tests/test_pdf_analyzer.py

#!/usr/bin/env python3
"""
Test script for PDF Analyzer Agent
"""

import asyncio
import os
from dotenv import load_dotenv
from langchain_openai import ChatOpenAI
from mistralai import Mistral

from agents.pdf_analyzer import PDFAnalyzerAgent

load_dotenv()


async def main():
    """Test the PDF analyzer agent"""
    
    # Initialize LLM (Cerebras via OpenAI-compatible endpoint)
    if os.getenv("OPENAI_API_KEY"):
        llm = ChatOpenAI(
            model=os.getenv("LLM_MODEL", "gpt-5-nano-2025-08-07"),
            reasoning_effort="low",
            api_key=os.getenv("OPENAI_API_KEY"),
            base_url=os.getenv("LLM_BINDING_HOST", "https://api.openai.com/v1"),
            default_headers={"X-Cerebras-3rd-Party-Integration": "langgraph"}
        )
    else:
        print("❌ No API key found. Please set OPENAI_API_KEY in .env file")
        return
    
    # Initialize Mistral client for OCR (optional)
    mistral_client = None
    if os.getenv("MISTRAL_API_KEY"):
        mistral_client = Mistral(api_key=os.getenv("MISTRAL_API_KEY"))
    # Create PDF analyzer agent
    agent = PDFAnalyzerAgent(llm, mistral_client=mistral_client)
    
    # Provide path to your PDF file
    pdf_path = "PublicWaterMassMailing.pdf"
    
    # Check if file exists
    if not os.path.exists(pdf_path):
        print(f"\n❌ PDF file not found: {pdf_path}")
        return
    
    print(f"\n{'='*80}")
    print(f"PDF ANALYSIS FOR: {pdf_path}")
    print(f"{'='*80}\n")
    
    try:
        # Analyze the PDF
        result = await agent.analyze_pdf(pdf_path)
        
        # Display results
        if result.get("actors"):
            print(f"\n{'─'*80}")
            print("👥 ACTORS INVOLVED")
            print(f"{'─'*80}")
            print(result["actors"])
        
        if result.get("key_details"):
            print(f"\n{'─'*80}")
            print("🔑 KEY DETAILS")
            print(f"{'─'*80}")
            print(result["key_details"])
        
        if result.get("summary"):
            print(f"\n{'─'*80}")
            print("📄 HIGH-LEVEL SUMMARY")
            print(f"{'─'*80}")
            print(result["summary"])
        
        print(f"\n{'='*80}")
        print(f"✅ ANALYSIS COMPLETE - Status: {result.get('processing_status')}")
        if result.get("ocr_used"):
            print(f"📷 OCR was used: {result.get('ocr_method')}")
        print(f"{'='*80}\n")
        
    except Exception as e:
        print(f"\n❌ Error during PDF analysis: {e}")
        import traceback
        traceback.print_exc()


if __name__ == "__main__":
    asyncio.run(main())