kreuzberg 3.11.0__tar.gz → 3.11.2__tar.gz

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/.github/workflows/ci.yaml

@@ -15,7 +15,7 @@ jobs:
     timeout-minutes: 10
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
 
       - name: Install uv
         uses: astral-sh/setup-uv@v6
@@ -58,7 +58,7 @@ jobs:
     timeout-minutes: 20
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
 
       - name: Install uv
         uses: astral-sh/setup-uv@v6
@@ -151,7 +151,7 @@ jobs:
     timeout-minutes: 30
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
 
       - name: Install uv
         uses: astral-sh/setup-uv@v6

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/.github/workflows/docs.yml

@@ -24,7 +24,7 @@ jobs:
     runs-on: ubuntu-latest
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
         with:
           fetch-depth: 0
 

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/.github/workflows/pr-title.yaml

@@ -15,6 +15,6 @@ jobs:
     name: Validate PR title
     runs-on: ubuntu-latest
     steps:
-      - uses: amannn/action-semantic-pull-request@v5
+      - uses: amannn/action-semantic-pull-request@v6
         env:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/.github/workflows/publish-docker.yml

@@ -46,7 +46,7 @@ jobs:
           df -h
 
       - name: Checkout repository
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
         with:
           ref: ${{ github.ref }}
 

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/.github/workflows/release.yaml

@@ -13,7 +13,7 @@ jobs:
       contents: read
     steps:
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
 
       - name: Install uv
         uses: astral-sh/setup-uv@v6

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/.pre-commit-config.yaml

@@ -6,12 +6,12 @@ repos:
         stages: [commit-msg]
         additional_dependencies: ["@commitlint/config-conventional"]
   - repo: https://github.com/Goldziher/ai-rulez
-    rev: v1.1.4
+    rev: v1.5.1
     hooks:
       - id: ai-rulez-validate
       - id: ai-rulez-generate
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v5.0.0
+    rev: v6.0.0
     hooks:
       - id: name-tests-test
         args:
@@ -53,7 +53,7 @@ repos:
     hooks:
       - id: pyproject-fmt
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.12.7
+    rev: v0.12.9
     hooks:
       - id: ruff
         args: ["--fix", "--unsafe-fixes"]
@@ -66,7 +66,7 @@ repos:
         additional_dependencies:
           - tomli
   - repo: https://github.com/jsh9/pydoclint
-    rev: 0.6.7
+    rev: 0.6.10
     hooks:
       - id: pydoclint
         args:

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kreuzberg
-Version: 3.11.0
+Version: 3.11.2
 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
@@ -28,13 +28,13 @@ Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Classifier: Topic :: Text Processing :: General
 Classifier: Typing :: Typed
 Requires-Python: >=3.10
-Requires-Dist: anyio>=4.9.0
+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]>=1.9.0
-Requires-Dist: mcp>=1.12.3
+Requires-Dist: mcp>=1.13.0
 Requires-Dist: msgspec>=0.18.0
-Requires-Dist: playa-pdf>=0.6.4
+Requires-Dist: playa-pdf>=0.7.0
 Requires-Dist: psutil>=7.0.0
 Requires-Dist: pypdfium2==4.30.0
 Requires-Dist: python-calamine>=0.3.2
@@ -50,19 +50,19 @@ Requires-Dist: easyocr>=1.7.2; extra == 'all'
 Requires-Dist: fast-langdetect>=0.3.2; extra == 'all'
 Requires-Dist: gmft>=0.4.2; extra == 'all'
 Requires-Dist: keybert>=0.9.0; extra == 'all'
-Requires-Dist: litestar[opentelemetry,standard,structlog]>=2.16.0; extra == 'all'
+Requires-Dist: litestar[opentelemetry,standard,structlog]>=2.17.0; extra == 'all'
 Requires-Dist: mailparse>=1.0.15; extra == 'all'
-Requires-Dist: paddleocr>=3.1.0; extra == 'all'
+Requires-Dist: paddleocr>=3.1.1; extra == 'all'
 Requires-Dist: paddlepaddle>=3.1.0; extra == 'all'
 Requires-Dist: pandas>=2.3.1; extra == 'all'
-Requires-Dist: playa-pdf[crypto]>=0.6.4; extra == 'all'
+Requires-Dist: playa-pdf[crypto]>=0.7.0; extra == 'all'
 Requires-Dist: rich>=14.1.0; extra == 'all'
 Requires-Dist: semantic-text-splitter>=0.27.0; extra == 'all'
 Requires-Dist: setuptools>=80.9.0; extra == 'all'
 Requires-Dist: spacy>=3.8.7; extra == 'all'
 Requires-Dist: tomli>=2.0.0; (python_version < '3.11') and extra == 'all'
 Provides-Extra: api
-Requires-Dist: litestar[opentelemetry,standard,structlog]>=2.16.0; extra == 'api'
+Requires-Dist: litestar[opentelemetry,standard,structlog]>=2.17.0; extra == 'api'
 Provides-Extra: chunking
 Requires-Dist: semantic-text-splitter>=0.27.0; extra == 'chunking'
 Provides-Extra: cli
@@ -70,7 +70,7 @@ Requires-Dist: click>=8.2.1; extra == 'cli'
 Requires-Dist: rich>=14.1.0; extra == 'cli'
 Requires-Dist: tomli>=2.0.0; (python_version < '3.11') and extra == 'cli'
 Provides-Extra: crypto
-Requires-Dist: playa-pdf[crypto]>=0.6.4; extra == 'crypto'
+Requires-Dist: playa-pdf[crypto]>=0.7.0; extra == 'crypto'
 Provides-Extra: document-classification
 Requires-Dist: deep-translator>=1.11.4; extra == 'document-classification'
 Requires-Dist: pandas>=2.3.1; extra == 'document-classification'
@@ -84,7 +84,7 @@ Requires-Dist: gmft>=0.4.2; extra == 'gmft'
 Provides-Extra: langdetect
 Requires-Dist: fast-langdetect>=0.3.2; extra == 'langdetect'
 Provides-Extra: paddleocr
-Requires-Dist: paddleocr>=3.1.0; extra == 'paddleocr'
+Requires-Dist: paddleocr>=3.1.1; extra == 'paddleocr'
 Requires-Dist: paddlepaddle>=3.1.0; extra == 'paddleocr'
 Requires-Dist: setuptools>=80.9.0; extra == 'paddleocr'
 Description-Content-Type: text/markdown

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/docs/examples/extraction-examples.md

@@ -132,15 +132,15 @@ async def extract_tables_from_pdf():
     # Process extracted tables
     print(f"Found {len(result.tables)} tables")
     for i, table in enumerate(result.tables):
-        print(f"Table {i+1} on page {table.page_number}:")
-        print(table.text)  # Markdown formatted table
+        print(f"Table {i+1} on page {table['page_number']}:")
+        print(table["text"])  # Markdown formatted table
 
         # Work with the pandas DataFrame
-        df = table.df
+        df = table["df"]
         print(f"Table shape: {df.shape}")
 
         # The cropped table image is also available
-        # table.cropped_image.save(f"table_{i+1}.png")
+        # table['cropped_image'].save(f"table_{i+1}.png")
 
     # With custom GMFT configuration
     custom_config = ExtractionConfig(

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/docs/user-guide/extraction-configuration.md

@@ -237,10 +237,10 @@ result = await extract_file("document_with_tables.pdf", config=config)
 
 # Access extracted tables
 for i, table in enumerate(result.tables):
-    print(f"Table {i+1} on page {table.page_number}:")
-    print(table.text)  # Markdown formatted table text
+    print(f"Table {i+1} on page {table['page_number']}:")
+    print(table["text"])  # Markdown formatted table text
     # You can also access the pandas DataFrame directly
-    df = table.df
+    df = table["df"]
     print(df.shape)  # (rows, columns)
 ```
 

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/kreuzberg/_gmft.py

@@ -444,14 +444,26 @@ def _extract_tables_in_process(
                 cropped_image.save(img_bytes, format="PNG")
                 img_bytes.seek(0)
 
-                results.append(
-                    {
-                        "cropped_image_bytes": img_bytes.getvalue(),
-                        "page_number": cropped_table.page.page_number,
-                        "text": data_frame.to_markdown(),
-                        "df_csv": data_frame.to_csv(index=False),
-                    }
-                )
+                if data_frame.empty:
+                    results.append(
+                        {
+                            "cropped_image_bytes": img_bytes.getvalue(),
+                            "page_number": cropped_table.page.page_number,
+                            "text": data_frame.to_markdown(),
+                            "df_columns": data_frame.columns.tolist(),
+                            "df_csv": None,
+                        }
+                    )
+                else:
+                    results.append(
+                        {
+                            "cropped_image_bytes": img_bytes.getvalue(),
+                            "page_number": cropped_table.page.page_number,
+                            "text": data_frame.to_markdown(),
+                            "df_columns": None,
+                            "df_csv": data_frame.to_csv(index=False),
+                        }
+                    )
 
             result_queue.put((True, results))
 
@@ -532,7 +544,10 @@ def _extract_tables_isolated(
                 img = Image.open(io.BytesIO(table_dict["cropped_image_bytes"]))
                 import pandas as pd  # noqa: PLC0415
 
-                df = pd.read_csv(StringIO(table_dict["df_csv"]))
+                if table_dict["df_csv"] is None:
+                    df = pd.DataFrame(columns=table_dict["df_columns"])
+                else:
+                    df = pd.read_csv(StringIO(table_dict["df_csv"]))
 
                 tables.append(
                     TableData(
@@ -638,7 +653,10 @@ async def _extract_tables_isolated_async(
                 img = Image.open(io.BytesIO(table_dict["cropped_image_bytes"]))
                 import pandas as pd  # noqa: PLC0415
 
-                df = pd.read_csv(StringIO(table_dict["df_csv"]))
+                if table_dict["df_csv"] is None:
+                    df = pd.DataFrame(columns=table_dict["df_columns"])
+                else:
+                    df = pd.read_csv(StringIO(table_dict["df_csv"]))
 
                 tables.append(
                     TableData(

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/kreuzberg/_ocr/_easyocr.py

@@ -4,7 +4,6 @@ import warnings
 from dataclasses import dataclass
 from typing import TYPE_CHECKING, Any, ClassVar, Final, Literal
 
-import numpy as np
 from PIL import Image
 
 from kreuzberg._mime_types import PLAIN_TEXT_MIME_TYPE
@@ -188,6 +187,9 @@ class EasyOCRBackend(OCRBackend[EasyOCRConfig]):
 
         kwargs.pop("language", None)
         kwargs.pop("use_gpu", None)
+        kwargs.pop("device", None)
+        kwargs.pop("gpu_memory_limit", None)
+        kwargs.pop("fallback_to_cpu", None)
 
         try:
             result = await run_sync(
@@ -455,11 +457,16 @@ class EasyOCRBackend(OCRBackend[EasyOCRConfig]):
         Raises:
             OCRError: If OCR processing fails.
         """
+        import numpy as np  # noqa: PLC0415
+
         self._init_easyocr_sync(**kwargs)
 
         beam_width = kwargs.pop("beam_width")
         kwargs.pop("language", None)
         kwargs.pop("use_gpu", None)
+        kwargs.pop("device", None)
+        kwargs.pop("gpu_memory_limit", None)
+        kwargs.pop("fallback_to_cpu", None)
 
         try:
             result = self._reader.readtext(

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/kreuzberg/_ocr/_paddleocr.py

@@ -7,7 +7,6 @@ from importlib.util import find_spec
 from pathlib import Path
 from typing import TYPE_CHECKING, Any, ClassVar, Final, Literal
 
-import numpy as np
 from PIL import Image
 
 from kreuzberg._mime_types import PLAIN_TEXT_MIME_TYPE
@@ -380,6 +379,8 @@ class PaddleBackend(OCRBackend[PaddleOCRConfig]):
         Raises:
             OCRError: If OCR processing fails.
         """
+        import numpy as np  # noqa: PLC0415
+
         self._init_paddle_ocr_sync(**kwargs)
 
         if image.mode != "RGB":

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/pyproject.toml

@@ -5,7 +5,7 @@ requires = [ "hatchling" ]
 
 [project]
 name = "kreuzberg"
-version = "3.11.0"
+version = "3.11.2"
 description = "Document intelligence framework for Python - Extract text, metadata, and structured data from diverse file formats"
 readme = "README.md"
 keywords = [
@@ -57,13 +57,13 @@ classifiers = [
 ]
 
 dependencies = [
-  "anyio>=4.9.0",
+  "anyio>=4.10.0",
   "chardetng-py>=0.3.5",
   "exceptiongroup>=1.2.2; python_version<'3.11'",
   "html-to-markdown[lxml]>=1.9.0",
-  "mcp>=1.12.3",
+  "mcp>=1.13.0",
   "msgspec>=0.18.0",
-  "playa-pdf>=0.6.4",                                 # pinned due to breaking changes in 0.5.0
+  "playa-pdf>=0.7.0",                                 # pinned due to breaking changes in 0.5.0
   "psutil>=7.0.0",
   "pypdfium2==4.30.0",                                # pinned due to bug in 4.30.1, until v5 is stable
   "python-calamine>=0.3.2",
@@ -79,7 +79,7 @@ optional-dependencies.all = [
   "kreuzberg[additional-extensions,api,chunking,cli,crypto,document-classification,easyocr,entity-extraction,gmft,langdetect,paddleocr]",
 ]
 optional-dependencies.api = [
-  "litestar[standard,structlog,opentelemetry]>=2.16.0",
+  "litestar[standard,structlog,opentelemetry]>=2.17.0",
 ]
 optional-dependencies.chunking = [ "semantic-text-splitter>=0.27.0" ]
 optional-dependencies.cli = [
@@ -87,7 +87,7 @@ optional-dependencies.cli = [
   "rich>=14.1.0",
   "tomli>=2.0.0; python_version<'3.11'",
 ]
-optional-dependencies.crypto = [ "playa-pdf[crypto]>=0.6.4" ]
+optional-dependencies.crypto = [ "playa-pdf[crypto]>=0.7.0" ]
 optional-dependencies.document-classification = [
   "deep-translator>=1.11.4",
   "pandas>=2.3.1",
@@ -97,7 +97,7 @@ optional-dependencies.entity-extraction = [ "keybert>=0.9.0", "spacy>=3.8.7" ]
 optional-dependencies.gmft = [ "gmft>=0.4.2" ]
 optional-dependencies.langdetect = [ "fast-langdetect>=0.3.2" ]
 optional-dependencies.paddleocr = [
-  "paddleocr>=3.1.0",
+  "paddleocr>=3.1.1",
   "paddlepaddle>=3.1.0",
   "setuptools>=80.9.0",
 ]
@@ -111,13 +111,13 @@ scripts.kreuzberg-mcp = "kreuzberg._mcp.server:main"
 dev = [
   "covdefaults>=2.3.0",
   "mypy>=1.16.1",
-  "pre-commit>=4.2.0",
+  "pre-commit>=4.3.0",
   "pytest>=8.4.1",
   "pytest-cov>=6.2.1",
   "pytest-mock>=3.14.0",
   "pytest-rerunfailures>=15.1",
   "pytest-timeout>=2.4.0",
-  "ruff>=0.12.5",
+  "ruff>=0.12.9",
   "trio>=0.30.0",
   "uv-bump",
 ]

{kreuzberg-3.11.0 → kreuzberg-3.11.2}/tests/gmft_test.py

@@ -669,6 +669,125 @@ class TestGMFTInlineExtractionEdgeCases:
             pytest.skip("GMFT dependency not available for inline testing")
 
 
+class TestGMFTWithoutTables:
+    """Test GMFT behavior with PDFs that have no tables - issue #104."""
+
+    @pytest.mark.anyio
+    async def test_extract_tables_pdf_without_tables_async(self) -> None:
+        """Test that extract_tables handles PDFs without tables gracefully (async)."""
+        pdf_path = Path("tests/test_source_files/searchable.pdf")
+
+        try:
+            tables = await extract_tables(pdf_path)
+
+            assert isinstance(tables, list)
+
+            for table in tables:
+                assert "page_number" in table
+                assert "text" in table
+                assert "df" in table
+                assert "cropped_image" in table
+
+                import pandas as pd
+
+                assert isinstance(table["df"], pd.DataFrame)
+        except MissingDependencyError:
+            pytest.skip("GMFT dependency not installed")
+
+    def test_extract_tables_pdf_without_tables_sync(self) -> None:
+        """Test that extract_tables_sync handles PDFs without tables gracefully (sync)."""
+        # Using searchable.pdf which is a simple text PDF without tables
+        pdf_path = Path("tests/test_source_files/searchable.pdf")
+
+        try:
+            tables = extract_tables_sync(pdf_path)
+
+            assert isinstance(tables, list)
+
+            for table in tables:
+                assert "page_number" in table
+                assert "text" in table
+                assert "df" in table
+                assert "cropped_image" in table
+
+                import pandas as pd
+
+                assert isinstance(table["df"], pd.DataFrame)
+        except MissingDependencyError:
+            pytest.skip("GMFT dependency not installed")
+
+    @pytest.mark.anyio
+    async def test_extract_file_with_gmft_pdf_without_tables(self) -> None:
+        """Test that extract_file with extract_tables=True handles PDFs without tables gracefully."""
+        pdf_path = Path("tests/test_source_files/searchable.pdf")
+
+        config = ExtractionConfig(
+            extract_tables=True,
+            gmft_config=GMFTConfig(
+                detector_base_threshold=0.85,
+                remove_null_rows=True,
+                enable_multi_header=True,
+            ),
+        )
+
+        try:
+            result = await extract_file(pdf_path, config=config)
+
+            assert result.content
+            assert "Sample PDF" in result.content
+
+            assert hasattr(result, "tables")
+            assert isinstance(result.tables, list)
+
+            for table in result.tables:
+                assert "page_number" in table
+                assert "text" in table
+                assert "df" in table
+                assert "cropped_image" in table
+
+                import pandas as pd
+
+                assert isinstance(table["df"], pd.DataFrame)
+        except MissingDependencyError:
+            pytest.skip("GMFT dependency not installed")
+
+    def test_extract_file_sync_with_gmft_pdf_without_tables(self) -> None:
+        """Test that extract_file_sync with extract_tables=True handles PDFs without tables gracefully."""
+        pdf_path = Path("tests/test_source_files/searchable.pdf")
+
+        from kreuzberg.extraction import extract_file_sync
+
+        config = ExtractionConfig(
+            extract_tables=True,
+            gmft_config=GMFTConfig(
+                detector_base_threshold=0.85,
+                remove_null_rows=True,
+                enable_multi_header=True,
+            ),
+        )
+
+        try:
+            result = extract_file_sync(pdf_path, config=config)
+
+            assert result.content
+            assert "Sample PDF" in result.content
+
+            assert hasattr(result, "tables")
+            assert isinstance(result.tables, list)
+
+            for table in result.tables:
+                assert "page_number" in table
+                assert "text" in table
+                assert "df" in table
+                assert "cropped_image" in table
+
+                import pandas as pd
+
+                assert isinstance(table["df"], pd.DataFrame)
+        except MissingDependencyError:
+            pytest.skip("GMFT dependency not installed")
+
+
 class TestGMFTConfigSerialization:
     """Test GMFTConfig serialization for multiprocessing."""