kreuzberg 3.14.1__py3-none-any.whl → 3.16.0__py3-none-any.whl

kreuzberg/extraction.py

@@ -3,7 +3,7 @@ from __future__ import annotations
 import multiprocessing as mp
 from concurrent.futures import ThreadPoolExecutor, as_completed
 from pathlib import Path
-from typing import TYPE_CHECKING, Any, Final, cast
+from typing import TYPE_CHECKING, Final, cast
 
 import anyio
 
@@ -30,6 +30,31 @@ if TYPE_CHECKING:
 DEFAULT_CONFIG: Final[ExtractionConfig] = ExtractionConfig()
 
 
+async def _handle_cache_async(path: Path, config: ExtractionConfig) -> ExtractionResult | None:
+    """Handle cache lookup and coordination with other processes.
+
+    Args:
+        path: Path to the file being processed
+        config: Extraction configuration
+
+    Returns:
+        Cached result if available, None otherwise
+    """
+    cache = get_document_cache()
+
+    cached_result = cache.get(path, config)
+    if cached_result is not None:
+        return cached_result
+
+    if cache.is_processing(path, config):
+        event = cache.mark_processing(path, config)  # pragma: no cover
+        await anyio.to_thread.run_sync(event.wait)  # pragma: no cover
+
+        return cache.get(path, config)  # pragma: no cover
+
+    return None
+
+
 def _validate_and_post_process_helper(
     result: ExtractionResult, config: ExtractionConfig, file_path: Path | None = None
 ) -> ExtractionResult:
@@ -102,9 +127,9 @@ def _handle_chunk_content(
     mime_type: str,
     config: ExtractionConfig,
     content: str,
-) -> Any:
+) -> list[str]:
     chunker = get_chunker(mime_type=mime_type, max_characters=config.max_chars, overlap_characters=config.max_overlap)
-    return chunker.chunks(content)
+    return list(chunker.chunks(content))
 
 
 async def extract_bytes(content: bytes, mime_type: str, config: ExtractionConfig = DEFAULT_CONFIG) -> ExtractionResult:
@@ -153,19 +178,9 @@ async def extract_file(
     path = Path(file_path)
 
     if config.use_cache:
-        cached_result = cache.get(path, config)
+        cached_result = await _handle_cache_async(path, config)
         if cached_result is not None:
             return cached_result
-
-        if cache.is_processing(path, config):
-            event = cache.mark_processing(path, config)
-            await anyio.to_thread.run_sync(event.wait)  # pragma: no cover
-
-            # Try cache again after waiting for other process to complete  # ~keep
-            cached_result = cache.get(path, config)  # pragma: no cover
-            if cached_result is not None:  # pragma: no cover
-                return cached_result
-
         cache.mark_processing(path, config)
 
     try:
@@ -227,11 +242,11 @@ async def batch_extract_file(
                 error_result = ExtractionResult(
                     content=f"Error: {type(e).__name__}: {e!s}",
                     mime_type="text/plain",
-                    metadata={  # type: ignore[typeddict-unknown-key]
+                    metadata={
                         "error": f"{type(e).__name__}: {e!s}",
                         "error_context": create_error_context(
                             operation="batch_extract_file",
-                            file_path=path,
+                            file_path=str(path),
                             error=e,
                             index=index,
                         ),
@@ -276,7 +291,7 @@ async def batch_extract_bytes(
                 error_result = ExtractionResult(
                     content=f"Error: {type(e).__name__}: {e!s}",
                     mime_type="text/plain",
-                    metadata={  # type: ignore[typeddict-unknown-key]
+                    metadata={
                         "error": f"{type(e).__name__}: {e!s}",
                         "error_context": create_error_context(
                             operation="batch_extract_bytes",
@@ -347,7 +362,7 @@ def extract_file_sync(
             return cached_result
 
         if cache.is_processing(path, config):
-            event = cache.mark_processing(path, config)
+            event = cache.mark_processing(path, config)  # pragma: no cover
             event.wait()  # pragma: no cover
 
             # Try cache again after waiting for other process to complete  # ~keep
@@ -400,31 +415,31 @@ def batch_extract_file_sync(
 
     max_workers = min(len(file_paths), mp.cpu_count())
 
-    def extract_single(file_path: PathLike[str] | str) -> tuple[int, ExtractionResult]:
+    def extract_single(index: int, file_path: PathLike[str] | str) -> tuple[int, ExtractionResult]:
         """Extract single file with index for ordering."""
         try:
             return (
-                file_paths.index(file_path),
+                index,
                 extract_file_sync(file_path=Path(file_path), mime_type=None, config=config),
             )
         except Exception as e:  # noqa: BLE001
             error_result = ExtractionResult(
                 content=f"Error: {type(e).__name__}: {e!s}",
                 mime_type="text/plain",
-                metadata={  # type: ignore[typeddict-unknown-key]
+                metadata={
                     "error": f"{type(e).__name__}: {e!s}",
                     "error_context": create_error_context(
                         operation="batch_extract_file_sync",
-                        file_path=file_path,
+                        file_path=str(file_path),
                         error=e,
                     ),
                 },
                 chunks=[],
             )
-            return (file_paths.index(file_path), error_result)
+            return (index, error_result)
 
     with ThreadPoolExecutor(max_workers=max_workers) as executor:
-        future_to_index = {executor.submit(extract_single, fp): i for i, fp in enumerate(file_paths)}
+        future_to_index = {executor.submit(extract_single, i, fp): i for i, fp in enumerate(file_paths)}
 
         results: list[ExtractionResult | None] = [None] * len(file_paths)
         for future in as_completed(future_to_index):
@@ -453,16 +468,15 @@ def batch_extract_bytes_sync(
 
     max_workers = min(len(contents), mp.cpu_count())
 
-    def extract_single(index_and_content: tuple[int, tuple[bytes, str]]) -> tuple[int, ExtractionResult]:
+    def extract_single(index: int, content: bytes, mime_type: str) -> tuple[int, ExtractionResult]:
         """Extract single content with index for ordering."""
-        index, (content, mime_type) = index_and_content
         try:
             return (index, extract_bytes_sync(content=content, mime_type=mime_type, config=config))
         except Exception as e:  # noqa: BLE001
             error_result = ExtractionResult(
                 content=f"Error: {type(e).__name__}: {e!s}",
                 mime_type="text/plain",
-                metadata={  # type: ignore[typeddict-unknown-key]
+                metadata={
                     "error": f"{type(e).__name__}: {e!s}",
                     "error_context": create_error_context(
                         operation="batch_extract_bytes_sync",
@@ -477,7 +491,9 @@ def batch_extract_bytes_sync(
             return (index, error_result)
 
     with ThreadPoolExecutor(max_workers=max_workers) as executor:
-        future_to_index = {executor.submit(extract_single, (i, content)): i for i, content in enumerate(contents)}
+        future_to_index = {
+            executor.submit(extract_single, i, content, mime_type): i for i, (content, mime_type) in enumerate(contents)
+        }
 
         results: list[ExtractionResult | None] = [None] * len(contents)
         for future in as_completed(future_to_index):

{kreuzberg-3.14.1.dist-info → kreuzberg-3.16.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kreuzberg
-Version: 3.14.1
+Version: 3.16.0
 Summary: Document intelligence framework for Python - Extract text, metadata, and structured data from diverse file formats
 Project-URL: documentation, https://kreuzberg.dev
 Project-URL: homepage, https://github.com/Goldziher/kreuzberg
@@ -31,7 +31,7 @@ Requires-Python: >=3.10
 Requires-Dist: anyio>=4.10.0
 Requires-Dist: chardetng-py>=0.3.5
 Requires-Dist: exceptiongroup>=1.2.2; python_version < '3.11'
-Requires-Dist: html-to-markdown[lxml]>=1.11.0
+Requires-Dist: html-to-markdown[lxml]>=1.13.0
 Requires-Dist: mcp>=1.14.0
 Requires-Dist: msgspec>=0.18.0
 Requires-Dist: numpy>=2.0.0
@@ -107,8 +107,9 @@ Description-Content-Type: text/markdown
 ### Document Intelligence Capabilities
 
 - **Text Extraction**: High-fidelity text extraction preserving document structure and formatting
+- **Image Extraction**: Extract embedded images from PDFs, presentations, HTML, and Office documents with optional OCR
 - **Metadata Extraction**: Comprehensive metadata including author, creation date, language, and document properties
-- **Format Support**: 18 document types including PDF, Microsoft Office, images, HTML, and structured data formats
+- **Format Support**: 21 document types including PDF, Microsoft Office, images, HTML, and structured data formats
 - **OCR Integration**: Tesseract OCR with markdown output (default) and table extraction from scanned documents
 - **Document Classification**: Automatic document type detection (contracts, forms, invoices, receipts, reports)
 
@@ -226,14 +227,15 @@ claude mcp add kreuzberg uvx kreuzberg-mcp
 
 ## Supported Formats
 
-| Category          | Formats                        |
-| ----------------- | ------------------------------ |
-| **Documents**     | PDF, DOCX, DOC, RTF, TXT, EPUB |
-| **Images**        | JPG, PNG, TIFF, BMP, GIF, WEBP |
-| **Spreadsheets**  | XLSX, XLS, CSV, ODS            |
-| **Presentations** | PPTX, PPT, ODP                 |
-| **Web**           | HTML, XML, MHTML               |
-| **Archives**      | Support via extraction         |
+| Category            | Formats                        |
+| ------------------- | ------------------------------ |
+| **Documents**       | PDF, DOCX, DOC, RTF, TXT, EPUB |
+| **Images**          | JPG, PNG, TIFF, BMP, GIF, WEBP |
+| **Spreadsheets**    | XLSX, XLS, CSV, ODS            |
+| **Presentations**   | PPTX, PPT, ODP                 |
+| **Web**             | HTML, XML, MHTML               |
+| **Structured Data** | JSON, YAML, TOML               |
+| **Archives**        | Support via extraction         |
 
 ## 📊 Performance Characteristics
 

kreuzberg-3.16.0.dist-info/RECORD

@@ -0,0 +1,61 @@
+kreuzberg/__init__.py,sha256=EE6ENEjyKlt0o6QN1cG3Z_1isCtminVOjQT7ii5eBHA,1575
+kreuzberg/__main__.py,sha256=3cIDdzTggj2kj8uKx4WShWHmCWqdZazdM3BxUGbAuSI,104
+kreuzberg/_chunker.py,sha256=lRXvVN60vmWaTxa1b3QzvE-jBmOqYzh5dY-3Kl6pSqI,1427
+kreuzberg/_config.py,sha256=H4jUAL0fNY-YE61GbGq5UtAUtXHbZA4-9W3YwcT_hu8,12988
+kreuzberg/_constants.py,sha256=gY6SpCi9za59ghRuLX_z7xfSok6qqvPbvEnv4BLczqI,265
+kreuzberg/_document_classification.py,sha256=55aDxDIJ65qK6yEXt-fRYTn8LgALvYsWssjWSheVpR0,5697
+kreuzberg/_entity_extraction.py,sha256=YvcELIo3kV8A_WbzwNjhKn7rPhkZXjbpNMgm2UK0oJw,3621
+kreuzberg/_gmft.py,sha256=XI8vdBG0tdEVwFiabVieCuvxM5esqTSiFtsEwJ0YT5g,20787
+kreuzberg/_language_detection.py,sha256=T9p6aimB7QFXAQiEntIMZeH_Z62E52E8fBQ43hWuyhs,1960
+kreuzberg/_mime_types.py,sha256=-05mBS5AoF4LUmfB_WyLoce0y4peiOyOf2JucF714WQ,8602
+kreuzberg/_playa.py,sha256=p4G5ymSSCbQoDeXJjH-yuVzdd4y-wKcolqDthjPtqok,11413
+kreuzberg/_registry.py,sha256=8XYT-vPhNYMAbB5RBIUKz-1Zdg48OCnBcdVZzBq6YwY,3307
+kreuzberg/_types.py,sha256=Xht1_TcvsbIpdmLYMy6Pa_HpbQuF9MBOo-BrKkZ7cLA,47358
+kreuzberg/cli.py,sha256=OoHA5MiIcRBATFJpb-FZYlZfpohxL2AbVgamyhnEMFo,14342
+kreuzberg/exceptions.py,sha256=PTiAZgQwcG9hXbgYg2W7sfxksFhq5_wzOFgZGnTJAoc,2991
+kreuzberg/extraction.py,sha256=5TuuRqLRmboLaTS0x9eZ2lrYOHKJBSHuTT_U-5nn6ek,17829
+kreuzberg/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+kreuzberg/_api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+kreuzberg/_api/_config_cache.py,sha256=gX_ezGNq6SCpTn02yFkn24zMVrQwfIk8-u5XkKJiHFg,8774
+kreuzberg/_api/main.py,sha256=_tBZaRiq7qq7x4nXkVRgU5FBivLFJ_dmadAc7aT0H_k,13901
+kreuzberg/_extractors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+kreuzberg/_extractors/_base.py,sha256=39E7R7hV6C2uMJdQKLBVSWK3tN-mtK0LaayU10-8Fqo,11191
+kreuzberg/_extractors/_email.py,sha256=DzNthVbmbdlajDUfs0nNwbHNvG0CAQVqJsRfsatHtf0,8799
+kreuzberg/_extractors/_html.py,sha256=zZ9WZmmoIG9B5dGF25ulm_GmW9RsYFI1HddDUUp3hOE,6351
+kreuzberg/_extractors/_image.py,sha256=7rKEGhUAmdzO0YcBKQVhVme4PqyKIi2UCn4esmmFXOY,4300
+kreuzberg/_extractors/_pandoc.py,sha256=CPEJxKTZdfyb7jPacZkiAsR2NEGL6KyiHzOr88tprJY,24142
+kreuzberg/_extractors/_pdf.py,sha256=78gPO7m8nPFIOskqqRpUfyOhKUk6f5rjJ0cZDnL9Vdk,23224
+kreuzberg/_extractors/_presentation.py,sha256=2g6PJnpgUpUfMjQJh-7_gHywDulE8QE8ypH__BrEUTQ,10692
+kreuzberg/_extractors/_spread_sheet.py,sha256=TJOM70DLN0HzcOkAowZJogAx7QFrouohvU5V0OIliag,12738
+kreuzberg/_extractors/_structured.py,sha256=YkTOfSQJOe127ZURrAYAomNrIkKoAYC4gt0P9ypY3RY,8919
+kreuzberg/_mcp/__init__.py,sha256=h6DgLFO4TMUk7_wCJ2jn2Y6IkFmfzb-Z7jX-G5UCYVc,43
+kreuzberg/_mcp/server.py,sha256=vJWCXbBiv0ktIPZeLedSWZEwKF46p6642H6lxhTnjek,16723
+kreuzberg/_ocr/__init__.py,sha256=grshVFwVQl2rMvH1hg1JNlYXjy5-Tdb_rusLD1Cselk,706
+kreuzberg/_ocr/_base.py,sha256=ZvOJvW8DtylQJZdCPk9vlVNZiBFK-dC4Oj7Kb6-mWkY,1419
+kreuzberg/_ocr/_easyocr.py,sha256=7bkMM_zN0h7ZiX0-VHxxnwNOhQloI-dlOOibpRc-vNs,15710
+kreuzberg/_ocr/_paddleocr.py,sha256=XyYc3gtmnvOGfQ0qBQYFphJa1kSv5hZ_LJ0weD2hQ08,15006
+kreuzberg/_ocr/_table_extractor.py,sha256=LhBiCX8R_xR-uK1FH3ONA_vqOmqUWANZJ2HMCBLsmNY,5513
+kreuzberg/_ocr/_tesseract.py,sha256=BjTKE6ilUpSEKarHdgP3PbsE6I89JeqgDtpQ-XHniBA,51452
+kreuzberg/_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+kreuzberg/_utils/_cache.py,sha256=AtANbs1MWR4WLB2MhatVGhlh7kM-yjSfFuDnSVSNp50,14110
+kreuzberg/_utils/_device.py,sha256=o03rLiHiRX6TKhJ55LO1Vj2Map1Po5YdjuMdA63tGOE,8249
+kreuzberg/_utils/_document_cache.py,sha256=tfk9_Yc1cQkT5_uM5R1uaI4w-2SjNn7QyAd6AmWkSz8,4851
+kreuzberg/_utils/_errors.py,sha256=aQYEnp8oJ-WJVmCNo7YY-25y1KZZFEwjAmxVRfw4a_M,4920
+kreuzberg/_utils/_html_streaming.py,sha256=ywQgEQfEGm6MSotS1g_HXgl0e7V59yLmf2wytALuZko,648
+kreuzberg/_utils/_image_preprocessing.py,sha256=arl4UDDiD_Z6SKM-jTXENaOaaHZBVFTsueb6DcpFXOo,10934
+kreuzberg/_utils/_ocr_cache.py,sha256=uCCZfdY7EiqMhCnhNwqirFOr-Wfaobd2Ntc-F07TKec,3425
+kreuzberg/_utils/_pdf_lock.py,sha256=Ytvds30aZf3yXeZFo27ZenrhUoU-GZlR2rKEkhJ_wlk,1349
+kreuzberg/_utils/_process_pool.py,sha256=fqlxNsxDoqS28BLrZeDBH743HdaUBuGPYFH5hjSajIg,7493
+kreuzberg/_utils/_quality.py,sha256=FCVh9KieWUYgT1klLxudbslzKuqbOTBbTsHbvIuru7M,5510
+kreuzberg/_utils/_ref.py,sha256=iOflvjTUc_F0XaL28Bd6fpvL6qkeoURGA4B77Nqky7I,840
+kreuzberg/_utils/_resource_managers.py,sha256=N3-VeHDj6sKBeg3UL-PqRtKGExUBoVcEB5UuQ8FncY8,2079
+kreuzberg/_utils/_serialization.py,sha256=G-kxtCPDPGFqBMyHfzvAPo-bNUmPdaXYdeg1dnBLfN4,1789
+kreuzberg/_utils/_string.py,sha256=wVyvEHByHBeu_6evmqJGv9Ml-NAwkyz60n8l-7L5Cw0,4366
+kreuzberg/_utils/_sync.py,sha256=gb828WYfVtkB4wKslJrPMmrdeI1h3htWceq-gywHtO4,3184
+kreuzberg/_utils/_table.py,sha256=OVg6T2QnerMhVNb1juLTBSIjyjFiE5-OrUWr5NSCgnQ,6493
+kreuzberg/_utils/_tmp.py,sha256=mwZ0BFzhGPfYa2tt8qSjUjfcHnSYvbQT4VlPRCRc_q8,2038
+kreuzberg-3.16.0.dist-info/METADATA,sha256=d1sUA7WBl0VcXHX0jPGzTHeXmj7yyJzTWjzHUmT-Dp4,12319
+kreuzberg-3.16.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+kreuzberg-3.16.0.dist-info/entry_points.txt,sha256=GplGhFryCP7kyAG_k-Mdahznvo2fwi73qLFg5yQfH_A,91
+kreuzberg-3.16.0.dist-info/licenses/LICENSE,sha256=-8caMvpCK8SgZ5LlRKhGCMtYDEXqTKH9X8pFEhl91_4,1066
+kreuzberg-3.16.0.dist-info/RECORD,,

kreuzberg-3.14.1.dist-info/RECORD

@@ -1,58 +0,0 @@
-kreuzberg/__init__.py,sha256=Oh_NTp8wf0BlvD8CSBad2A493nEWH4jTE0x8v7v1Y9w,1341
-kreuzberg/__main__.py,sha256=3cIDdzTggj2kj8uKx4WShWHmCWqdZazdM3BxUGbAuSI,104
-kreuzberg/_chunker.py,sha256=tr9_KUYTSLauFois3MsB-A-0hGcTT8hTQFrqNRTii-I,1373
-kreuzberg/_config.py,sha256=2LI5z9gXniqO4afrMmbZfMdhlT2701O5OlGKkrMo-bM,12385
-kreuzberg/_constants.py,sha256=Bxc8oiN-wHwnWXT9bEiJhTUcu1ygPpra5qHirAif3b4,191
-kreuzberg/_document_classification.py,sha256=Mz_s2GJGsEl7MQ-67BPoGYCZibTy9Sw0PScUZKBjKOA,5736
-kreuzberg/_entity_extraction.py,sha256=5YpPnqoJ5aiHd_sy4bN4-Ngiq79RhCV6yaUQE8joGXo,3503
-kreuzberg/_gmft.py,sha256=a7KDXbZM0PxyFpAIjM0xMRvxzoMo4fTQuGlFNa8uXBU,20502
-kreuzberg/_language_detection.py,sha256=T9p6aimB7QFXAQiEntIMZeH_Z62E52E8fBQ43hWuyhs,1960
-kreuzberg/_mime_types.py,sha256=kGBDSMO4XPgzUKC7iaBeChCtRQXZ9_zXq6eJydejX_k,7739
-kreuzberg/_playa.py,sha256=p4G5ymSSCbQoDeXJjH-yuVzdd4y-wKcolqDthjPtqok,11413
-kreuzberg/_registry.py,sha256=8XYT-vPhNYMAbB5RBIUKz-1Zdg48OCnBcdVZzBq6YwY,3307
-kreuzberg/_types.py,sha256=BEMTnA8fvHL0dDCnjq7g9Jjd2Ze8NFq988YkMH4zQ9g,39163
-kreuzberg/cli.py,sha256=Ob0IfqWcaiM09pFdC6wTpdSeql0SGZDxBxfrEhJAGmo,13501
-kreuzberg/exceptions.py,sha256=PTiAZgQwcG9hXbgYg2W7sfxksFhq5_wzOFgZGnTJAoc,2991
-kreuzberg/extraction.py,sha256=qT-Ziw5FmMqcPT88VrglikL1RASSJCf5W7xP6L9Vi5s,17673
-kreuzberg/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg/_api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg/_api/main.py,sha256=8g_8j8Dp2e70_yYYUUrJNC5Ku9fuyNgyjUuIgJTRUW8,12500
-kreuzberg/_extractors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg/_extractors/_base.py,sha256=i2FvAhRnamEtBb4a-C7pfcdWIXnkEBw0saMQu7h1_RQ,2069
-kreuzberg/_extractors/_email.py,sha256=jn_8J4BASKJ7zFHBG0PgxNe3OT4pjmEM2tTKX8y_0AE,5887
-kreuzberg/_extractors/_html.py,sha256=NyQKChNLvaSUC_5x1qTYlIQGwL4lEbgUF7BgH9ejEVY,1583
-kreuzberg/_extractors/_image.py,sha256=lFPoxAf7_Zbx-1t8W4vU2bhHauiNGOAFbZxr_2gNUsw,3991
-kreuzberg/_extractors/_pandoc.py,sha256=-Ai4S1cXs7F6yeonb_7Y7_ZoWHn29E2oP1WlPtM-4HM,22505
-kreuzberg/_extractors/_pdf.py,sha256=naJ_AgtAgtGIjAqiU4_G7lgftKWhUjZDLVILSG2AyVc,18757
-kreuzberg/_extractors/_presentation.py,sha256=ULGkt7dzeA9sYSEhpAucKZmkdv9EubzeZtOjoLP3Z2E,6994
-kreuzberg/_extractors/_spread_sheet.py,sha256=eBAx_OwoyRqMzmD4Z07UlOBwcXckymgvj_0o7di6thA,12715
-kreuzberg/_extractors/_structured.py,sha256=PpefI_GDrdLyUgnElrbdB-MeTMKVWium4Ckxm5Zg100,5536
-kreuzberg/_mcp/__init__.py,sha256=h6DgLFO4TMUk7_wCJ2jn2Y6IkFmfzb-Z7jX-G5UCYVc,43
-kreuzberg/_mcp/server.py,sha256=YPMJp6xnZ3DC32NEdX5Gqf3vwxsHZxXxUxZ6jghpv6I,5688
-kreuzberg/_ocr/__init__.py,sha256=grshVFwVQl2rMvH1hg1JNlYXjy5-Tdb_rusLD1Cselk,706
-kreuzberg/_ocr/_base.py,sha256=ZvOJvW8DtylQJZdCPk9vlVNZiBFK-dC4Oj7Kb6-mWkY,1419
-kreuzberg/_ocr/_easyocr.py,sha256=XbgpGt5tkE4xHleIGvV1cHlpOQTp43rSXBO1CyIyKTg,14599
-kreuzberg/_ocr/_paddleocr.py,sha256=hfc6Zi2eSUYTVVF9y9D1P2_pLiLXPfFRoJ6QDJ6oZag,15017
-kreuzberg/_ocr/_table_extractor.py,sha256=LhBiCX8R_xR-uK1FH3ONA_vqOmqUWANZJ2HMCBLsmNY,5513
-kreuzberg/_ocr/_tesseract.py,sha256=QEKK_PDZnNiZRgpklOgMXB-cObJy6C-HuxL6Gza5Z3c,49136
-kreuzberg/_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg/_utils/_cache.py,sha256=qeyI6rJOQlKtdHjJeOjUxx31eItak_drrNn8Cf8HbN8,13956
-kreuzberg/_utils/_device.py,sha256=UxGkSTN3Up-Zn43CSyvf8CozW2xAF05Cm01LWA2FZmg,8263
-kreuzberg/_utils/_document_cache.py,sha256=tfk9_Yc1cQkT5_uM5R1uaI4w-2SjNn7QyAd6AmWkSz8,4851
-kreuzberg/_utils/_errors.py,sha256=aQYEnp8oJ-WJVmCNo7YY-25y1KZZFEwjAmxVRfw4a_M,4920
-kreuzberg/_utils/_image_preprocessing.py,sha256=2u0A28M07F9XlYebTG5salOUVEE3YT3m8fiR8Z2ZM8E,12326
-kreuzberg/_utils/_ocr_cache.py,sha256=uCCZfdY7EiqMhCnhNwqirFOr-Wfaobd2Ntc-F07TKec,3425
-kreuzberg/_utils/_pdf_lock.py,sha256=Ytvds30aZf3yXeZFo27ZenrhUoU-GZlR2rKEkhJ_wlk,1349
-kreuzberg/_utils/_process_pool.py,sha256=7p8Co1w-Tvh2MUdxMcPMpvOikumrb0nN2ApQVytV-_c,6726
-kreuzberg/_utils/_quality.py,sha256=f7NbyZysyJQD8jKCNWhogvluU9A7GdEYhMsDBeMbGAA,5412
-kreuzberg/_utils/_ref.py,sha256=iOflvjTUc_F0XaL28Bd6fpvL6qkeoURGA4B77Nqky7I,840
-kreuzberg/_utils/_serialization.py,sha256=97iIgdcxdbym-BEvy0J6HAduBCUXyCGwhuEHCT_l7I4,1513
-kreuzberg/_utils/_string.py,sha256=wVyvEHByHBeu_6evmqJGv9Ml-NAwkyz60n8l-7L5Cw0,4366
-kreuzberg/_utils/_sync.py,sha256=OWiciXPTGHIxgiGoHI2AglZ1siTNT-nU_JCgHPNzzHk,2196
-kreuzberg/_utils/_table.py,sha256=OVg6T2QnerMhVNb1juLTBSIjyjFiE5-OrUWr5NSCgnQ,6493
-kreuzberg/_utils/_tmp.py,sha256=wnOInBkcuQoxI1vBLvNv9NqbRCEu9Y03qfOjqQuAk3s,841
-kreuzberg-3.14.1.dist-info/METADATA,sha256=4sG9L9AtvBHFxjv84obrcaYNToc_sO0-AHnnpo1-ZGY,12127
-kreuzberg-3.14.1.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-kreuzberg-3.14.1.dist-info/entry_points.txt,sha256=GplGhFryCP7kyAG_k-Mdahznvo2fwi73qLFg5yQfH_A,91
-kreuzberg-3.14.1.dist-info/licenses/LICENSE,sha256=-8caMvpCK8SgZ5LlRKhGCMtYDEXqTKH9X8pFEhl91_4,1066
-kreuzberg-3.14.1.dist-info/RECORD,,