kreuzberg 3.13.0__py3-none-any.whl → 3.13.2__py3-none-any.whl

kreuzberg/_utils/_string.py

@@ -21,20 +21,10 @@ _encoding_cache: dict[str, str] = {}
 
 @lru_cache(maxsize=128)
 def _get_encoding_cache_key(data_hash: str, size: int) -> str:
-    """Generate cache key for encoding detection."""
     return f"{data_hash}:{size}"
 
 
 def safe_decode(byte_data: bytes, encoding: str | None = None) -> str:
-    """Decode a byte string safely with mojibake detection and correction.
-
-    Args:
-        byte_data: The byte string to decode.
-        encoding: The encoding to use when decoding the byte string.
-
-    Returns:
-        The decoded string with mojibake detection and correction.
-    """
     if not byte_data:
         return ""
 
@@ -88,7 +78,6 @@ def safe_decode(byte_data: bytes, encoding: str | None = None) -> str:
 
 
 def _calculate_text_confidence(text: str) -> float:
-    """Calculate confidence score for decoded text quality."""
     if not text:
         return 0.0
 
@@ -114,7 +103,6 @@ def _calculate_text_confidence(text: str) -> float:
 
 
 def _fix_mojibake(text: str) -> str:
-    """Attempt to fix common mojibake patterns."""
     if not text:
         return text
 
@@ -131,14 +119,6 @@ def _fix_mojibake(text: str) -> str:
 
 
 def normalize_spaces(text: str) -> str:
-    """Normalize spaces while preserving line breaks and paragraph structure.
-
-    Args:
-        text: The text to normalize.
-
-    Returns:
-        The normalized text with proper spacing.
-    """
     if not text or not text.strip():
         return ""
 

kreuzberg/_utils/_sync.py

@@ -18,16 +18,6 @@ P = ParamSpec("P")
 
 
 async def run_sync(sync_fn: Callable[P, T], *args: P.args, **kwargs: P.kwargs) -> T:
-    """Run a synchronous function in an asynchronous context.
-
-    Args:
-        sync_fn: The synchronous function to run.
-        *args: The positional arguments to pass to the function.
-        **kwargs: The keyword arguments to pass to the function.
-
-    Returns:
-        The result of the synchronous function.
-    """
     if kwargs:
         handler = partial(sync_fn, **kwargs)
         return cast("T", await any_io_run_sync(handler, *args, abandon_on_cancel=True))  # pyright: ignore [reportCallIssue]
@@ -35,14 +25,6 @@ async def run_sync(sync_fn: Callable[P, T], *args: P.args, **kwargs: P.kwargs) -
 
 
 async def run_taskgroup(*async_tasks: Awaitable[Any]) -> list[Any]:
-    """Run a list of coroutines concurrently.
-
-    Args:
-        *async_tasks: The list of coroutines to run.
-
-    Returns:
-        The results of the coroutines.
-    """
     results: list[Any] = [None] * len(async_tasks)
 
     async def run_task(index: int, task: Awaitable[T]) -> None:
@@ -56,15 +38,6 @@ async def run_taskgroup(*async_tasks: Awaitable[Any]) -> list[Any]:
 
 
 async def run_taskgroup_batched(*async_tasks: Awaitable[Any], batch_size: int) -> list[Any]:
-    """Run a list of coroutines concurrently in batches.
-
-    Args:
-        *async_tasks: The list of coroutines to run.
-        batch_size: The size of each batch.
-
-    Returns:
-        The results of the coroutines.
-    """
     results: list[Any] = []
 
     for i in range(0, len(async_tasks), batch_size):
@@ -75,25 +48,6 @@ async def run_taskgroup_batched(*async_tasks: Awaitable[Any], batch_size: int) -
 
 
 async def run_maybe_sync(fn: Callable[P, T | Awaitable[T]], *args: P.args, **kwargs: P.kwargs) -> T:
-    """Executes a callable function and handles both synchronous and asynchronous
-    results.
-
-    This function invokes the provided callable `sync_fn` with the given
-    arguments and keyword arguments. If the result of `sync_fn` is awaitable,
-    it awaits the result before returning it. Otherwise, the result is returned
-    directly.
-
-    Args:
-        fn: The callable to be executed. It can produce either a
-            synchronous or asynchronous result.
-        *args: Positional arguments to pass to `sync_fn`.
-        **kwargs: Keyword arguments to pass to `sync_fn`.
-
-    Returns:
-        The result of `sync_fn` invocation. If the result is awaitable, the
-        awaited value is returned. Otherwise, the synchronous result is
-        returned.
-    """
     result = fn(*args, **kwargs)
     if isawaitable(result):
         return cast("T", await result)
@@ -101,40 +55,10 @@ async def run_maybe_sync(fn: Callable[P, T | Awaitable[T]], *args: P.args, **kwa
 
 
 def run_maybe_async(fn: Callable[P, T | Awaitable[T]], *args: P.args, **kwargs: P.kwargs) -> T:
-    """Runs a synchronous or asynchronous function, resolving the output.
-
-    Determines if the provided function is synchronous or asynchronous. If synchronous,
-    executes it directly. If asynchronous, it runs the function within the event loop
-    using anyio. The return value is resolved regardless of the function type.
-
-    Args:
-        fn: The function to be executed, which can
-            either be synchronous or asynchronous.
-        *args: Positional arguments to be passed to the function.
-        **kwargs: Keyword arguments to be passed to the function.
-
-    Returns:
-        T: The return value of the executed function, resolved if asynchronous.
-    """
     return cast("T", fn(*args, **kwargs) if not iscoroutinefunction(fn) else anyio.run(partial(fn, **kwargs), *args))
 
 
 def run_sync_only(fn: Callable[P, T | Awaitable[T]], *args: P.args, **kwargs: P.kwargs) -> T:
-    """Runs a function, but only if it's synchronous. Raises error if async.
-
-    This is used for pure sync code paths where we cannot handle async functions.
-
-    Args:
-        fn: The function to be executed, must be synchronous.
-        *args: Positional arguments to be passed to the function.
-        **kwargs: Keyword arguments to be passed to the function.
-
-    Returns:
-        T: The return value of the executed function.
-
-    Raises:
-        RuntimeError: If the function is asynchronous.
-    """
     if iscoroutinefunction(fn):
         raise RuntimeError(f"Cannot run async function {fn.__name__} in sync-only context")
     return cast("T", fn(*args, **kwargs))

kreuzberg/_utils/_table.py

@@ -8,15 +8,6 @@ if TYPE_CHECKING:
 
 
 def export_table_to_csv(table: TableData, separator: str = ",") -> str:
-    r"""Export a TableData object to CSV/TSV format.
-
-    Args:
-        table: TableData object containing DataFrame
-        separator: Field separator ("," for CSV, "\t" for TSV)
-
-    Returns:
-        String representation in CSV/TSV format
-    """
     if "df" not in table or table["df"] is None:
         return ""
 
@@ -27,26 +18,10 @@ def export_table_to_csv(table: TableData, separator: str = ",") -> str:
 
 
 def export_table_to_tsv(table: TableData) -> str:
-    """Export a TableData object to TSV format.
-
-    Args:
-        table: TableData object containing DataFrame
-
-    Returns:
-        String representation in TSV format
-    """
     return export_table_to_csv(table, separator="\t")
 
 
 def enhance_table_markdown(table: TableData) -> str:
-    """Generate enhanced markdown table with better formatting.
-
-    Args:
-        table: TableData object
-
-    Returns:
-        Enhanced markdown table string
-    """
     if "df" not in table or table["df"] is None:
         return table.get("text", "")
 
@@ -72,7 +47,6 @@ def enhance_table_markdown(table: TableData) -> str:
 
 
 def _generate_separator_row(df: Any) -> str:
-    """Generate separator row with proper alignment hints."""
     separators = []
     for col in df.columns:
         dtype_str = str(df[col].dtype)
@@ -84,7 +58,6 @@ def _generate_separator_row(df: Any) -> str:
 
 
 def _analyze_float_columns(df: Any) -> dict[str, str]:
-    """Analyze float columns to determine formatting strategy."""
     float_col_formatting = {}
     for col in df.columns:
         dtype_str = str(df[col].dtype)
@@ -103,7 +76,6 @@ def _analyze_float_columns(df: Any) -> dict[str, str]:
 
 
 def _format_table_row(row: Any, df: Any, float_col_formatting: dict[str, str]) -> list[str]:
-    """Format a single table row with proper value formatting."""
     formatted_row = []
     for col_name, value in row.items():
         if value is None:
@@ -124,7 +96,6 @@ def _format_table_row(row: Any, df: Any, float_col_formatting: dict[str, str]) -
 
 
 def _is_numeric_column(series: Any) -> bool:
-    """Check if a polars Series contains mostly numeric values."""
     if len(series) == 0:
         return False
 
@@ -161,14 +132,6 @@ def _is_numeric_column(series: Any) -> bool:
 
 
 def generate_table_summary(tables: list[TableData]) -> dict[str, Any]:
-    """Generate summary statistics for extracted tables.
-
-    Args:
-        tables: List of TableData objects
-
-    Returns:
-        Dictionary with table statistics
-    """
     if not tables:
         return {
             "table_count": 0,
@@ -208,14 +171,6 @@ def generate_table_summary(tables: list[TableData]) -> dict[str, Any]:
 
 
 def extract_table_structure_info(table: TableData) -> dict[str, Any]:
-    """Extract structural information from a table.
-
-    Args:
-        table: TableData object
-
-    Returns:
-        Dictionary with structural information
-    """
     info = {
         "has_headers": False,
         "row_count": 0,

kreuzberg/_utils/_tmp.py

@@ -16,15 +16,6 @@ if TYPE_CHECKING:  # pragma: no cover
 async def create_temp_file(
     extension: str, content: bytes | None = None
 ) -> tuple[Path, Callable[[], Coroutine[None, None, None]]]:
-    """Create a temporary file that is closed.
-
-    Args:
-        extension: The file extension.
-        content: The content to write to the file.
-
-    Returns:
-        The temporary file path.
-    """
     file = await run_sync(NamedTemporaryFile, suffix=extension, delete=False)
     if content:
         await AsyncPath(file.name).write_bytes(content)

kreuzberg/cli.py

@@ -265,7 +265,7 @@ def cli(ctx: click.Context) -> None:
 @click.option("--easyocr-languages", help="EasyOCR language codes (comma-separated, e.g., 'en,de')")
 @click.option("--paddleocr-languages", help="PaddleOCR language codes (comma-separated, e.g., 'en,german')")
 @click.pass_context
-def extract(ctx: click.Context) -> None:
+def extract(ctx: click.Context, /, **kwargs: Any) -> None:
     """Extract text from a document.
 
     FILE can be a path to a document or '-' to read from stdin.
@@ -279,7 +279,7 @@ def extract(ctx: click.Context) -> None:
 
         extraction_config = build_extraction_config(file_config, cli_args)
 
-        result = _perform_extraction(params["file"], extraction_config, params["verbose"])
+        result = _perform_extraction(kwargs.get("file"), extraction_config, params["verbose"])
 
         _write_output(result, params["output"], params["show_metadata"], params["output_format"], params["verbose"])
 

{kreuzberg-3.13.0.dist-info → kreuzberg-3.13.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kreuzberg
-Version: 3.13.0
+Version: 3.13.2
 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
@@ -34,11 +34,12 @@ Requires-Dist: exceptiongroup>=1.2.2; python_version < '3.11'
 Requires-Dist: html-to-markdown[lxml]>=1.9.1
 Requires-Dist: mcp>=1.13.0
 Requires-Dist: msgspec>=0.18.0
+Requires-Dist: numpy>=1.24.0
 Requires-Dist: playa-pdf>=0.7.0
 Requires-Dist: polars>=1.33.0
 Requires-Dist: psutil>=7.0.0
 Requires-Dist: pypdfium2==4.30.0
-Requires-Dist: python-calamine>=0.3.2
+Requires-Dist: python-calamine>=0.5.2
 Requires-Dist: python-pptx>=1.0.2
 Requires-Dist: typing-extensions>=4.15.0; python_version < '3.12'
 Provides-Extra: additional-extensions

kreuzberg-3.13.2.dist-info/RECORD

@@ -0,0 +1,57 @@
+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=Q5oiJE1XRf8ITuYcO8LZAOB3G2zNlXz2458rgPSth-U,12257
+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=jKbD7V_KP9XTLjT9SBgSgE3CyDjqbRDm9BAiWV2sAC0,19542
+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=8cPpz3oZVnMwWDT2v_Q7wf-GHd5YuHmc-nkLtvPfE1I,2433
+kreuzberg/_types.py,sha256=D-2d_WG8HyByA163izGhjk7t-e4FL_N-_6UzlVso8Dg,36020
+kreuzberg/cli.py,sha256=nPH4FDW6WkoF4gtH0s4RWmxjAveJ_-Unb6fev6x0Sko,12752
+kreuzberg/exceptions.py,sha256=PTiAZgQwcG9hXbgYg2W7sfxksFhq5_wzOFgZGnTJAoc,2991
+kreuzberg/extraction.py,sha256=jiMKiDyTf3sHyk76sMffHR-eH-_yg-DFRMuXEKufRYI,17649
+kreuzberg/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+kreuzberg/_api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+kreuzberg/_api/main.py,sha256=JALYRD0qwyoZloWk5dNNuslBtG4GlVNc0G2oADm6cAc,7578
+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=UqPoYfvDRX6Rd1yPhcLHJLDw6d2cUzgkqOGjh2eleJM,3301
+kreuzberg/_extractors/_pandoc.py,sha256=-Ai4S1cXs7F6yeonb_7Y7_ZoWHn29E2oP1WlPtM-4HM,22505
+kreuzberg/_extractors/_pdf.py,sha256=Yv_c3xYzrGAjgTbwCGqbiQTDLjIUP_Pu7Z3GmMOqgqg,17865
+kreuzberg/_extractors/_presentation.py,sha256=ULGkt7dzeA9sYSEhpAucKZmkdv9EubzeZtOjoLP3Z2E,6994
+kreuzberg/_extractors/_spread_sheet.py,sha256=x25u2M-ufxpDd7_qrjhMEz1yFftIcOISE1qwPW09Zm0,11962
+kreuzberg/_extractors/_structured.py,sha256=PpefI_GDrdLyUgnElrbdB-MeTMKVWium4Ckxm5Zg100,5536
+kreuzberg/_mcp/__init__.py,sha256=h6DgLFO4TMUk7_wCJ2jn2Y6IkFmfzb-Z7jX-G5UCYVc,43
+kreuzberg/_mcp/server.py,sha256=iYJG6g0u7I6mWtC4R1XlxydBrPpgnp5dGJzpm9QAZig,8438
+kreuzberg/_ocr/__init__.py,sha256=grshVFwVQl2rMvH1hg1JNlYXjy5-Tdb_rusLD1Cselk,706
+kreuzberg/_ocr/_base.py,sha256=5ef2g8JuSaZF2sDiAmoaODHbeG4MT0LtNzbtW0n9BnU,1445
+kreuzberg/_ocr/_easyocr.py,sha256=XbgpGt5tkE4xHleIGvV1cHlpOQTp43rSXBO1CyIyKTg,14599
+kreuzberg/_ocr/_paddleocr.py,sha256=58sKOHfKCHGFJNlRLrJwey8G_7xbsAAPBXB4n3hKc7k,14052
+kreuzberg/_ocr/_table_extractor.py,sha256=LhBiCX8R_xR-uK1FH3ONA_vqOmqUWANZJ2HMCBLsmNY,5513
+kreuzberg/_ocr/_tesseract.py,sha256=xGML3ygY5xMN5T3YznrKDVAH_DWfaFiteFBo_-GpjCs,48931
+kreuzberg/_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+kreuzberg/_utils/_cache.py,sha256=S6Oc4TJamiuuWeJ2ABxDFbbQh4o8w38AUyZeBEc1NN8,12767
+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/_ocr_cache.py,sha256=8_-qmPlK2adQKsH4OO4Mlk8wmqBMl3XxkcV_NsXVyFs,3501
+kreuzberg/_utils/_pdf_lock.py,sha256=Ytvds30aZf3yXeZFo27ZenrhUoU-GZlR2rKEkhJ_wlk,1349
+kreuzberg/_utils/_process_pool.py,sha256=9dPMD_gBocQ5VaeCIrlSJfPXKyXNuyKaATmqOPExxiE,6723
+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=R-6owHjvcvHGhem_vDsFH7S2yMHGoUUO2PFcj-Idptk,6361
+kreuzberg/_utils/_tmp.py,sha256=wnOInBkcuQoxI1vBLvNv9NqbRCEu9Y03qfOjqQuAk3s,841
+kreuzberg-3.13.2.dist-info/METADATA,sha256=c1w8iB_Frnzr0DHY-X-a9rk5S9vQPICPIniPzwfvHV8,12127
+kreuzberg-3.13.2.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
+kreuzberg-3.13.2.dist-info/entry_points.txt,sha256=GplGhFryCP7kyAG_k-Mdahznvo2fwi73qLFg5yQfH_A,91
+kreuzberg-3.13.2.dist-info/licenses/LICENSE,sha256=-8caMvpCK8SgZ5LlRKhGCMtYDEXqTKH9X8pFEhl91_4,1066
+kreuzberg-3.13.2.dist-info/RECORD,,

kreuzberg-3.13.0.dist-info/RECORD

@@ -1,56 +0,0 @@
-kreuzberg/__init__.py,sha256=Oh_NTp8wf0BlvD8CSBad2A493nEWH4jTE0x8v7v1Y9w,1341
-kreuzberg/__main__.py,sha256=3cIDdzTggj2kj8uKx4WShWHmCWqdZazdM3BxUGbAuSI,104
-kreuzberg/_chunker.py,sha256=y4-dX6ILjjBkkC1gkCzXb7v7vbi8844m7vz1gIzbmv4,1952
-kreuzberg/_config.py,sha256=dSTumnpleMeUjUabWgAH7WlhTkdNG3eeMv8FSFmUaEI,15776
-kreuzberg/_constants.py,sha256=Bxc8oiN-wHwnWXT9bEiJhTUcu1ygPpra5qHirAif3b4,191
-kreuzberg/_document_classification.py,sha256=NZ-6tQtVa1OgigC7xf30hAsnL5_gi9ak9X2XYdsCfTI,6361
-kreuzberg/_entity_extraction.py,sha256=QFIPQ_fovEnEezpS6W4pwpjTA2PqS7TUCD9AKf8sAyc,4666
-kreuzberg/_gmft.py,sha256=60WpPTf7jocU-kmkBe-pBytl7l58aQzd-Aw2_Hlioug,21481
-kreuzberg/_language_detection.py,sha256=yLUliJOUyofVma_q6FwzG9Ck4-XX3AEjxleTHrqi8R4,2445
-kreuzberg/_mime_types.py,sha256=fwtPKtp2XhCLT686qF26PBMeOqcVJroKPwkp7JgaM0E,8462
-kreuzberg/_playa.py,sha256=1viLRqgcDWvaPo5ZsDPO2gqHFSBApOYortTV_SPVK9k,12190
-kreuzberg/_registry.py,sha256=wGSlkS0U1zqruWQCLE95vj4a2mw1yyvf0j6rgz80sJg,3473
-kreuzberg/_types.py,sha256=WFUFY1S7SL7kTfHCX-zGASLYT94FxLD71C9vGUzFOiA,38922
-kreuzberg/cli.py,sha256=MLeWoMcLoN6WnkbyRbOY-2dqp-vNZf7Nb-K_R5F5CoU,12730
-kreuzberg/exceptions.py,sha256=PTiAZgQwcG9hXbgYg2W7sfxksFhq5_wzOFgZGnTJAoc,2991
-kreuzberg/extraction.py,sha256=jiMKiDyTf3sHyk76sMffHR-eH-_yg-DFRMuXEKufRYI,17649
-kreuzberg/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg/_api/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg/_api/main.py,sha256=JALYRD0qwyoZloWk5dNNuslBtG4GlVNc0G2oADm6cAc,7578
-kreuzberg/_extractors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg/_extractors/_base.py,sha256=EZTEJzwJxwu_yYFQ5QlZVNQMPCcli7yyUB4T5mFotCY,4209
-kreuzberg/_extractors/_email.py,sha256=mVi_VDmiFhe6NgiWxJDYt4DQiP6jVs5dP8BsPClm3WQ,6108
-kreuzberg/_extractors/_html.py,sha256=NyQKChNLvaSUC_5x1qTYlIQGwL4lEbgUF7BgH9ejEVY,1583
-kreuzberg/_extractors/_image.py,sha256=UZEOmKNAS4KjaX38iYq2Ux6Mta3juCF1MzWNeBxpPE8,3414
-kreuzberg/_extractors/_pandoc.py,sha256=zumwImIXwD3ziPhYxt0EQct5sSMy5lQiY6KnPSDxBTU,24183
-kreuzberg/_extractors/_pdf.py,sha256=766O7rXAeAJ42vPpWbGpW_WgHXm48eWwX09l3aqjKeM,18064
-kreuzberg/_extractors/_presentation.py,sha256=BJdEM9jsuAd0vb-PIRwNMcRj4xVjItb5kpOpnjsCBi0,10175
-kreuzberg/_extractors/_spread_sheet.py,sha256=wqAV-Stqfd4hXs5ock-chqBEdzv4voSgT1uFUO1cIU0,12075
-kreuzberg/_extractors/_structured.py,sha256=PpefI_GDrdLyUgnElrbdB-MeTMKVWium4Ckxm5Zg100,5536
-kreuzberg/_mcp/__init__.py,sha256=h6DgLFO4TMUk7_wCJ2jn2Y6IkFmfzb-Z7jX-G5UCYVc,43
-kreuzberg/_mcp/server.py,sha256=iYJG6g0u7I6mWtC4R1XlxydBrPpgnp5dGJzpm9QAZig,8438
-kreuzberg/_ocr/__init__.py,sha256=grshVFwVQl2rMvH1hg1JNlYXjy5-Tdb_rusLD1Cselk,706
-kreuzberg/_ocr/_base.py,sha256=IkONqwG6zxZoVMni1JlYugBoyONahlRny7J2_7Dy69c,3953
-kreuzberg/_ocr/_easyocr.py,sha256=CtiHGx_BmuUwZhC7bScYF9mwnAxRrLWJ-X70fuwFTjk,14079
-kreuzberg/_ocr/_paddleocr.py,sha256=wCuIQ_yxPWE9hukiehYNRdt00Rb2h6pWdfqPS8hI2s0,14297
-kreuzberg/_ocr/_table_extractor.py,sha256=MeQLQn_bRco5OAcUoy613ZbZLCDBRJY8uHH_bUBSP8I,7613
-kreuzberg/_ocr/_tesseract.py,sha256=i_UTjOmrFxZbtmXxrQIsE78wtZLTyZph0i0jDQc4EMA,56916
-kreuzberg/_utils/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kreuzberg/_utils/_cache.py,sha256=fDqFp_54-Kyn3_4VkXkhovvNIB2osKqXlS13MlORrU8,14539
-kreuzberg/_utils/_device.py,sha256=35xQvrLSPISJlWicQGknoBjkwdalwVxiJbzyxwuwOVo,9747
-kreuzberg/_utils/_document_cache.py,sha256=CpCdJVd8SYLjfwm0ozSM8mx5x8i9vVDet3BlEUpzuZY,6920
-kreuzberg/_utils/_errors.py,sha256=ctD-s1q7vbEgqHQ3OVJiEOODDLTd2LvrM3z6o37zrGI,6395
-kreuzberg/_utils/_pdf_lock.py,sha256=mHB1A4Fo_nSfgdqUNEWODH9b5tNFqpEHcNE6rT41dGE,1886
-kreuzberg/_utils/_process_pool.py,sha256=ebuMPmHXPkWaLWjgAkeaONvAZo974PhfENN8pnPTCco,8415
-kreuzberg/_utils/_quality.py,sha256=m3SIXGDY9pfRmh3XeKdZWT1vBz7issH0SfKsutEuRxw,5833
-kreuzberg/_utils/_ref.py,sha256=uP_S3x0AQH2Nyjo1tYEj7N_u9hGzYVewdjch6a8Fv5I,1458
-kreuzberg/_utils/_serialization.py,sha256=duKP5OuBvi-m6ljQOhoyuJU7sl2WPnov8yJDpYuDArw,2052
-kreuzberg/_utils/_string.py,sha256=yrcwHHl23FxWrNoFXkmR3icgivfvbLRvkqQek8F3qqI,5020
-kreuzberg/_utils/_sync.py,sha256=mc-K2y_sc6mG-HOswlHTXAWaEzgisEERvq9PPw2dAw4,4869
-kreuzberg/_utils/_table.py,sha256=dYM_dWNHRCXcWOhSQBnahOJaBXyuQFyYX9arHrH4TF8,7555
-kreuzberg/_utils/_tmp.py,sha256=hVn-VVijIg2FM7EZJ899gc7wZg-TGoJZoeAcxMX-Cxg,1044
-kreuzberg-3.13.0.dist-info/METADATA,sha256=896BWDLD6ApGiOQFKXMqQezC4qgKRUxjMqbZVWxBoJ0,12098
-kreuzberg-3.13.0.dist-info/WHEEL,sha256=qtCwoSJWgHk21S1Kb4ihdzI2rlJ1ZKaIurTj_ngOhyQ,87
-kreuzberg-3.13.0.dist-info/entry_points.txt,sha256=GplGhFryCP7kyAG_k-Mdahznvo2fwi73qLFg5yQfH_A,91
-kreuzberg-3.13.0.dist-info/licenses/LICENSE,sha256=-8caMvpCK8SgZ5LlRKhGCMtYDEXqTKH9X8pFEhl91_4,1066
-kreuzberg-3.13.0.dist-info/RECORD,,