Update app.py

app.py

@@ -1,49 +1,76 @@
 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):
+import gradio as gr
+from extract_pdf import extract_text_from_pdf, extract_text_from_scanned_pdf
+from vector_store import add_document, search
+from stock_data import get_stock_data
+from sentiment import analyze_sentiment
+
+# Create folders if not exist
+DATA_FOLDER = "data"
+FAISS_FOLDER = "faiss_index"
+os.makedirs(DATA_FOLDER, exist_ok=True)
+os.makedirs(FAISS_FOLDER, exist_ok=True)
+
+
+# 1️⃣ Process uploaded PDF and add to FAISS
+def process_pdf(file):
+    """Process uploaded PDF and add to FAISS"""
+    file_path = os.path.join(DATA_FOLDER, file.name)
+    with open(file_path, "wb") as f:
+        f.write(file.read())
+
+    # Extract text using available methods
+    text = extract_text_from_pdf(file_path) or extract_text_from_scanned_pdf(file_path)
+    add_document(text, file.name)
+    return f"✅ PDF '{file.name}' indexed successfully!"
+
+
+# 2️⃣ Search FAISS for relevant PDFs
+def query_pdf(query):
+    """Search FAISS index for relevant documents."""
     results = search(query)
 
-    print("📌 Top Matching Documents:")
+    response = "📌 **Top Matching Documents:**\n\n"
     for res in results:
-        print(f"\n📄 Document: {res.metadata['id']}\n{text[:500]}...")
+        response += f"📄 **{res.metadata['id']}**\n{text[:500]}...\n\n"
 
-    # Perform sentiment analysis
-    print("\n📊 Sentiment Analysis:")
+    # Sentiment analysis
+    response += "\n📊 **Sentiment Analysis:**\n"
     for res in results:
         sentiment = analyze_sentiment(res.page_content)
-        print(f"📄 Document: {res.metadata['id']}, Sentiment: {sentiment}")
+        response += f"📄 **{res.metadata['id']}** → Sentiment: {sentiment}\n"
 
-# Fetch stock data
+    return response
+
+
+# 3️⃣ Fetch Live Stock Market Data
 def get_stock_info(ticker):
+    """Fetch real-time stock price from Yahoo Finance."""
     stock_data = get_stock_data(ticker)
-    print(f"\n📈 Stock Price for {ticker}: ${stock_data['price']}")
+    return f"\n📈 Stock Price for {ticker}: ${stock_data['price']}"
+
+
+# Gradio UI
+with gr.Blocks() as demo:
+    gr.Markdown("# Stock Analysis PDF Indexer")
+
+    # File Upload for PDFs
+    pdf_upload = gr.File(label="Upload PDFs", file_count="multiple")
+    upload_button = gr.Button("Upload and Index PDF")
+    upload_message = gr.Textbox(label="Upload Status", interactive=False)
+
+    # PDF Query Input
+    query_input = gr.Textbox(label="Query PDFs")
+    query_output = gr.Textbox(label="Query Results", interactive=False)
 
-# Run the application
-if __name__ == "__main__":
-    print("🔄 Indexing PDFs...")
-    process_pdfs()
+    # Stock Ticker Input
+    stock_input = gr.Textbox(label="Enter Stock Ticker (e.g., AAPL)")
+    stock_output = gr.Textbox(label="Stock Data", interactive=False)
 
-    query = input("\n🔍 Enter your query: ")
-    query_system(query)
+    # Actions
+    upload_button.click(process_pdf, inputs=pdf_upload, outputs=upload_message)
+    query_input.submit(query_pdf, inputs=query_input, outputs=query_output)
+    stock_input.submit(get_stock_info, inputs=stock_input, outputs=stock_output)
 
-    ticker = input("\n📊 Enter a stock ticker (e.g., AAPL, TSLA): ")
-    get_stock_info(ticker)
+# Launch the Gradio interface
+demo.launch()