Version5

app.py

@@ -3,11 +3,24 @@ from sentence_transformers import SentenceTransformer
 import faiss
 import re
 import gradio as gr
+import PyPDF2
+import io
 
-import numpy as np
-from sentence_transformers import SentenceTransformer
-import faiss
-import re
+def extract_text_from_pdf(pdf_file):
+    """
+    Extract text from a PDF file
+    """
+    if pdf_file is None:
+        return "Please upload a PDF file."
+    
+    pdf_text = ""
+    pdf_reader = PyPDF2.PdfReader(io.BytesIO(pdf_file))
+    
+    for page_num in range(len(pdf_reader.pages)):
+        page = pdf_reader.pages[page_num]
+        pdf_text += page.extract_text() + "\n"
+    
+    return pdf_text
 
 def preprocess_text(text):
     """
@@ -110,8 +123,45 @@ def ask_question(question, model, index, text_chunks):
         print(result['full_text'])
         print(f"Best match confidence: {result['confidence']:.2f}")
         return result
-# Initialize the system
-model, index, text_chunks = create_qa_system(pdf_text)
+
+# Global variables to store model, index, and text chunks
+global_model = None
+global_index = None
+global_text_chunks = None
+
+def upload_file(file):
+    global global_model, global_index, global_text_chunks
+    if file is not None:
+        try:
+            # Extract text from PDF
+            pdf_text = extract_text_from_pdf(file)
+            
+            # Initialize QA system
+            global_model, global_index, global_text_chunks = create_qa_system(pdf_text)
+            
+            return "✅ Document processed successfully! You can now ask questions."
+        except Exception as e:
+            return f"❌ Error processing document: {str(e)}"
+    else:
+        return "❌ Please upload a PDF file."
+
+def answer_question(question):
+    global global_model, global_index, global_text_chunks
+    
+    if global_model is None or global_index is None or global_text_chunks is None:
+        return "Please upload and process a document first."
+    
+    if not question.strip():
+        return "Please enter a question."
+    
+    result = query_qa_system(question, global_model, global_index, global_text_chunks)
+    
+    if result['found_answer']:
+        response = f"Found matching section (confidence: {result['confidence']:.2f}):\n\n{result['full_text']}"
+    else:
+        response = f"{result['full_text']}\nBest match confidence: {result['confidence']:.2f}"
+    
+    return response
 
 # Custom CSS for professional styling
 custom_css = """
@@ -262,7 +312,7 @@ with gr.Blocks(title="Interview Q&A Assistant", css=custom_css) as demo:
         </div>
         """)
     
-    # Set up events (keeping the same functionality)
+    # Set up events
     upload_button.click(upload_file, inputs=pdf_upload, outputs=status_text)
     submit_button.click(answer_question, inputs=question_input, outputs=answer_output)