Update app.py

app.py

@@ -84,53 +84,59 @@ def export_conversation_to_pdf(conversation_history):
 
 def main():
     st.set_page_config(page_title="Gemini Chat QA", layout="wide")
-    st.title("📄 Chat with Your Document using Gemini (Sayandip+Dhiraj+Fazal)")
+    st.title("📄 Chat with Your Documents using Gemini (Sayandip+Dhiraj+Fazal)")
 
     # Initialize session state attributes if they don't exist
     if "conversation" not in st.session_state:
         st.session_state.conversation = []
-    if "document_text" not in st.session_state:
-        st.session_state.document_text = ""
+    if "documents_text" not in st.session_state:
+        st.session_state.documents_text = []
     if "chat_active" not in st.session_state:
         st.session_state.chat_active = True
     if "chat_history" not in st.session_state:  # Initialize chat_history
         st.session_state.chat_history = []
 
-    uploaded_file = st.file_uploader(
-        "Upload a file (.txt, .docx, .csv, .xlsx, .pptx, .pdf, .html, .htm, .tex, .jpg, .jpeg, .png):",
-        type=["txt", "docx", "csv", "xlsx", "pptx", "pdf", "html", "htm", "tex", "jpg", "jpeg", "png"]
+    uploaded_files = st.file_uploader(
+        "Upload files (.txt, .docx, .csv, .xlsx, .pptx, .pdf, .html, .htm, .tex, .jpg, .jpeg, .png):",
+        type=["txt", "docx", "csv", "xlsx", "pptx", "pdf", "html", "htm", "tex", "jpg", "jpeg", "png"],
+        accept_multiple_files=True
     )
 
-    if uploaded_file and not st.session_state.document_text:
-        filetype = uploaded_file.name.split(".")[-1].lower()
-        try:
-            if filetype == "txt":
-                st.session_state.document_text = uploaded_file.read().decode("utf-8")
-            elif filetype == "docx":
-                st.session_state.document_text = extract_text_from_docx(uploaded_file)
-            elif filetype == "csv":
-                st.session_state.document_text = extract_text_from_csv(uploaded_file)
-            elif filetype == "xlsx":
-                st.session_state.document_text = extract_text_from_xlsx(uploaded_file)
-            elif filetype == "pptx":
-                st.session_state.document_text = extract_text_from_pptx(uploaded_file)
-            elif filetype == "pdf":
-                st.session_state.document_text = extract_text_from_pdf(uploaded_file)
-            elif filetype in ["html", "htm"]:
-                st.session_state.document_text = extract_text_from_html(uploaded_file)
-            elif filetype == "tex":
-                st.session_state.document_text = extract_text_from_tex(uploaded_file)
-            elif filetype in ["jpg", "jpeg", "png"]:
-                image_data = process_image(uploaded_file)
-                st.session_state.document_text = "Image uploaded and processed."
-                st.image(uploaded_file, caption=uploaded_file.name, use_container_width=True)
-                st.session_state.image_data = image_data  # Store the image data for later use
-            else:
-                st.error("Unsupported file format.")
-        except Exception as e:
-            st.error(f"Failed to extract text: {e}")
-
-    if st.session_state.document_text:
+    # Process each uploaded file
+    if uploaded_files:
+        all_text_content = ""
+        for uploaded_file in uploaded_files:
+            filetype = uploaded_file.name.split(".")[-1].lower()
+            try:
+                if filetype == "txt":
+                    all_text_content += uploaded_file.read().decode("utf-8") + "\n"
+                elif filetype == "docx":
+                    all_text_content += extract_text_from_docx(uploaded_file) + "\n"
+                elif filetype == "csv":
+                    all_text_content += extract_text_from_csv(uploaded_file) + "\n"
+                elif filetype == "xlsx":
+                    all_text_content += extract_text_from_xlsx(uploaded_file) + "\n"
+                elif filetype == "pptx":
+                    all_text_content += extract_text_from_pptx(uploaded_file) + "\n"
+                elif filetype == "pdf":
+                    all_text_content += extract_text_from_pdf(uploaded_file) + "\n"
+                elif filetype in ["html", "htm"]:
+                    all_text_content += extract_text_from_html(uploaded_file) + "\n"
+                elif filetype == "tex":
+                    all_text_content += extract_text_from_tex(uploaded_file) + "\n"
+                elif filetype in ["jpg", "jpeg", "png"]:
+                    image_data = process_image(uploaded_file)
+                    st.session_state.image_data = image_data  # Store the image data for later use
+                    all_text_content += "Image uploaded and processed.\n"
+                    st.image(uploaded_file, caption=uploaded_file.name, use_container_width=True)
+                else:
+                    st.error(f"Unsupported file format: {uploaded_file.name}")
+            except Exception as e:
+                st.error(f"Failed to extract text from {uploaded_file.name}: {e}")
+        
+        st.session_state.documents_text = all_text_content  # Store the compiled text from all documents
+
+    if st.session_state.documents_text:
         st.markdown("### 💬 Conversation")
         for user_q, gemini_a in st.session_state.conversation:
             st.markdown(f"**You:** {user_q}")
@@ -138,7 +144,7 @@ def main():
 
         if st.session_state.chat_active:
             with st.form(key="chat_form", clear_on_submit=True):
-                user_input = st.text_input("Ask a question about the document (type 'exit' to stop):")
+                user_input = st.text_input("Ask a question about the documents (type 'exit' to stop):")
                 submit = st.form_submit_button("Send")
 
                 if submit and user_input:
@@ -157,9 +163,9 @@ def main():
                                 )
                                 content_blocks.append({"text": f"Previous conversation:\n{history_text}"})
 
-                            # Add file text content
-                            if st.session_state.document_text.strip():
-                                content_blocks.append({"text": f"Context:\n{st.session_state.document_text}"})
+                            # Add all documents' text content
+                            if st.session_state.documents_text.strip():
+                                content_blocks.append({"text": f"Context:\n{st.session_state.documents_text}"})
 
                             # Add image blocks (if image data exists)
                             if "image_data" in st.session_state: