# Docling Analysis Framework
[](https://www.python.org/downloads/)
[](https://opensource.org/licenses/MIT)
[](https://github.com/DS4SD/docling)
[](https://github.com/rdwj/docling-analysis-framework)
AI-ready analysis framework for **PDF and Office documents** using Docling for content extraction. Transforms Docling's output into optimized chunks and structured analysis for AI/ML pipelines.
> **📝 For text, code, and configuration files**, use our companion [document-analysis-framework](https://github.com/redhat-ai-americas/document-analysis-framework) which uses only Python standard library.
## 🚀 Quick Start
### Simple API - Get Started in Seconds
```python
import docling_analysis_framework as daf
# 🎯 One-line analysis with Docling extraction
result = daf.analyze("document.pdf")
print(f"Document type: {result['document_type'].type_name}")
print(f"Pages: {result['document_type'].pages}")
print(f"Handler used: {result['handler_used']}")
# ✂️ Smart chunking for AI/ML
chunks = daf.chunk("document.pdf", strategy="auto")
print(f"Created {len(chunks)} optimized chunks")
# 🚀 Enhanced analysis with both analysis and chunking
enhanced = daf.analyze_enhanced("document.pdf", chunking_strategy="structural")
print(f"Document: {enhanced['analysis']['document_type'].type_name}")
print(f"Chunks: {len(enhanced['chunks'])}")
# 💾 Save chunks to JSON
daf.save_chunks_to_json(chunks, "chunks_output.json")
# 💾 Save analysis to JSON
daf.save_analysis_to_json(result, "analysis_output.json")
```
### Advanced Usage
```python
import docling_analysis_framework as daf
# Enhanced analysis with full results
enhanced = daf.analyze_enhanced("research_paper.pdf")
print(f"Type: {enhanced['analysis']['document_type'].type_name}")
print(f"Confidence: {enhanced['analysis']['confidence']:.2f}")
print(f"AI use cases: {len(enhanced['analysis']['analysis'].ai_use_cases)}")
if enhanced['analysis']['analysis'].quality_metrics:
for metric, score in enhanced['analysis']['analysis'].quality_metrics.items():
print(f"{metric}: {score:.2f}")
# Different chunking strategies
hierarchical_chunks = daf.chunk("document.pdf", strategy="structural")
table_aware_chunks = daf.chunk("document.pdf", strategy="table_aware")
page_aware_chunks = daf.chunk("document.pdf", strategy="page_aware")
# Process chunks
for chunk in hierarchical_chunks:
print(f"Chunk {chunk.chunk_id}: {chunk.token_count} tokens")
print(f"Type: {chunk.chunk_type}")
# Send to your AI model...
# 💾 Save different strategies to separate files
strategies = {
"structural": hierarchical_chunks,
"table_aware": table_aware_chunks,
"page_aware": page_aware_chunks
}
for strategy_name, chunks in strategies.items():
daf.save_chunks_to_json(chunks, f"chunks_{strategy_name}.json")
print(f"Saved {len(chunks)} chunks to chunks_{strategy_name}.json")
```
### Expert Usage - Direct Class Access
```python
# For advanced customization, use the classes directly
from docling_analysis_framework import DoclingAnalyzer, DoclingChunkingOrchestrator, ChunkingConfig
analyzer = DoclingAnalyzer(max_file_size_mb=100)
result = analyzer.analyze_document("document.pdf")
# Custom chunking with config
config = ChunkingConfig(
max_chunk_size=1500,
min_chunk_size=200,
overlap_size=100,
preserve_structure=True
)
orchestrator = DoclingChunkingOrchestrator(config=config)
# Note: In advanced usage, you'd pass the actual docling_result object
```
## 📋 Supported Document Types
| Format | Extensions | Confidence | Powered By |
|--------|------------|------------|------------|
| **📄 PDF Documents** | .pdf | 95% | Docling extraction |
| **📝 Word Documents** | .docx | 95% | Docling extraction |
| **📊 Spreadsheets** | .xlsx | 70% | Docling extraction |
| **📅 Presentations** | .pptx | 70% | Docling extraction |
| **🖼️ Images with Text** | .png, .jpg, .tiff | 70% | Docling OCR |
## 🎯 Key Features
### **🔍 Docling-Powered Extraction**
- **PDF text extraction** - High-quality content extraction
- **Table detection** - Preserves table structure in markdown
- **Figure references** - Maintains image/figure relationships
- **Header hierarchy** - Document structure preservation
### **🤖 AI Preparation Layer**
- **Quality assessment** - Extraction quality scoring
- **Structure analysis** - Document type detection and analysis
- **Chunk optimization** - AI-ready segmentation strategies
- **Rich metadata** - Page counts, figures, tables, quality metrics
### **⚡ Smart Chunking Strategies**
- **Structural chunking** - Respects document hierarchy (headers, sections)
- **Table-aware chunking** - Separates tables from text content
- **Page-aware chunking** - Considers original page boundaries
- **Auto-selection** - Document-type-aware strategy selection
### **📊 Extraction Quality Analysis**
- **Text coverage** - How much text was successfully extracted
- **Structure preservation** - Whether headers, lists, tables are maintained
- **Overall confidence** - Combined quality score for AI processing
## 🎉 Framework Status - COMPLETED
The Docling Analysis Framework is now **fully functional** and follows the same successful patterns as the XML Analysis Framework:
### ✅ Completed Features
- **🎯 Simple API**: One-line functions for `analyze()`, `chunk()`, `analyze_enhanced()`
- **🔧 Advanced API**: Direct class access for customization with `DoclingAnalyzer`, `DoclingChunkingOrchestrator`
- **⚙️ Configurable Chunking**: `ChunkingConfig` class for fine-tuning chunk parameters
- **📦 Multiple Strategies**: Structural, table-aware, page-aware, and auto-selection chunking
- **💾 JSON Export**: Easy export of analysis results and chunks to JSON files
- **🛡️ Enhanced Error Handling**: Comprehensive logging and error reporting
- **📊 Quality Metrics**: Extraction quality assessment and content analysis
- **🧪 Testing Framework**: Complete Jupyter notebook for validation
- **📚 Documentation**: Comprehensive README with examples and usage patterns
### 🚀 Ready for AI/ML Integration
The framework provides everything needed for AI/ML pipelines:
- **Token-optimized chunks** sized for LLM context windows
- **Rich metadata** with document structure and quality metrics
- **JSON export** for vector database ingestion
- **Multiple chunking strategies** for different document types
- **Quality assessment** to determine AI suitability
### 🔄 Comparison with XML Framework
| Feature | XML Framework | Docling Framework | Status |
|---------|---------------|-------------------|---------|
| Simple API | ✅ | ✅ | Complete |
| Advanced Classes | ✅ | ✅ | Complete |
| Multiple Strategies | ✅ | ✅ | Complete |
| Configuration | ✅ | ✅ | Complete |
| JSON Export | ✅ | ✅ | Complete |
| Error Handling | ✅ | ✅ | Complete |
| Testing Notebooks | ✅ | ✅ | Complete |
| Quality Metrics | ✅ | ✅ | Complete |
Both frameworks now provide identical API patterns and functionality!
## 🔧 Installation
```bash
# Install Docling first
pip install docling
# Install framework
git clone https://github.com/rdwj/docling-analysis-framework.git
cd docling-analysis-framework
pip install -e .
```
## 📖 Usage Examples
### Document Analysis with Quality Assessment
```python
from core.analyzer import DoclingAnalyzer
analyzer = DoclingAnalyzer()
result = analyzer.analyze_document("contract.pdf")
# Access extraction quality
quality = result['analysis'].extraction_quality
print(f"Text coverage: {quality['text_coverage']:.2f}")
print(f"Structure score: {quality['structure_score']:.2f}")
print(f"Overall quality: {quality['overall_score']:.2f}")
# Access document insights
findings = result['analysis'].key_findings
print(f"Pages: {findings['page_count']}")
print(f"Tables: {findings['table_rows']}")
print(f"Figures: {findings['figure_count']}")
```
### Advanced Chunking for Academic Papers
```python
# Perfect for research papers with complex structure
chunks = orchestrator.chunk_document(
"paper.pdf",
markdown_content,
docling_result,
strategy='structural'
)
# Chunks respect paper structure
for chunk in chunks:
if chunk.chunk_type == 'section':
print(f"Section: {chunk.metadata['section_title']}")
elif chunk.chunk_type == 'table':
print(f"Table data: {len(chunk.content)} chars")
elif chunk.chunk_type == 'figure':
print(f"Figure reference preserved")
```
### File Size Management
```python
from core.analyzer import DoclingAnalyzer
# Large PDF processing with limits
analyzer = DoclingAnalyzer(max_file_size_mb=100.0)
result = analyzer.analyze_document("large_manual.pdf")
if 'error' in result:
print(f"File too large: {result['error']}")
else:
print(f"Successfully processed {result['file_size']} bytes")
```
## 🧪 Framework Ecosystem
This framework is part of a larger document analysis ecosystem:
- **`xml-analysis-framework`** - Specialized XML document analysis
- **`docling-analysis-framework`** - PDF and Office documents (this package)
- **`document-analysis-framework`** - Text, code, and configuration files
- **`unified-analysis-orchestrator`** - Routes documents to appropriate frameworks
## 🔗 Integration with Docling
This framework acts as an **AI preparation layer** on top of Docling:
1. **Docling** handles the heavy lifting of document extraction
2. **Our framework** adds AI-specific analysis and chunking
3. **Result** is AI-ready structured data and optimized chunks
```python
# What Docling provides:
docling_result = docling.convert("document.pdf")
markdown_content = docling_result.document.export_to_markdown()
# What we add:
ai_analysis = our_analyzer.analyze(docling_result)
ai_chunks = our_chunker.chunk(markdown_content, ai_analysis)
quality_scores = our_quality_assessor.assess(docling_result)
```
## 📄 License
MIT License - see [LICENSE](LICENSE) file for details.
## 🙏 Acknowledgments
- Powered by [Docling](https://github.com/DS4SD/docling) for document extraction
- Built for modern AI/ML development workflows
- Part of the **AI Building Blocks** initiative
Raw data
{
"_id": null,
"home_page": "https://github.com/rdwj/docling-analysis-framework",
"name": "docling-analysis-framework",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": null,
"keywords": "docling, pdf, office, ai, ml, document-processing, semantic-search, chunking",
"author": "Wes Jackson",
"author_email": "AI Building Blocks <wjackson@redhat.com>",
"download_url": "https://files.pythonhosted.org/packages/00/3d/1eb48303ce2cd7630888aba9f7beb90575b7d3748f105fd527d33e98a37c/docling_analysis_framework-1.1.0.tar.gz",
"platform": null,
"description": "# Docling Analysis Framework\n\n[](https://www.python.org/downloads/)\n[](https://opensource.org/licenses/MIT)\n[](https://github.com/DS4SD/docling)\n[](https://github.com/rdwj/docling-analysis-framework)\n\nAI-ready analysis framework for **PDF and Office documents** using Docling for content extraction. Transforms Docling's output into optimized chunks and structured analysis for AI/ML pipelines.\n\n> **\ud83d\udcdd For text, code, and configuration files**, use our companion [document-analysis-framework](https://github.com/redhat-ai-americas/document-analysis-framework) which uses only Python standard library.\n\n## \ud83d\ude80 Quick Start\n\n### Simple API - Get Started in Seconds\n\n```python\nimport docling_analysis_framework as daf\n\n# \ud83c\udfaf One-line analysis with Docling extraction\nresult = daf.analyze(\"document.pdf\")\nprint(f\"Document type: {result['document_type'].type_name}\")\nprint(f\"Pages: {result['document_type'].pages}\")\nprint(f\"Handler used: {result['handler_used']}\")\n\n# \u2702\ufe0f Smart chunking for AI/ML\nchunks = daf.chunk(\"document.pdf\", strategy=\"auto\")\nprint(f\"Created {len(chunks)} optimized chunks\")\n\n# \ud83d\ude80 Enhanced analysis with both analysis and chunking\nenhanced = daf.analyze_enhanced(\"document.pdf\", chunking_strategy=\"structural\")\nprint(f\"Document: {enhanced['analysis']['document_type'].type_name}\")\nprint(f\"Chunks: {len(enhanced['chunks'])}\")\n\n# \ud83d\udcbe Save chunks to JSON\ndaf.save_chunks_to_json(chunks, \"chunks_output.json\")\n\n# \ud83d\udcbe Save analysis to JSON \ndaf.save_analysis_to_json(result, \"analysis_output.json\")\n```\n\n### Advanced Usage\n\n```python\nimport docling_analysis_framework as daf\n\n# Enhanced analysis with full results\nenhanced = daf.analyze_enhanced(\"research_paper.pdf\")\n\nprint(f\"Type: {enhanced['analysis']['document_type'].type_name}\")\nprint(f\"Confidence: {enhanced['analysis']['confidence']:.2f}\")\nprint(f\"AI use cases: {len(enhanced['analysis']['analysis'].ai_use_cases)}\")\nif enhanced['analysis']['analysis'].quality_metrics:\n for metric, score in enhanced['analysis']['analysis'].quality_metrics.items():\n print(f\"{metric}: {score:.2f}\")\n\n# Different chunking strategies\nhierarchical_chunks = daf.chunk(\"document.pdf\", strategy=\"structural\")\ntable_aware_chunks = daf.chunk(\"document.pdf\", strategy=\"table_aware\") \npage_aware_chunks = daf.chunk(\"document.pdf\", strategy=\"page_aware\")\n\n# Process chunks\nfor chunk in hierarchical_chunks:\n print(f\"Chunk {chunk.chunk_id}: {chunk.token_count} tokens\")\n print(f\"Type: {chunk.chunk_type}\")\n # Send to your AI model...\n\n# \ud83d\udcbe Save different strategies to separate files\nstrategies = {\n \"structural\": hierarchical_chunks,\n \"table_aware\": table_aware_chunks,\n \"page_aware\": page_aware_chunks\n}\n\nfor strategy_name, chunks in strategies.items():\n daf.save_chunks_to_json(chunks, f\"chunks_{strategy_name}.json\")\n print(f\"Saved {len(chunks)} chunks to chunks_{strategy_name}.json\")\n```\n\n### Expert Usage - Direct Class Access\n\n```python\n# For advanced customization, use the classes directly\nfrom docling_analysis_framework import DoclingAnalyzer, DoclingChunkingOrchestrator, ChunkingConfig\n\nanalyzer = DoclingAnalyzer(max_file_size_mb=100)\nresult = analyzer.analyze_document(\"document.pdf\")\n\n# Custom chunking with config\nconfig = ChunkingConfig(\n max_chunk_size=1500,\n min_chunk_size=200,\n overlap_size=100,\n preserve_structure=True\n)\n\norchestrator = DoclingChunkingOrchestrator(config=config)\n# Note: In advanced usage, you'd pass the actual docling_result object\n```\n\n## \ud83d\udccb Supported Document Types\n\n| Format | Extensions | Confidence | Powered By |\n|--------|------------|------------|------------|\n| **\ud83d\udcc4 PDF Documents** | .pdf | 95% | Docling extraction |\n| **\ud83d\udcdd Word Documents** | .docx | 95% | Docling extraction |\n| **\ud83d\udcca Spreadsheets** | .xlsx | 70% | Docling extraction |\n| **\ud83d\udcc5 Presentations** | .pptx | 70% | Docling extraction |\n| **\ud83d\uddbc\ufe0f Images with Text** | .png, .jpg, .tiff | 70% | Docling OCR |\n\n## \ud83c\udfaf Key Features\n\n### **\ud83d\udd0d Docling-Powered Extraction**\n- **PDF text extraction** - High-quality content extraction\n- **Table detection** - Preserves table structure in markdown\n- **Figure references** - Maintains image/figure relationships \n- **Header hierarchy** - Document structure preservation\n\n### **\ud83e\udd16 AI Preparation Layer**\n- **Quality assessment** - Extraction quality scoring\n- **Structure analysis** - Document type detection and analysis\n- **Chunk optimization** - AI-ready segmentation strategies\n- **Rich metadata** - Page counts, figures, tables, quality metrics\n\n### **\u26a1 Smart Chunking Strategies**\n- **Structural chunking** - Respects document hierarchy (headers, sections)\n- **Table-aware chunking** - Separates tables from text content\n- **Page-aware chunking** - Considers original page boundaries\n- **Auto-selection** - Document-type-aware strategy selection\n\n### **\ud83d\udcca Extraction Quality Analysis**\n- **Text coverage** - How much text was successfully extracted\n- **Structure preservation** - Whether headers, lists, tables are maintained\n- **Overall confidence** - Combined quality score for AI processing\n\n## \ud83c\udf89 Framework Status - COMPLETED\n\nThe Docling Analysis Framework is now **fully functional** and follows the same successful patterns as the XML Analysis Framework:\n\n### \u2705 Completed Features\n\n- **\ud83c\udfaf Simple API**: One-line functions for `analyze()`, `chunk()`, `analyze_enhanced()`\n- **\ud83d\udd27 Advanced API**: Direct class access for customization with `DoclingAnalyzer`, `DoclingChunkingOrchestrator`\n- **\u2699\ufe0f Configurable Chunking**: `ChunkingConfig` class for fine-tuning chunk parameters\n- **\ud83d\udce6 Multiple Strategies**: Structural, table-aware, page-aware, and auto-selection chunking\n- **\ud83d\udcbe JSON Export**: Easy export of analysis results and chunks to JSON files\n- **\ud83d\udee1\ufe0f Enhanced Error Handling**: Comprehensive logging and error reporting\n- **\ud83d\udcca Quality Metrics**: Extraction quality assessment and content analysis\n- **\ud83e\uddea Testing Framework**: Complete Jupyter notebook for validation\n- **\ud83d\udcda Documentation**: Comprehensive README with examples and usage patterns\n\n### \ud83d\ude80 Ready for AI/ML Integration\n\nThe framework provides everything needed for AI/ML pipelines:\n\n- **Token-optimized chunks** sized for LLM context windows\n- **Rich metadata** with document structure and quality metrics \n- **JSON export** for vector database ingestion\n- **Multiple chunking strategies** for different document types\n- **Quality assessment** to determine AI suitability\n\n### \ud83d\udd04 Comparison with XML Framework\n\n| Feature | XML Framework | Docling Framework | Status |\n|---------|---------------|-------------------|---------|\n| Simple API | \u2705 | \u2705 | Complete |\n| Advanced Classes | \u2705 | \u2705 | Complete |\n| Multiple Strategies | \u2705 | \u2705 | Complete |\n| Configuration | \u2705 | \u2705 | Complete |\n| JSON Export | \u2705 | \u2705 | Complete |\n| Error Handling | \u2705 | \u2705 | Complete |\n| Testing Notebooks | \u2705 | \u2705 | Complete |\n| Quality Metrics | \u2705 | \u2705 | Complete |\n\nBoth frameworks now provide identical API patterns and functionality!\n\n## \ud83d\udd27 Installation\n\n```bash\n# Install Docling first\npip install docling\n\n# Install framework\ngit clone https://github.com/rdwj/docling-analysis-framework.git\ncd docling-analysis-framework \npip install -e .\n```\n\n## \ud83d\udcd6 Usage Examples\n\n### Document Analysis with Quality Assessment\n```python\nfrom core.analyzer import DoclingAnalyzer\n\nanalyzer = DoclingAnalyzer()\nresult = analyzer.analyze_document(\"contract.pdf\")\n\n# Access extraction quality\nquality = result['analysis'].extraction_quality\nprint(f\"Text coverage: {quality['text_coverage']:.2f}\")\nprint(f\"Structure score: {quality['structure_score']:.2f}\")\nprint(f\"Overall quality: {quality['overall_score']:.2f}\")\n\n# Access document insights\nfindings = result['analysis'].key_findings\nprint(f\"Pages: {findings['page_count']}\")\nprint(f\"Tables: {findings['table_rows']}\")\nprint(f\"Figures: {findings['figure_count']}\")\n```\n\n### Advanced Chunking for Academic Papers\n```python\n# Perfect for research papers with complex structure\nchunks = orchestrator.chunk_document(\n \"paper.pdf\",\n markdown_content,\n docling_result,\n strategy='structural'\n)\n\n# Chunks respect paper structure\nfor chunk in chunks:\n if chunk.chunk_type == 'section':\n print(f\"Section: {chunk.metadata['section_title']}\")\n elif chunk.chunk_type == 'table':\n print(f\"Table data: {len(chunk.content)} chars\")\n elif chunk.chunk_type == 'figure':\n print(f\"Figure reference preserved\")\n```\n\n### File Size Management\n```python\nfrom core.analyzer import DoclingAnalyzer\n\n# Large PDF processing with limits\nanalyzer = DoclingAnalyzer(max_file_size_mb=100.0)\n\nresult = analyzer.analyze_document(\"large_manual.pdf\")\nif 'error' in result:\n print(f\"File too large: {result['error']}\")\nelse:\n print(f\"Successfully processed {result['file_size']} bytes\")\n```\n\n## \ud83e\uddea Framework Ecosystem\n\nThis framework is part of a larger document analysis ecosystem:\n\n- **`xml-analysis-framework`** - Specialized XML document analysis\n- **`docling-analysis-framework`** - PDF and Office documents (this package)\n- **`document-analysis-framework`** - Text, code, and configuration files \n- **`unified-analysis-orchestrator`** - Routes documents to appropriate frameworks\n\n## \ud83d\udd17 Integration with Docling\n\nThis framework acts as an **AI preparation layer** on top of Docling:\n\n1. **Docling** handles the heavy lifting of document extraction\n2. **Our framework** adds AI-specific analysis and chunking\n3. **Result** is AI-ready structured data and optimized chunks\n\n```python\n# What Docling provides:\ndocling_result = docling.convert(\"document.pdf\")\nmarkdown_content = docling_result.document.export_to_markdown()\n\n# What we add:\nai_analysis = our_analyzer.analyze(docling_result)\nai_chunks = our_chunker.chunk(markdown_content, ai_analysis)\nquality_scores = our_quality_assessor.assess(docling_result)\n```\n\n## \ud83d\udcc4 License\n\nMIT License - see [LICENSE](LICENSE) file for details.\n\n## \ud83d\ude4f Acknowledgments\n\n- Powered by [Docling](https://github.com/DS4SD/docling) for document extraction\n- Built for modern AI/ML development workflows\n- Part of the **AI Building Blocks** initiative\n",
"bugtrack_url": null,
"license": "MIT",
"summary": "AI-ready analysis framework for PDF and Office documents using Docling for content extraction",
"version": "1.1.0",
"project_urls": {
"Documentation": "https://github.com/rdwj/docling-analysis-framework/blob/main/README.md",
"Homepage": "https://github.com/rdwj/docling-analysis-framework",
"Issues": "https://github.com/rdwj/docling-analysis-framework/issues",
"Repository": "https://github.com/rdwj/docling-analysis-framework"
},
"split_keywords": [
"docling",
" pdf",
" office",
" ai",
" ml",
" document-processing",
" semantic-search",
" chunking"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "5c47e3b29bb62616ae5f09335d09a8723bbebff020155d30f2aff8c03e7e81dd",
"md5": "8898f4a0c35d2c8761b5f9a551f0a76f",
"sha256": "0f2fd86091bdb9233ca2744957d0e18f38bca32491f46aef2feda7d3497d12f3"
},
"downloads": -1,
"filename": "docling_analysis_framework-1.1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "8898f4a0c35d2c8761b5f9a551f0a76f",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 21187,
"upload_time": "2025-07-29T14:33:56",
"upload_time_iso_8601": "2025-07-29T14:33:56.494383Z",
"url": "https://files.pythonhosted.org/packages/5c/47/e3b29bb62616ae5f09335d09a8723bbebff020155d30f2aff8c03e7e81dd/docling_analysis_framework-1.1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "003d1eb48303ce2cd7630888aba9f7beb90575b7d3748f105fd527d33e98a37c",
"md5": "f37327ff007f083af5519dd56e1977d7",
"sha256": "9d8d4de0821326cab9994c452a2aad143d66df473b14aef691ba0caac50271a7"
},
"downloads": -1,
"filename": "docling_analysis_framework-1.1.0.tar.gz",
"has_sig": false,
"md5_digest": "f37327ff007f083af5519dd56e1977d7",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 9259597,
"upload_time": "2025-07-29T14:34:10",
"upload_time_iso_8601": "2025-07-29T14:34:10.442370Z",
"url": "https://files.pythonhosted.org/packages/00/3d/1eb48303ce2cd7630888aba9f7beb90575b7d3748f105fd527d33e98a37c/docling_analysis_framework-1.1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-29 14:34:10",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "rdwj",
"github_project": "docling-analysis-framework",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"requirements": [
{
"name": "docling",
"specs": [
[
">=",
"1.0.0"
]
]
}
],
"lcname": "docling-analysis-framework"
}
JSON
