Robert-kelly19/237law

A whatsapp chat board for cameroon law explaination # Cameroonian Law RAG System A production-ready NestJS backend that implements a legal RAG (Retrieval Augmented Generation) system for Cameroonian law. ## Features - **PDF Ingestion**: Upload PDF law books (Penal Code, Criminal Procedure Code, Constitution) - **Text Extraction**: Extracts text from PDF using pdf-parse - **Structured Parsing**: Splits text into Articles and Sections - **Smart Chunking**: Chunks long articles into 800-word pieces for optimal embedding - **Vector Storage**: Stores OpenAI embeddings in PostgreSQL using pgvector - **Similarity Search**: Cosine similarity search using pgvector - **AI Answers**: RAG-powered question answering that prevents hallucination ## Tech Stack - **NestJS** - Node.js framework - **PostgreSQL** - Database - **pgvector** - Vector similarity search - **Prisma ORM** - Database access - **OpenAI API** - Embeddings and chat completion - **pdf-parse** - PDF text extraction - **class-validator** - DTO validation ## Prerequisites 1. Node.js 18+ 2. PostgreSQL with pgvector extension 3. OpenAI API key ## Setup ### 1. Install Dependencies ```bash npm install ``` ### 2. Configure Environment Copy `.env.example` to `.env` and fill in your values: ```env DATABASE_URL="postgresql://postgres:password@localhost:5432/cm_law?schema=public" OPENAI_API_KEY="your-openai-api-key" PORT=3000 NODE_ENV="development" ``` ### 3. Set Up Database Run the pgvector migration: ```bash psql -U postgres -d cm_law -f prisma/migrations/001_enable_pgvector.sql ``` ### 4. Generate Prisma Client ```bash npm run prisma:generate ``` ### 5. Run the Application ```bash npm run start:dev ``` The API will be available at `localhost` ## API Endpoints ### 1. Ingest PDF Upload a PDF law document. ```bash curl -X POST localhost \ -F "file=@./penal-code.pdf" \ -F "lawName=Penal Code" ``` ### 2. Ingest Text Ingest raw text content directly. ```bash curl -X POST localhost \ -H "C …