documentation.md

📋 Document Finder Backend API

A comprehensive REST API for searching and retrieving technical documents and specifications from 3GPP, ETSI and GlobalPlatform organizations.

🚀 Key Features

• Document Retrieval: Get direct download URLs and metadata for specific documents
• Batch Processing: Handle multiple document requests simultaneously
• Advanced Search: Multiple search modes with keyword matching
• BM25 Scoring: State-of-the-art relevance ranking using BM25 algorithm
• Cross-Organization: Search across both 3GPP and ETSI document repositories

📚 Supported Document Types

3GPP Documents

• TDocs (Technical Documents):
  • Format: S1-123456, C4-234567, R2-345678
  • Working group documents from SA, CT, RAN groups
• Technical Specifications:
  • Format: 23.401, 38.331-16
  • Official published specifications

ETSI Documents

• TDocs (Technical Documents):
  • Format: SET(25)000001, SCPTEQ(19)000011
  • Committee working documents
• Technical Specifications:
  • Format: 131 102, 188 008-2
  • Published ETSI standards

GP Documents

• Specifications
  • Format: GPC_SPE_106, GP_REQ_004
  • Published Global Platform specifications

🔍 Search Capabilities

• Quick Search: Lightning-fast metadata-only search
• Deep Search: Comprehensive content-based search within document sections
• BM25 Search: Advanced relevance scoring with normalization
• Flexible Filtering: By source organization, document type, and specification category

🛡️ Data Sources

This API indexes and searches through:

• 3GPP specification metadata and content (from the Datasets section)
• ETSI specification metadata and content (from the Datasets section)
• 3GPP TDoc location mappings (from the Datasets section)
• GP TDoc location mappings (from Python app.py)
• Pre-built BM25 search indices (from the Models section)

🔧 Technical Stack

Built with FastAPI, featuring automatic OpenAPI documentation, request validation, and comprehensive error handling.