kreuzberg 3.8.2__tar.gz → 3.9.1__tar.gz

kreuzberg-3.9.1/.deepsource.toml

@@ -0,0 +1,54 @@
+version = 1
+
+test_patterns = ["tests/**"]
+
+exclude_patterns = [
+    # Virtual environments
+    ".venv/**",
+    "venv/**",
+
+    # Build and distribution artifacts
+    "dist/**",
+    "build/**",
+    "*.egg-info/**",
+
+    # Documentation
+    "docs/**",
+    "site/**",
+
+    # Cache directories
+    "**/__pycache__/**",
+    ".pytest_cache/**",
+    ".mypy_cache/**",
+    ".ruff_cache/**",
+    ".coverage",
+    "htmlcov/**",
+
+    # Benchmarks and performance tests
+    "benchmarks/**",
+
+    # IDE and editor files
+    ".idea/**",
+    ".vscode/**",
+
+    # Version control
+    ".git/**",
+
+    # Temporary and generated files
+    "*.pyc",
+    ".DS_Store",
+    "*.swp",
+    "*.swo",
+]
+
+[[analyzers]]
+name = "test-coverage"
+
+[[analyzers]]
+name = "python"
+
+[analyzers.meta]
+runtime_version = "3.x.x"
+
+[[transformers]]
+name = "ruff"

kreuzberg-3.9.1/.github/workflows/ci.yaml

@@ -0,0 +1,197 @@
+name: CI
+
+on:
+  pull_request:
+    branches:
+      - main
+  push:
+    branches:
+      - main
+      - feat/smart-multiprocessing
+
+jobs:
+  validate:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v6
+        with:
+          enable-cache: true
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version-file: "pyproject.toml"
+
+      - name: Install Dependencies
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 5
+          max_attempts: 3
+          retry_wait_seconds: 30
+          command: |
+            if [[ "${{ runner.os }}" == "Windows" ]] && [[ -d ".venv" ]]; then
+              echo "Removing existing .venv directory on Windows"
+              rm -rf .venv
+            fi
+            uv sync --all-packages --all-extras --dev
+          shell: bash
+
+      - name: Load Cached Pre-Commit Dependencies
+        id: cached-pre-commit-dependencies
+        uses: actions/cache@v4
+        with:
+          path: ~/.cache/pre-commit/
+          key: pre-commit|${{ env.pythonLocation }}|${{ hashFiles('.pre-commit-config.yaml') }}
+
+      - name: Execute Pre-Commit
+        run: uv run pre-commit run --show-diff-on-failure --color=always --all-files
+
+  test:
+    strategy:
+      matrix:
+        os: [ ubuntu-latest, macOS-latest, windows-latest ]
+        python: ${{ github.event_name == 'pull_request' && fromJSON('["3.13"]') || fromJSON('["3.10", "3.11", "3.12", "3.13"]') }}
+    runs-on: ${{ matrix.os }}
+    timeout-minutes: 30
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+
+      - name: Install uv
+        uses: astral-sh/setup-uv@v6
+        with:
+          enable-cache: true
+
+      - name: Install Python
+        uses: actions/setup-python@v5
+        id: setup-python
+        with:
+          python-version: ${{ matrix.python }}
+
+      - name: Cache Python Dependencies
+        id: python-cache
+        uses: actions/cache@v4
+        with:
+          path: |
+            ~/.cache/uv
+            .venv
+          key: python-dependencies-${{ matrix.os }}-${{ steps.setup-python.outputs.python-version }}-${{ hashFiles('uv.lock') }}
+          restore-keys: |
+            python-dependencies-${{ matrix.os }}-${{ matrix.python }}-
+
+      - name: Install Dependencies
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 5
+          max_attempts: 3
+          retry_wait_seconds: 30
+          command: |
+            if [[ "${{ runner.os }}" == "Windows" ]] && [[ -d ".venv" ]]; then
+              echo "Removing existing .venv directory on Windows"
+              rm -rf .venv
+            fi
+            uv sync --all-packages --all-extras --dev
+          shell: bash
+
+      - name: Cache Test Artifacts
+        uses: actions/cache@v4
+        with:
+          path: .pytest_cache/
+          key: pytest-cache-${{ matrix.os }}-${{ matrix.python }}
+
+      - name: Cache and Install Homebrew (macOS)
+        if: runner.os == 'macOS'
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 10
+          max_attempts: 3
+          retry_wait_seconds: 30
+          command: |
+            # Using the underlying homebrew commands instead of the action
+            brew update || true
+            brew install tesseract tesseract-lang pandoc || brew upgrade tesseract tesseract-lang pandoc || true
+            brew list tesseract tesseract-lang pandoc
+          shell: bash
+
+      - name: Cache and Install APT Packages (Linux)
+        if: runner.os == 'Linux'
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 5
+          max_attempts: 3
+          retry_wait_seconds: 30
+          command: |
+            sudo apt-get update
+            sudo apt-get install -y tesseract-ocr tesseract-ocr-deu pandoc
+          shell: bash
+
+      - name: Install System Dependencies (Windows)
+        if: runner.os == 'Windows'
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 10
+          max_attempts: 3
+          retry_wait_seconds: 30
+          command: |
+            choco install -y tesseract pandoc --no-progress
+            Write-Output "C:\Program Files\Tesseract-OCR" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
+            Write-Output "C:\Program Files\Pandoc" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
+            $env:PATH = "C:\Program Files\Tesseract-OCR;C:\Program Files\Pandoc;" + $env:PATH
+            tesseract --version
+            pandoc --version
+          shell: pwsh
+
+      - name: Clean Coverage Data
+        run: |
+          rm -f .coverage .coverage.* coverage.lcov htmlcov/* || true
+        shell: bash
+
+      - name: Run Tests with Coverage
+        run: |
+          uv run coverage erase
+          uv run pytest -s -vvv --cov=kreuzberg --cov-report=lcov:coverage.lcov --cov-report=term --cov-config=pyproject.toml
+
+      - name: Upload Coverage Artifacts
+        if: matrix.os == 'ubuntu-latest' && matrix.python == '3.13'
+        uses: actions/upload-artifact@v4
+        with:
+          name: coverage-report
+          path: coverage.lcov
+          retention-days: 1
+
+  upload-coverage:
+    needs: test
+    runs-on: ubuntu-latest
+    if: github.event_name == 'push' || github.event_name == 'pull_request'
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v4
+        with:
+          ref: ${{ github.event.pull_request.head.sha || github.sha }}
+
+      - name: Download Coverage Artifacts
+        uses: actions/download-artifact@v4
+        with:
+          name: coverage-report
+          path: .
+
+      - name: Install DeepSource CLI
+        uses: nick-fields/retry@v3
+        with:
+          timeout_minutes: 3
+          max_attempts: 3
+          retry_wait_seconds: 10
+          command: |
+            curl -fsSL https://deepsource.io/cli | sh
+          shell: bash
+
+      - name: Upload Coverage to DeepSource
+        env:
+          DEEPSOURCE_DSN: ${{ secrets.DEEPSOURCE_DSN }}
+        run: |
+          ./bin/deepsource report --analyzer test-coverage --key python --value-file ./coverage.lcov

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/.gitignore

@@ -1,5 +1,6 @@
 *$py.class
 *.Cache
+.clause/
 *.cscfg
 *.egg-info/
 *.log
@@ -9,6 +10,8 @@
 *temp/
 .coverage
 .coverage*
+coverage.lcov
+htmlcov/
 .cursorrules
 .dist/
 .DS_store

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/.pre-commit-config.yaml

@@ -53,7 +53,7 @@ repos:
     hooks:
       - id: pyproject-fmt
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.12.2
+    rev: v0.12.5
     hooks:
       - id: ruff
         args: ["--fix", "--unsafe-fixes"]

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/PKG-INFO

@@ -1,13 +1,13 @@
 Metadata-Version: 2.4
 Name: kreuzberg
-Version: 3.8.2
+Version: 3.9.1
 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
 Author-email: Na'aman Hirschfeld <nhirschfed@gmail.com>
 License: MIT
 License-File: LICENSE
-Keywords: async,document-analysis,document-intelligence,document-processing,extensible,information-extraction,mcp,metadata-extraction,model-context-protocol,ocr,pandoc,pdf-extraction,pdfium,plugin-architecture,rag,retrieval-augmented-generation,structured-data,table-extraction,tesseract,text-extraction
+Keywords: async,document-analysis,document-classification,document-intelligence,document-processing,extensible,information-extraction,mcp,metadata-extraction,model-context-protocol,ocr,pandoc,pdf-extraction,pdfium,plugin-architecture,rag,retrieval-augmented-generation,structured-data,table-extraction,tesseract,text-extraction
 Classifier: Development Status :: 5 - Production/Stable
 Classifier: Intended Audience :: Developers
 Classifier: Intended Audience :: Information Technology
@@ -29,12 +29,12 @@ Classifier: Topic :: Text Processing :: General
 Classifier: Typing :: Typed
 Requires-Python: >=3.10
 Requires-Dist: anyio>=4.9.0
-Requires-Dist: chardetng-py>=0.3.4
+Requires-Dist: chardetng-py>=0.3.5
 Requires-Dist: exceptiongroup>=1.2.2; python_version < '3.11'
-Requires-Dist: html-to-markdown[lxml]>=1.8.0
-Requires-Dist: mcp>=1.11.0
+Requires-Dist: html-to-markdown[lxml]>=1.9.0
+Requires-Dist: mcp>=1.12.2
 Requires-Dist: msgspec>=0.18.0
-Requires-Dist: playa-pdf>=0.6.1
+Requires-Dist: playa-pdf>=0.6.4
 Requires-Dist: psutil>=7.0.0
 Requires-Dist: pypdfium2==4.30.0
 Requires-Dist: python-calamine>=0.3.2
@@ -53,18 +53,21 @@ Requires-Dist: litestar[opentelemetry,standard,structlog]>=2.16.0; extra == 'all
 Requires-Dist: mailparse>=1.0.15; extra == 'all'
 Requires-Dist: paddleocr>=3.1.0; extra == 'all'
 Requires-Dist: paddlepaddle>=3.1.0; extra == 'all'
-Requires-Dist: rich>=14.0.0; 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'
+Provides-Extra: auto-classify-document-type
+Requires-Dist: deep-translator>=1.11.4; extra == 'auto-classify-document-type'
+Requires-Dist: pandas>=2.3.1; extra == 'auto-classify-document-type'
 Provides-Extra: chunking
 Requires-Dist: semantic-text-splitter>=0.27.0; extra == 'chunking'
 Provides-Extra: cli
 Requires-Dist: click>=8.2.1; extra == 'cli'
-Requires-Dist: rich>=14.0.0; 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: easyocr
 Requires-Dist: easyocr>=1.7.2; extra == 'easyocr'
@@ -88,7 +91,7 @@ Description-Content-Type: text/markdown
 [![Documentation](https://img.shields.io/badge/docs-kreuzberg.dev-blue)](https://kreuzberg.dev/)
 [![Benchmarks](https://img.shields.io/badge/benchmarks-fastest%20CPU-orange)](https://benchmarks.kreuzberg.dev/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-[![Test Coverage](https://img.shields.io/badge/coverage-95%25-green)](https://github.com/Goldziher/kreuzberg)
+[![DeepSource](https://app.deepsource.com/gh/Goldziher/kreuzberg.svg/?label=code+coverage&show_trend=true&token=U8AW1VWWSLwVhrbtL8LmLBDN)](https://app.deepsource.com/gh/Goldziher/kreuzberg/)
 
 **A document intelligence framework for Python.** Extract text, metadata, and structured information from diverse document formats through a unified, extensible API. Built on established open source foundations including Pandoc, PDFium, and Tesseract.
 
@@ -103,6 +106,7 @@ Description-Content-Type: text/markdown
 - **Format Support**: 18 document types including PDF, Microsoft Office, images, HTML, and structured data formats
 - **OCR Integration**: Multiple OCR engines (Tesseract, EasyOCR, PaddleOCR) with automatic fallback
 - **Table Detection**: Structured table extraction with cell-level precision via GMFT integration
+- **Document Classification**: Automatic document type detection (contracts, forms, invoices, receipts, reports)
 
 ### Technical Architecture
 
@@ -126,14 +130,14 @@ Kreuzberg leverages established open source technologies:
 ### Extract Text with CLI
 
 ```bash
-# Extract text from any file to markdown
-uvx kreuzberg extract document.pdf > output.md
+# Extract text from any file to text format
+uvx kreuzberg extract document.pdf > output.txt
 
 # With all features (OCR, table extraction, etc.)
-uvx --from "kreuzberg[all]" kreuzberg extract invoice.pdf --ocr --format markdown
+uvx --from "kreuzberg[all]" kreuzberg extract invoice.pdf --ocr-backend tesseract --output-format text
 
 # Extract with rich metadata
-uvx kreuzberg extract report.pdf --show-metadata --format json
+uvx kreuzberg extract report.pdf --show-metadata --output-format json
 ```
 
 ### Python Usage

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/README.md

@@ -5,7 +5,7 @@
 [![Documentation](https://img.shields.io/badge/docs-kreuzberg.dev-blue)](https://kreuzberg.dev/)
 [![Benchmarks](https://img.shields.io/badge/benchmarks-fastest%20CPU-orange)](https://benchmarks.kreuzberg.dev/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-[![Test Coverage](https://img.shields.io/badge/coverage-95%25-green)](https://github.com/Goldziher/kreuzberg)
+[![DeepSource](https://app.deepsource.com/gh/Goldziher/kreuzberg.svg/?label=code+coverage&show_trend=true&token=U8AW1VWWSLwVhrbtL8LmLBDN)](https://app.deepsource.com/gh/Goldziher/kreuzberg/)
 
 **A document intelligence framework for Python.** Extract text, metadata, and structured information from diverse document formats through a unified, extensible API. Built on established open source foundations including Pandoc, PDFium, and Tesseract.
 
@@ -20,6 +20,7 @@
 - **Format Support**: 18 document types including PDF, Microsoft Office, images, HTML, and structured data formats
 - **OCR Integration**: Multiple OCR engines (Tesseract, EasyOCR, PaddleOCR) with automatic fallback
 - **Table Detection**: Structured table extraction with cell-level precision via GMFT integration
+- **Document Classification**: Automatic document type detection (contracts, forms, invoices, receipts, reports)
 
 ### Technical Architecture
 
@@ -43,14 +44,14 @@ Kreuzberg leverages established open source technologies:
 ### Extract Text with CLI
 
 ```bash
-# Extract text from any file to markdown
-uvx kreuzberg extract document.pdf > output.md
+# Extract text from any file to text format
+uvx kreuzberg extract document.pdf > output.txt
 
 # With all features (OCR, table extraction, etc.)
-uvx --from "kreuzberg[all]" kreuzberg extract invoice.pdf --ocr --format markdown
+uvx --from "kreuzberg[all]" kreuzberg extract invoice.pdf --ocr-backend tesseract --output-format text
 
 # Extract with rich metadata
-uvx kreuzberg extract report.pdf --show-metadata --format json
+uvx kreuzberg extract report.pdf --show-metadata --output-format json
 ```
 
 ### Python Usage

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/benchmarks/end_to_end_benchmark.py

@@ -43,7 +43,7 @@ async def run_end_to_end_benchmark(trials: int = 20) -> dict[str, Any]:
     print(f"Tables: {len(cold_result.tables)}")
     print(f"Chunks: {len(cold_result.chunks)}")
 
-    from kreuzberg._utils._cache import (
+    from kreuzberg._utils._cache import (  # noqa: PLC0415
         get_ocr_cache,
         get_table_cache,
         get_mime_cache,

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/benchmarks/statistical_benchmark.py

@@ -130,7 +130,7 @@ async def run_statistical_benchmark() -> dict[str, Any]:
         f"  Cache consistency: {'✅ STABLE' if warm_clean_stdev / warm_clean_mean < 0.1 else '⚠️ VARIABLE'}"
     )
 
-    from kreuzberg._utils._cache import (
+    from kreuzberg._utils._cache import (  # noqa: PLC0415
         get_ocr_cache,
         get_table_cache,
         get_mime_cache,

kreuzberg-3.9.1/docs/changelog.md

@@ -0,0 +1,49 @@
+# Changelog
+
+All notable changes to Kreuzberg will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [3.9.0] - 2025-01-17
+
+### Added
+
+- Automatic Document Type Detection (#88) - A new feature for classifying documents into categories (contract, form, invoice, receipt, report)
+    - Integration with Google Translate for multi-language support
+    - New optional dependency group `auto-classify-document-type` with `deep-translator` and `pandas`
+    - Comprehensive tests and documentation
+- DeepSource integration for code quality analysis
+
+### Fixed
+
+- PDF extraction handling when no OCR backend is available
+- Entity extraction test updated to use frozenset of tuples
+- Config handling for dataclasses with `slots=True` - replaced `config.__dict__` with `asdict(config)`
+- Coverage configuration and cleanup issues
+
+### Changed
+
+- CI/CD: Added retry logic for flaky steps across all platforms
+- Improved coverage gathering and cleanup in test runs
+- Updated dependencies in `uv.lock`
+
+## [3.8.2] - Previous Release
+
+### Added
+
+- Documentation site with comprehensive examples and API reference
+- Improved configuration for all OCR backends
+- Added hooks system for validation and post-processing
+- Language detection feature with `auto_detect_language` configuration option
+- New optional dependency group `langdetect` for automatic language detection
+
+### Changed
+
+- Refactored internal structure for better maintainability
+- Updated extraction functions to use config object instead of kwargs
+- Improved error messages and reporting
+
+## Previous Versions
+
+For a complete history of changes, please refer to the [GitHub releases page](https://github.com/strickvl/kreuzberg/releases).

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/docs/index.md

@@ -49,6 +49,7 @@ Kreuzberg addresses the complete document intelligence pipeline through a modula
 - **OCR Engines**: Tesseract (default), EasyOCR, PaddleOCR with automatic fallback strategies
 - **Data Extraction**: Text content, document metadata, table structures, and embedded resources
 - **Processing Capabilities**: Content chunking for RAG pipelines, language detection, format preservation
+- **Document Classification**: Automatic document type detection (contracts, forms, invoices, receipts, reports)
 - **Extensibility**: Plugin architecture for custom extractors and hooks
 
 ## Architecture Philosophy

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/docs/user-guide/basic-usage.md

@@ -131,3 +131,31 @@ async def show_metadata():
 
 asyncio.run(show_metadata())
 ```
+
+## Document Classification
+
+Kreuzberg can automatically classify documents into categories (contracts, forms, invoices, receipts, reports):
+
+```python
+import asyncio
+from kreuzberg import extract_file, ExtractionConfig
+
+async def classify_document():
+    config = ExtractionConfig(
+        auto_detect_document_type=True,
+        document_classification_mode="text",  # or "vision" for better accuracy
+        type_confidence_threshold=0.5,
+    )
+
+    result = await extract_file("invoice.pdf", config=config)
+
+    # Access classification results
+    if result.document_type:
+        print(f"Document type: {result.document_type}")
+        print(f"Confidence: {result.type_confidence:.2%}")
+
+    # The extracted content is still available
+    print(f"Content: {result.content[:200]}...")
+
+asyncio.run(classify_document())
+```

kreuzberg-3.9.1/docs/user-guide/document-classification.md

@@ -0,0 +1,53 @@
+# Automatic Document Classification
+
+Kreuzberg can automatically classify documents into common types like invoices, contracts, and receipts. This allows you to build custom processing pipelines tailored to each document type.
+
+## Enabling Document Classification
+
+To enable this feature, set `auto_detect_document_type=True` in your `ExtractionConfig`:
+
+```python
+from kreuzberg import ExtractionConfig, extract_file
+
+config = ExtractionConfig(auto_detect_document_type=True)
+result = await extract_file("path/to/your/document.pdf", config=config)
+
+if result.document_type:
+    print(f"Detected document type: {result.document_type}")
+    print(f"Confidence: {result.document_type_confidence:.2f}")
+```
+
+## Classification Modes
+
+You can choose between two classification modes using the `document_classification_mode` parameter in `ExtractionConfig`:
+
+- `"text"` (default): This mode uses a rule-based classifier that analyzes the extracted text for keywords and patterns. It's fast and works well for text-based documents.
+- `"vision"`: This mode uses layout information from OCR to identify document types. It's more accurate for scanned documents and images, but it requires the Tesseract OCR backend.
+
+Here's how to use the vision-based classifier:
+
+```python
+config = ExtractionConfig(
+    auto_detect_document_type=True,
+    document_classification_mode="vision",
+    force_ocr=True,  # Recommended for vision-based classification
+)
+```
+
+## Confidence Threshold
+
+You can control the minimum confidence required for a classification to be considered valid by setting the `type_confidence_threshold` in `ExtractionConfig`. The default value is `0.7`.
+
+```python
+config = ExtractionConfig(
+    auto_detect_document_type=True,
+    type_confidence_threshold=0.85,  # Require 85% confidence
+)
+```
+
+## Output
+
+The classification results are available in the `ExtractionResult` object:
+
+- `document_type`: The detected document type (e.g., `"invoice"`, `"contract"`) or `None` if no type was detected with sufficient confidence.
+- `type_confidence`: The confidence score of the detection (a float between 0.0 and 1.0) or `None`.

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/docs/user-guide/extraction-configuration.md

@@ -31,6 +31,9 @@ max_chars = 2000
 max_overlap = 100
 ocr_backend = "tesseract"
 auto_detect_language = true
+auto_detect_document_type = true
+document_classification_mode = "text"  # or "vision"
+type_confidence_threshold = 0.5
 
 # Tesseract OCR configuration
 [tesseract]
@@ -76,6 +79,9 @@ force_ocr = false
 chunk_content = true
 extract_tables = true
 auto_detect_language = true
+auto_detect_document_type = true
+document_classification_mode = "text"
+type_confidence_threshold = 0.5
 
 [tool.kreuzberg.tesseract]
 language = "eng"

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/docs/user-guide/index.md

@@ -8,6 +8,7 @@ This guide provides comprehensive documentation for the Kreuzberg document intel
 - [Extraction Configuration](extraction-configuration.md) - Configure the extraction process ([API](../api-reference/types.md#extractionconfig))
 - [Metadata Extraction](metadata-extraction.md) - Document metadata extraction ([API](../api-reference/types.md#metadata))
 - [Content Chunking](chunking.md) - Split documents into manageable chunks
+- [Document Classification](document-classification.md) - Automatic document type detection
 - [OCR Configuration](ocr-configuration.md) - Configure OCR settings ([API](../api-reference/ocr-configuration.md))
 - [OCR Backends](ocr-backends.md) - Choose and configure different OCR engines
 - [Supported Formats](supported-formats.md) - All supported document formats

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/kreuzberg/_chunker.py

@@ -2,9 +2,9 @@ from __future__ import annotations
 
 from typing import TYPE_CHECKING
 
-from kreuzberg import MissingDependencyError
 from kreuzberg._constants import DEFAULT_MAX_CHARACTERS, DEFAULT_MAX_OVERLAP
 from kreuzberg._mime_types import MARKDOWN_MIME_TYPE
+from kreuzberg.exceptions import MissingDependencyError
 
 if TYPE_CHECKING:
     from semantic_text_splitter import MarkdownSplitter, TextSplitter
@@ -36,11 +36,11 @@ def get_chunker(
     if key not in _chunkers:
         try:
             if mime_type == MARKDOWN_MIME_TYPE:
-                from semantic_text_splitter import MarkdownSplitter
+                from semantic_text_splitter import MarkdownSplitter  # noqa: PLC0415
 
                 _chunkers[key] = MarkdownSplitter(max_characters, overlap_characters)
             else:
-                from semantic_text_splitter import TextSplitter
+                from semantic_text_splitter import TextSplitter  # noqa: PLC0415
 
                 _chunkers[key] = TextSplitter(max_characters, overlap_characters)
         except ImportError as e:

{kreuzberg-3.8.2 → kreuzberg-3.9.1}/kreuzberg/_config.py

@@ -95,7 +95,7 @@ def parse_ocr_backend_config(
         # Convert psm integer to PSMMode enum if needed
         processed_config = backend_config.copy()
         if "psm" in processed_config and isinstance(processed_config["psm"], int):
-            from kreuzberg._ocr._tesseract import PSMMode
+            from kreuzberg._ocr._tesseract import PSMMode  # noqa: PLC0415
 
             processed_config["psm"] = PSMMode(processed_config["psm"])
         return TesseractConfig(**processed_config)