Update app.py

app.py

@@ -30,32 +30,40 @@ prompt = PromptTemplate(template=template, input_variables=["context", "question
 # upload PDF
 pdf_file = st.file_uploader("Upload your pdf",type="pdf")
 question = st.text_input("Ask a question:")
-if pdf_file is not None:  
-    # Load and process the PDF
-    loader = PyPDFLoader(pdf_file)
-    pdf_data = loader.load()
-    
-    # Split the text into chunks
-    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
-    docs = text_splitter.split_documents(pdf_data)
-    
-    # Create a Chroma vector store
-    embeddings = HuggingFaceEmbeddings(model_name="embaas/sentence-transformers-multilingual-e5-base")
-    db = Chroma.from_documents(docs, embeddings)
-    
-    # Initialize message history for conversation
-    message_history = ChatMessageHistory()
-    
-    # Memory for conversational context
-    memory = ConversationBufferMemory(
-           memory_key="chat_history",
-           output_key="answer",
-           chat_memory=message_history,
-           return_messages=True,
-       )
-    
-    # Create a chain that uses the Chroma vector store
-    chain = ConversationalRetrievalChain.from_llm(
+if pdf_file is not None:
+    try:
+        # Save the uploaded file to a temporary file
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as temp_file:
+            temp_file.write(pdf_file.read())
+            temp_file_path = temp_file.name
+        
+        # Load and process the PDF
+        loader = PyPDFLoader(temp_file_path)
+        pdf_data = loader.load()
+        st.write(f"Loaded {len(pdf_data)} documents from PDF.")
+        
+        # Split the text into chunks
+        text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+        docs = text_splitter.split_documents(pdf_data)
+        st.write(f"Split text into {len(docs)} chunks.")
+        
+        # Create a Chroma vector store
+        embeddings = HuggingFaceEmbeddings(model_name="embaas/sentence-transformers-multilingual-e5-base")
+        db = Chroma.from_documents(docs, embeddings)
+        
+        # Initialize message history for conversation
+        message_history = ChatMessageHistory()
+        
+        # Memory for conversational context
+        memory = ConversationBufferMemory(
+            memory_key="chat_history",
+            output_key="answer",
+            chat_memory=message_history,
+            return_messages=True,
+        )
+        
+        # Create a chain that uses the Chroma vector store
+        chain = ConversationalRetrievalChain.from_llm(
             llm=llm,
             chain_type="stuff",
             retriever=db.as_retriever(),
@@ -63,10 +71,52 @@ if pdf_file is not None:
             return_source_documents=False,
             combine_docs_chain_kwargs={'prompt': prompt}
         )
-    if question:
-        # Process the question
-        res = chain({"question": question})
-        answer = res["answer"]
-        st.write(f"{answer}")
+        
+        if question:
+            # Process the question
+            res = chain({"question": question})
+            answer = res.get("answer", "No answer found.")
+            st.write(f"Answer: {answer}")
+
+    except Exception as e:
+        st.error(f"An error occurred: {e}")
+# if pdf_file is not None:  
+#     # Load and process the PDF
+#     loader = PyPDFLoader(pdf_file)
+#     pdf_data = loader.load()
+    
+#     # Split the text into chunks
+#     text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+#     docs = text_splitter.split_documents(pdf_data)
+    
+#     # Create a Chroma vector store
+#     embeddings = HuggingFaceEmbeddings(model_name="embaas/sentence-transformers-multilingual-e5-base")
+#     db = Chroma.from_documents(docs, embeddings)
+    
+#     # Initialize message history for conversation
+#     message_history = ChatMessageHistory()
+    
+#     # Memory for conversational context
+#     memory = ConversationBufferMemory(
+#            memory_key="chat_history",
+#            output_key="answer",
+#            chat_memory=message_history,
+#            return_messages=True,
+#        )
+    
+#     # Create a chain that uses the Chroma vector store
+#     chain = ConversationalRetrievalChain.from_llm(
+#             llm=llm,
+#             chain_type="stuff",
+#             retriever=db.as_retriever(),
+#             memory=memory,
+#             return_source_documents=False,
+#             combine_docs_chain_kwargs={'prompt': prompt}
+#         )
+#     if question:
+#         # Process the question
+#         res = chain({"question": question})
+#         answer = res["answer"]
+#         st.write(f"{answer}")