v0.2.0: Amazon Neptune, Docling Integration & Robust Extraction

[KaifAhmad1]

🚀 Semantica v0.2.0 Release Notes

Date: 2026-01-10
Tag: v0.2.0

🌟 Highlights

This release represents a major step forward in enterprise readiness and robustness. Key additions include native Amazon Neptune support for scalable graph storage, Docling integration for high-fidelity document parsing, and a new Robust Extraction system that ensures reliable data extraction even in edge cases.

🌊 Amazon Neptune Support

We've added a production-ready AmazonNeptuneStore that integrates seamlessly with AWS.

• Native SigV4 Authentication: Secure connection handling with automatic token refresh.
• Resilient Connectivity: Built-in retry logic with backoff for transient network issues.
• Full GraphStore Compatibility: Passes the complete compliance test suite.

📄 Docling Integration

Parsing unstructured documents just got a major upgrade with the new DoclingParser.

• Multi-Format Support: High-fidelity parsing for PDF, DOCX, PPTX, XLSX, HTML, and Images.
• Structure Preservation: Superior table extraction and layout understanding.
• Flexible Export: Convert documents to Markdown, JSON, or HTML with ease.

🛡️ Robust Extraction Fallbacks

No more empty results. We've implemented a smart fallback chain:

1. ML/LLM: High-precision extraction.
2. Pattern: Regex and rule-based matching.
3. Last Resort: Heuristic fallback (e.g., capitalized words) to guarantee output.

---

✨ New Features

Provenance & Tracking

• Traceability: Added batch_index and document_id metadata to all extracted entities and relations.
• Observability: Enhanced logging with count tracking for batch processes.

Semantic Extraction

• Auto-Chunking: Automatically handles long text inputs for LLM extraction.
• Resilient Providers: Added retry logic (3 attempts w/ backoff) and robust JSON parsing for all LLM calls.
• Groq Enhancements: Better diagnostics and connectivity testing for Groq users.

🛠️ Technical Improvements

• Schema Standardization: Introduced canonical Pydantic models in semantica/semantic_extract/schemas.py.
• Testing: Added comprehensive test suites for robustness, model switching, and end-to-end pipelines.
• Dependencies: Added GitPython and chardet for better environment stability.

---

🐛 Bug Fixes & Refactoring

• Critical: Fixed issue where extractors returned empty lists on primary method failure.
• Critical: Resolved NameError in extraction_validator.py.
• Embeddings: Fixed model switching bug in TextEmbedder to ensure state clears correctly.
• Graph Analysis: Fixed TypeError when processing graphs with raw Entity objects.
• Cleanup: Removed legacy deduplication logic in favor of the unified semantica/conflicts module.

---

📦 Installation & Upgrade

To install the latest version:

pip install semantica==0.2.0

To include Amazon Neptune support:

pip install semantica[graph-amazon-neptune]

To include Docling support:

pip install semantica[docling]

---

Thank you to all our contributors! If you encounter any issues, please report them on GitHub Issues.

---

This discussion was created from the release v0.2.0: Amazon Neptune, Docling Integration & Robust Extraction.