValueToWasm0(data_list, wasm.__wbindgen_malloc_command_export);
     const len0 = WASM_VECTOR_LEN;
     const ptr1 = passArrayJsValueToWasm0(mime_types, wasm.__wbindgen_malloc_command_export);
     const len1 = WASM_VECTOR_LEN;
-    const ret = wasm.batchExtractBytes(ptr0, len0, ptr1, len1, isLikeNone(config) ? 0 : addToExternrefTable0(config));
+    var ptr2 = isLikeNone(file_configs) ? 0 : passArrayJsValueToWasm0(file_configs, wasm.__wbindgen_malloc_command_export);
+    var len2 = WASM_VECTOR_LEN;
+    const ret = wasm.batchExtractBytes(ptr0, len0, ptr1, len1, isLikeNone(config) ? 0 : addToExternrefTable0(config), ptr2, len2);
     return ret;
 }
 
@@ -102,13 +111,14 @@ export function batchExtractBytes(data_list, mime_types, config) {
  * Batch extract from multiple byte arrays (synchronous).
  *
  * Processes multiple document byte arrays in parallel. All documents use the
- * same extraction configuration.
+ * same extraction configuration unless per-file configs are provided.
  *
  * # JavaScript Parameters
  *
  * * `dataList: Uint8Array[]` - Array of document bytes
  * * `mimeTypes: string[]` - Array of MIME types (must match dataList length)
  * * `config?: object` - Optional extraction configuration (applied to all)
+ * * `fileConfigs?: (object | null)[]` - Optional per-file config overrides (must match dataList length if provided)
  *
  * # Returns
  *
@@ -116,7 +126,8 @@ export function batchExtractBytes(data_list, mime_types, config) {
  *
  * # Throws
  *
- * Throws if dataList and mimeTypes lengths don't match.
+ * Throws if dataList and mimeTypes lengths don't match, or if fileConfigs
+ * is provided and its length doesn't match dataList.
  *
  * # Example
  *
@@ -130,18 +141,25 @@ export function batchExtractBytes(data_list, mime_types, config) {
  * results.forEach((result, i) => {
  *   console.log(`Document ${i}: ${result.content.substring(0, 50)}...`);
  * });
+ *
+ * // With per-file configs:
+ * const fileConfigs = [{ ocrConfig: { language: 'eng' } }, null, null];
+ * const results2 = batchExtractBytesSync(buffers, mimeTypes, null, fileConfigs);
  * ```
  * @param {Uint8Array[]} data_list
  * @param {string[]} mime_types
  * @param {any | null} [config]
+ * @param {any[] | null} [file_configs]
  * @returns {any}
  */
-export function batchExtractBytesSync(data_list, mime_types, config) {
+export function batchExtractBytesSync(data_list, mime_types, config, file_configs) {
     const ptr0 = passArrayJsValueToWasm0(data_list, wasm.__wbindgen_malloc_command_export);
     const len0 = WASM_VECTOR_LEN;
     const ptr1 = passArrayJsValueToWasm0(mime_types, wasm.__wbindgen_malloc_command_export);
     const len1 = WASM_VECTOR_LEN;
-    const ret = wasm.batchExtractBytesSync(ptr0, len0, ptr1, len1, isLikeNone(config) ? 0 : addToExternrefTable0(config));
+    var ptr2 = isLikeNone(file_configs) ? 0 : passArrayJsValueToWasm0(file_configs, wasm.__wbindgen_malloc_command_export);
+    var len2 = WASM_VECTOR_LEN;
+    const ret = wasm.batchExtractBytesSync(ptr0, len0, ptr1, len1, isLikeNone(config) ? 0 : addToExternrefTable0(config), ptr2, len2);
     if (ret[2]) {
         throw takeFromExternrefTable0(ret[1]);
     }
@@ -1345,6 +1363,10 @@ function __wbg_get_imports() {
             const ret = typeof(arg0) === 'function';
             return ret;
         },
+        __wbg___wbindgen_is_null_0b605fc6b167c56f: function(arg0) {
+            const ret = arg0 === null;
+            return ret;
+        },
         __wbg___wbindgen_is_object_781bc9f159099513: function(arg0) {
             const val = arg0;
             const ret = typeof(val) === 'object' && val !== null;
@@ -1448,6 +1470,9 @@ function __wbg_get_imports() {
         __wbg_getRandomValues_3f44b700395062e5: function() { return handleError(function (arg0, arg1) {
             globalThis.crypto.getRandomValues(getArrayU8FromWasm0(arg0, arg1));
         }, arguments); },
+        __wbg_getRandomValues_76dfc69825c9c552: function() { return handleError(function (arg0, arg1) {
+            globalThis.crypto.getRandomValues(getArrayU8FromWasm0(arg0, arg1));
+        }, arguments); },
         __wbg_getTime_1dad7b5386ddd2d9: function(arg0) {
             const ret = arg0.getTime();
             return ret;
@@ -1575,7 +1600,7 @@ function __wbg_get_imports() {
                     const a = state0.a;
                     state0.a = 0;
                     try {
-                        return wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_(a, state0.b, arg0, arg1);
+                        return wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2(a, state0.b, arg0, arg1);
                     } finally {
                         state0.a = a;
                     }
@@ -1597,7 +1622,7 @@ function __wbg_get_imports() {
                     const a = state0.a;
                     state0.a = 0;
                     try {
-                        return wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_(a, state0.b, arg0, arg1);
+                        return wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2(a, state0.b, arg0, arg1);
                     } finally {
                         state0.a = a;
                     }
@@ -1760,13 +1785,13 @@ function __wbg_get_imports() {
             console.warn(arg0);
         },
         __wbindgen_cast_0000000000000001: function(arg0, arg1) {
-            // Cast intrinsic for `Closure(Closure { dtor_idx: 21, function: Function { arguments: [Externref], shim_idx: 23, ret: Unit, inner_ret: Some(Unit) }, mutable: true }) -> Externref`.
-            const ret = makeMutClosure(arg0, arg1, wasm.wasm_bindgen_179aebea8749b817___closure__destroy___dyn_core_dde6c4b55a98adc4___ops__function__FnMut__wasm_bindgen_179aebea8749b817___JsValue____Output_______, wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue______true_);
+            // Cast intrinsic for `Closure(Closure { dtor_idx: 3637, function: Function { arguments: [Externref], shim_idx: 3638, ret: Result(Unit), inner_ret: Some(Result(Unit)) }, mutable: true }) -> Externref`.
+            const ret = makeMutClosure(arg0, arg1, wasm.wasm_bindgen__closure__destroy__hcbea9f36b367d3d8, wasm_bindgen__convert__closures_____invoke__h240eec0b57535315);
             return ret;
         },
         __wbindgen_cast_0000000000000002: function(arg0, arg1) {
-            // Cast intrinsic for `Closure(Closure { dtor_idx: 3543, function: Function { arguments: [Externref], shim_idx: 3544, ret: Result(Unit), inner_ret: Some(Result(Unit)) }, mutable: true }) -> Externref`.
-            const ret = makeMutClosure(arg0, arg1, wasm.wasm_bindgen_179aebea8749b817___closure__destroy___dyn_core_dde6c4b55a98adc4___ops__function__FnMut__wasm_bindgen_179aebea8749b817___JsValue____Output___core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___, wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue__core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___true_);
+            // Cast intrinsic for `Closure(Closure { dtor_idx: 4, function: Function { arguments: [Externref], shim_idx: 5, ret: Unit, inner_ret: Some(Unit) }, mutable: true }) -> Externref`.
+            const ret = makeMutClosure(arg0, arg1, wasm.wasm_bindgen__closure__destroy__h60b03db66f836c8f, wasm_bindgen__convert__closures_____invoke__h4c966767d2820559);
             return ret;
         },
         __wbindgen_cast_0000000000000003: function(arg0) {
@@ -1808,19 +1833,19 @@ function __wbg_get_imports() {
     };
 }
 
-function wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue______true_(arg0, arg1, arg2) {
-    wasm.wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue______true_(arg0, arg1, arg2);
+function wasm_bindgen__convert__closures_____invoke__h4c966767d2820559(arg0, arg1, arg2) {
+    wasm.wasm_bindgen__convert__closures_____invoke__h4c966767d2820559(arg0, arg1, arg2);
 }
 
-function wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue__core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___true_(arg0, arg1, arg2) {
-    const ret = wasm.wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue__core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___true_(arg0, arg1, arg2);
+function wasm_bindgen__convert__closures_____invoke__h240eec0b57535315(arg0, arg1, arg2) {
+    const ret = wasm.wasm_bindgen__convert__closures_____invoke__h240eec0b57535315(arg0, arg1, arg2);
     if (ret[1]) {
         throw takeFromExternrefTable0(ret[0]);
     }
 }
 
-function wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_(arg0, arg1, arg2, arg3) {
-    wasm.wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_(arg0, arg1, arg2, arg3);
+function wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2(arg0, arg1, arg2, arg3) {
+    wasm.wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2(arg0, arg1, arg2, arg3);
 }
 
 const ModuleInfoFinalization = (typeof FinalizationRegistry === 'undefined')

package/dist/pkg/kreuzberg_wasm_bg.js

@@ -53,6 +53,7 @@ if (Symbol.dispose) ModuleInfo.prototype[Symbol.dispose] = ModuleInfo.prototype.
  * * `dataList: Uint8Array[]` - Array of document bytes
  * * `mimeTypes: string[]` - Array of MIME types (must match dataList length)
  * * `config?: object` - Optional extraction configuration (applied to all)
+ * * `fileConfigs?: (object | null)[]` - Optional per-file config overrides (must match dataList length if provided)
  *
  * # Returns
  *
@@ -60,7 +61,8 @@ if (Symbol.dispose) ModuleInfo.prototype[Symbol.dispose] = ModuleInfo.prototype.
  *
  * # Throws
  *
- * Rejects if dataList and mimeTypes lengths don't match.
+ * Rejects if dataList and mimeTypes lengths don't match, or if fileConfigs
+ * is provided and its length doesn't match dataList.
  *
  * # Example
  *
@@ -81,18 +83,25 @@ if (Symbol.dispose) ModuleInfo.prototype[Symbol.dispose] = ModuleInfo.prototype.
  *   ['application/pdf', 'application/vnd.openxmlformats-officedocument.wordprocessingml.document'],
  *   null
  * );
+ *
+ * // With per-file configs:
+ * const fileConfigs = [{ ocrConfig: { language: 'eng' } }, null];
+ * const results2 = await batchExtractBytes(buffers, mimeTypes, null, fileConfigs);
  * ```
  * @param {Uint8Array[]} data_list
  * @param {string[]} mime_types
  * @param {any | null} [config]
+ * @param {any[] | null} [file_configs]
  * @returns {Promise<any>}
  */
-export function batchExtractBytes(data_list, mime_types, config) {
+export function batchExtractBytes(data_list, mime_types, config, file_configs) {
     const ptr0 = passArrayJsValueToWasm0(data_list, wasm.__wbindgen_malloc_command_export);
     const len0 = WASM_VECTOR_LEN;
     const ptr1 = passArrayJsValueToWasm0(mime_types, wasm.__wbindgen_malloc_command_export);
     const len1 = WASM_VECTOR_LEN;
-    const ret = wasm.batchExtractBytes(ptr0, len0, ptr1, len1, isLikeNone(config) ? 0 : addToExternrefTable0(config));
+    var ptr2 = isLikeNone(file_configs) ? 0 : passArrayJsValueToWasm0(file_configs, wasm.__wbindgen_malloc_command_export);
+    var len2 = WASM_VECTOR_LEN;
+    const ret = wasm.batchExtractBytes(ptr0, len0, ptr1, len1, isLikeNone(config) ? 0 : addToExternrefTable0(config), ptr2, len2);
     return ret;
 }
 
@@ -100,13 +109,14 @@ export function batchExtractBytes(data_list, mime_types, config) {
  * Batch extract from multiple byte arrays (synchronous).
  *
  * Processes multiple document byte arrays in parallel. All documents use the
- * same extraction configuration.
+ * same extraction configuration unless per-file configs are provided.
  *
  * # JavaScript Parameters
  *
  * * `dataList: Uint8Array[]` - Array of document bytes
  * * `mimeTypes: string[]` - Array of MIME types (must match dataList length)
  * * `config?: object` - Optional extraction configuration (applied to all)
+ * * `fileConfigs?: (object | null)[]` - Optional per-file config overrides (must match dataList length if provided)
  *
  * # Returns
  *
@@ -114,7 +124,8 @@ export function batchExtractBytes(data_list, mime_types, config) {
  *
  * # Throws
  *
- * Throws if dataList and mimeTypes lengths don't match.
+ * Throws if dataList and mimeTypes lengths don't match, or if fileConfigs
+ * is provided and its length doesn't match dataList.
  *
  * # Example
  *
@@ -128,18 +139,25 @@ export function batchExtractBytes(data_list, mime_types, config) {
  * results.forEach((result, i) => {
  *   console.log(`Document ${i}: ${result.content.substring(0, 50)}...`);
  * });
+ *
+ * // With per-file configs:
+ * const fileConfigs = [{ ocrConfig: { language: 'eng' } }, null, null];
+ * const results2 = batchExtractBytesSync(buffers, mimeTypes, null, fileConfigs);
  * ```
  * @param {Uint8Array[]} data_list
  * @param {string[]} mime_types
  * @param {any | null} [config]
+ * @param {any[] | null} [file_configs]
  * @returns {any}
  */
-export function batchExtractBytesSync(data_list, mime_types, config) {
+export function batchExtractBytesSync(data_list, mime_types, config, file_configs) {
     const ptr0 = passArrayJsValueToWasm0(data_list, wasm.__wbindgen_malloc_command_export);
     const len0 = WASM_VECTOR_LEN;
     const ptr1 = passArrayJsValueToWasm0(mime_types, wasm.__wbindgen_malloc_command_export);
     const len1 = WASM_VECTOR_LEN;
-    const ret = wasm.batchExtractBytesSync(ptr0, len0, ptr1, len1, isLikeNone(config) ? 0 : addToExternrefTable0(config));
+    var ptr2 = isLikeNone(file_configs) ? 0 : passArrayJsValueToWasm0(file_configs, wasm.__wbindgen_malloc_command_export);
+    var len2 = WASM_VECTOR_LEN;
+    const ret = wasm.batchExtractBytesSync(ptr0, len0, ptr1, len1, isLikeNone(config) ? 0 : addToExternrefTable0(config), ptr2, len2);
     if (ret[2]) {
         throw takeFromExternrefTable0(ret[1]);
     }
@@ -1251,6 +1269,10 @@ export function __wbg___wbindgen_is_function_3c846841762788c1(arg0) {
     const ret = typeof(arg0) === 'function';
     return ret;
 }
+export function __wbg___wbindgen_is_null_0b605fc6b167c56f(arg0) {
+    const ret = arg0 === null;
+    return ret;
+}
 export function __wbg___wbindgen_is_object_781bc9f159099513(arg0) {
     const val = arg0;
     const ret = typeof(val) === 'object' && val !== null;
@@ -1354,6 +1376,9 @@ export function __wbg_from_4bdf88943703fd48(arg0) {
 export function __wbg_getRandomValues_3f44b700395062e5() { return handleError(function (arg0, arg1) {
     globalThis.crypto.getRandomValues(getArrayU8FromWasm0(arg0, arg1));
 }, arguments); }
+export function __wbg_getRandomValues_76dfc69825c9c552() { return handleError(function (arg0, arg1) {
+    globalThis.crypto.getRandomValues(getArrayU8FromWasm0(arg0, arg1));
+}, arguments); }
 export function __wbg_getTime_1dad7b5386ddd2d9(arg0) {
     const ret = arg0.getTime();
     return ret;
@@ -1481,7 +1506,7 @@ export function __wbg_new_d098e265629cd10f(arg0, arg1) {
             const a = state0.a;
             state0.a = 0;
             try {
-                return wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_(a, state0.b, arg0, arg1);
+                return wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2(a, state0.b, arg0, arg1);
             } finally {
                 state0.a = a;
             }
@@ -1503,7 +1528,7 @@ export function __wbg_new_typed_aaaeaf29cf802876(arg0, arg1) {
             const a = state0.a;
             state0.a = 0;
             try {
-                return wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_(a, state0.b, arg0, arg1);
+                return wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2(a, state0.b, arg0, arg1);
             } finally {
                 state0.a = a;
             }
@@ -1666,13 +1691,13 @@ export function __wbg_warn_69424c2d92a2fa73(arg0) {
     console.warn(arg0);
 }
 export function __wbindgen_cast_0000000000000001(arg0, arg1) {
-    // Cast intrinsic for `Closure(Closure { dtor_idx: 21, function: Function { arguments: [Externref], shim_idx: 23, ret: Unit, inner_ret: Some(Unit) }, mutable: true }) -> Externref`.
-    const ret = makeMutClosure(arg0, arg1, wasm.wasm_bindgen_179aebea8749b817___closure__destroy___dyn_core_dde6c4b55a98adc4___ops__function__FnMut__wasm_bindgen_179aebea8749b817___JsValue____Output_______, wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue______true_);
+    // Cast intrinsic for `Closure(Closure { dtor_idx: 3637, function: Function { arguments: [Externref], shim_idx: 3638, ret: Result(Unit), inner_ret: Some(Result(Unit)) }, mutable: true }) -> Externref`.
+    const ret = makeMutClosure(arg0, arg1, wasm.wasm_bindgen__closure__destroy__hcbea9f36b367d3d8, wasm_bindgen__convert__closures_____invoke__h240eec0b57535315);
     return ret;
 }
 export function __wbindgen_cast_0000000000000002(arg0, arg1) {
-    // Cast intrinsic for `Closure(Closure { dtor_idx: 3543, function: Function { arguments: [Externref], shim_idx: 3544, ret: Result(Unit), inner_ret: Some(Result(Unit)) }, mutable: true }) -> Externref`.
-    const ret = makeMutClosure(arg0, arg1, wasm.wasm_bindgen_179aebea8749b817___closure__destroy___dyn_core_dde6c4b55a98adc4___ops__function__FnMut__wasm_bindgen_179aebea8749b817___JsValue____Output___core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___, wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue__core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___true_);
+    // Cast intrinsic for `Closure(Closure { dtor_idx: 4, function: Function { arguments: [Externref], shim_idx: 5, ret: Unit, inner_ret: Some(Unit) }, mutable: true }) -> Externref`.
+    const ret = makeMutClosure(arg0, arg1, wasm.wasm_bindgen__closure__destroy__h60b03db66f836c8f, wasm_bindgen__convert__closures_____invoke__h4c966767d2820559);
     return ret;
 }
 export function __wbindgen_cast_0000000000000003(arg0) {
@@ -1705,19 +1730,19 @@ export function __wbindgen_init_externref_table() {
     table.set(offset + 2, true);
     table.set(offset + 3, false);
 }
-function wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue______true_(arg0, arg1, arg2) {
-    wasm.wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue______true_(arg0, arg1, arg2);
+function wasm_bindgen__convert__closures_____invoke__h4c966767d2820559(arg0, arg1, arg2) {
+    wasm.wasm_bindgen__convert__closures_____invoke__h4c966767d2820559(arg0, arg1, arg2);
 }
 
-function wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue__core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___true_(arg0, arg1, arg2) {
-    const ret = wasm.wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue__core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___true_(arg0, arg1, arg2);
+function wasm_bindgen__convert__closures_____invoke__h240eec0b57535315(arg0, arg1, arg2) {
+    const ret = wasm.wasm_bindgen__convert__closures_____invoke__h240eec0b57535315(arg0, arg1, arg2);
     if (ret[1]) {
         throw takeFromExternrefTable0(ret[0]);
     }
 }
 
-function wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_(arg0, arg1, arg2, arg3) {
-    wasm.wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_(arg0, arg1, arg2, arg3);
+function wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2(arg0, arg1, arg2, arg3) {
+    wasm.wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2(arg0, arg1, arg2, arg3);
 }
 
 const ModuleInfoFinalization = (typeof FinalizationRegistry === 'undefined')

package/dist/pkg/kreuzberg_wasm_bg.wasm.d.ts

@@ -2,8 +2,8 @@
 /* eslint-disable */
 export const memory: WebAssembly.Memory;
 export const __wbg_moduleinfo_free: (a: number, b: number) => void;
-export const batchExtractBytes: (a: number, b: number, c: number, d: number, e: number) => any;
-export const batchExtractBytesSync: (a: number, b: number, c: number, d: number, e: number) => [number, number, number];
+export const batchExtractBytes: (a: number, b: number, c: number, d: number, e: number, f: number, g: number) => any;
+export const batchExtractBytesSync: (a: number, b: number, c: number, d: number, e: number, f: number, g: number) => [number, number, number];
 export const batchExtractFiles: (a: number, b: number, c: number) => any;
 export const batchExtractFilesSync: () => [number, number, number];
 export const clear_ocr_backends: () => [number, number];
@@ -45,11 +45,11 @@ export const read_block_from_callback_wasm: (a: number, b: number, c: number, d:
 export const write_block_from_callback_wasm: (a: number, b: number, c: number) => number;
 export const compress: (a: number, b: number, c: number, d: number) => [number, number, number];
 export const decompress: (a: any, b: number, c: number, d: any) => [number, number];
-export const wasm_bindgen_179aebea8749b817___closure__destroy___dyn_core_dde6c4b55a98adc4___ops__function__FnMut__wasm_bindgen_179aebea8749b817___JsValue____Output_______: (a: number, b: number) => void;
-export const wasm_bindgen_179aebea8749b817___closure__destroy___dyn_core_dde6c4b55a98adc4___ops__function__FnMut__wasm_bindgen_179aebea8749b817___JsValue____Output___core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___: (a: number, b: number) => void;
-export const wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue__core_dde6c4b55a98adc4___result__Result_____wasm_bindgen_179aebea8749b817___JsError___true_: (a: number, b: number, c: any) => [number, number];
-export const wasm_bindgen_179aebea8749b817___convert__closures_____invoke___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined___js_sys_3900621f03eac4bc___Function_fn_wasm_bindgen_179aebea8749b817___JsValue_____wasm_bindgen_179aebea8749b817___sys__Undefined_______true_: (a: number, b: number, c: any, d: any) => void;
-export const wasm_bindgen_179aebea8749b817___convert__closures_____invoke___wasm_bindgen_179aebea8749b817___JsValue______true_: (a: number, b: number, c: any) => void;
+export const wasm_bindgen__closure__destroy__hcbea9f36b367d3d8: (a: number, b: number) => void;
+export const wasm_bindgen__closure__destroy__h60b03db66f836c8f: (a: number, b: number) => void;
+export const wasm_bindgen__convert__closures_____invoke__h240eec0b57535315: (a: number, b: number, c: any) => [number, number];
+export const wasm_bindgen__convert__closures_____invoke__h8e9fc28c4e841be2: (a: number, b: number, c: any, d: any) => void;
+export const wasm_bindgen__convert__closures_____invoke__h4c966767d2820559: (a: number, b: number, c: any) => void;
 export const __wbindgen_externrefs: WebAssembly.Table;
 export const __wbindgen_malloc_command_export: (a: number, b: number) => number;
 export const __wbindgen_realloc_command_export: (a: number, b: number, c: number, d: number) => number;