app.py

import os
from utils.pdf_processing import extract_text_from_pdf, extract_text_from_scanned_pdf
from utils.vector_store import add_document, search
from utils.stock_data import get_stock_data
from utils.sentiment import analyze_sentiment

# Define PDF folder
PDF_FOLDER = "data"

# Process and index all PDFs
def process_pdfs():
    for pdf_file in os.listdir(PDF_FOLDER):
        pdf_path = os.path.join(PDF_FOLDER, pdf_file)
        
        # Extract text from normal or scanned PDFs
        text = extract_text_from_pdf(pdf_path) or extract_text_from_scanned_pdf(pdf_path)
        
        # Store in FAISS vector store
        add_document(text, pdf_file)

# Query the system
def query_system(query):
    results = search(query)

    print("📌 Top Matching Documents:")
    for res in results:
        print(f"\n📄 Document: {res.metadata['id']}\n{text[:500]}...")

    # Perform sentiment analysis
    print("\n📊 Sentiment Analysis:")
    for res in results:
        sentiment = analyze_sentiment(res.page_content)
        print(f"📄 Document: {res.metadata['id']}, Sentiment: {sentiment}")

# Fetch stock data
def get_stock_info(ticker):
    stock_data = get_stock_data(ticker)
    print(f"\n📈 Stock Price for {ticker}: ${stock_data['price']}")

# Run the application
if __name__ == "__main__":
    print("🔄 Indexing PDFs...")
    process_pdfs()

    query = input("\n🔍 Enter your query: ")
    query_system(query)

    ticker = input("\n📊 Enter a stock ticker (e.g., AAPL, TSLA): ")
    get_stock_info(ticker)