Update app.py

app.py

@@ -1,168 +1,329 @@
 import openai
-import streamlit as st
 from openai import OpenAI
 import io
 import time
 import os
 from dotenv import load_dotenv
 
-# OpenAI keys
-OPENAI_API_KEY = st.secrets["OPENAI_API_KEY"]
-# VECTOR_STORE_ID = st.secrets["VECTOR_STORE_ID"]
-# VECTOR_STORE_ID = os.getenv("VECTOR_STORE_ID")
+# Load environment variables
+load_dotenv()
+
+# Securely fetch environment variables for Hugging Face Spaces
+def get_env_variable(name: str) -> str:
+    value = os.environ.get(name)
+    if value is None:
+        raise ValueError(f"Environment variable '{name}' is not set.")
+    return value
+
+# Initialize OpenAI client
+OPENAI_API_KEY = get_env_variable("OPENAI_API_KEY")
+VECTOR_STORE_ID = get_env_variable("VECTOR_STORE_ID")
+ASSISTANT_ID = get_env_variable("ASSISTANT_ID")
 
 client = OpenAI(api_key=OPENAI_API_KEY)
-# vector_store_id =VECTOR_STORE_ID  # Vector Store ID to use
-vector_store_id = "vs_sHLBqA88ng66RDZYMgaBmDuX"
-st.write(vector_store_id)
-# all_files = list(client.beta.vector_stores.files.list(vector_store_id))
-# for file in all_files:
-#   # print(file)
-#   file_id = file.id
-#   st.write(file_id)
-
-# Set the assistant ID
-assistant_id = "asst_UeIHpkpVejAgrIz3RnisxSLm"  # Replace with your own assistant ID
-
-def ensure_single_thread_id():
-    if "thread_id" not in st.session_state:
+
+# Utility Functions
+def ensure_single_thread_id(session_state):
+    if "thread_id" not in session_state:
         thread = client.beta.threads.create()
-        st.session_state.thread_id = thread.id
-    return st.session_state.thread_id
+        session_state.thread_id = thread.id
+    return session_state.thread_id
+
 
 def safe_message_send(prompt, thread_id):
     try:
-        message = client.beta.threads.messages.create(
+        return client.beta.threads.messages.create(
             thread_id=thread_id,
             role="user",
             content=prompt
         )
-        return message
     except Exception as e:
         if "active" in str(e):
-            print("Waiting for the current run to finish...")
-            time.sleep(1)  # wait a bit before retrying
-            return safe_message_send(prompt, thread_id)  # retry sending the message
+            time.sleep(1)
+            return safe_message_send(prompt, thread_id)
         else:
             raise e
 
+
 def stream_generator(prompt, thread_id):
-    # print(f'First time thread in the function {thread_id}')
-    message = safe_message_send(prompt, thread_id)  # use the new safe send function
-    
-    with st.spinner("Wait... Generating response..."):
-        try:
-            stream = client.beta.threads.runs.create(
-                thread_id=thread_id,
-                assistant_id=assistant_id,
-                stream=True
-            )
-
-            for event in stream:
-                if event.data.object == "thread.message.delta":
-                    for content in event.data.delta.content:
-                        if content.type == 'text':
-                            yield content.text.value
-                            time.sleep(0.01)
-                elif event.data.object == "thread.run.stop":
-                    break  # Break if the run stops
-        except Exception as e:
-            print(f"Error during streaming: {str(e)}")
+    safe_message_send(prompt, thread_id)
+    try:
+        stream = client.beta.threads.runs.create(
+            thread_id=thread_id,
+            assistant_id=ASSISTANT_ID,
+            stream=True
+        )
+        for event in stream:
+            if event.data.object == "thread.message.delta":
+                for content in event.data.delta.content:
+                    if content.type == 'text':
+                        yield content.text.value
+                        time.sleep(0.01)
+            elif event.data.object == "thread.run.stop":
+                break
+    except Exception as e:
+        print(f"Streaming error: {str(e)}")
+
 
 def upload_and_add_to_vector_store(uploaded_file):
-    """Upload a file to OpenAI and add it to the specified vector store."""
     try:
-        # Convert the uploaded file to a BytesIO stream for uploading
         file_stream = io.BytesIO(uploaded_file.getvalue())
-        file_stream.name = uploaded_file.name  # Preserve the file name    
-        # Upload the file to the vector store
-        file_batch = client.beta.vector_stores.file_batches.upload_and_poll(
-            vector_store_id=vector_store_id,
+        file_stream.name = uploaded_file.name
+        client.vector_stores.file_batches.upload_and_poll(
+            vector_store_id=VECTOR_STORE_ID,
             files=[file_stream]
         )
-
-
-        st.success(f"File '{uploaded_file.name}' processed and added to vector store. Status: {file_batch.status}")
+        print(f"File '{uploaded_file.name}' uploaded successfully.")
     except Exception as e:
-        st.error(f"Failed to process file: {str(e)}")
-    
+        print(f"Upload failed: {str(e)}")
+
 
 def list_all_files_in_vector_store():
-    """List all files in the specified vector store."""
     try:
-        all_files = list(client.beta.vector_stores.files.list(vector_store_id=vector_store_id))
-        # st.write(all_files)
-        for file in all_files:
-            file_id = file.id
-            st.write(file_id)   
+        all_files = client.vector_stores.files.list(vector_store_id=VECTOR_STORE_ID)
+        for file in all_files.data:
+            print(f"File ID: {file.id}")
     except Exception as e:
-        st.error(f"Failed to list files: {str(e)}")
-        return {}
+        print(f"Failed to list files: {str(e)}")
+
 
-def delete_file_from_vector_store(vector_store_id, file_id):
-    """Delete a file from the specified vector store."""
+def delete_file_from_vector_store(file_id):
     try:
-        client.beta.vector_stores.files.delete(
-            vector_store_id=vector_store_id,
+        client.vector_stores.files.delete(
+            vector_store_id=VECTOR_STORE_ID,
             file_id=file_id
         )
-        st.success(f"File with ID '{file_id}' deleted from vector store '{vector_store_id}'.")
+        print(f"File '{file_id}' deleted successfully.")
     except Exception as e:
-        st.error(f"Failed to delete file. File id is not Found.")
+        print(f"Delete failed: {str(e)}")
+
+# Simple CLI-based Interface for Testing
+def main():
+    import sys
+
+    session_state = {}
+
+    print("Welcome to the CLI Chatbot!")
+    while True:
+        print("\nOptions:")
+        print("1. Upload file")
+        print("2. List files")
+        print("3. Delete file")
+        print("4. Chat")
+        print("5. Exit")
+        choice = input("Choose an option: ")
 
-# Interface to delete files from vector store
-st.sidebar.subheader("Delete File from Vector Store")
-file_id_to_delete = st.sidebar.text_input("Enter File ID to Delete", "")
-if st.sidebar.button("Delete File"):
-    delete_file_from_vector_store(vector_store_id, file_id_to_delete)
+        if choice == "1":
+            file_path = input("Enter path to PDF or DOCX file: ")
+            try:
+                with open(file_path, "rb") as f:
+                    class UploadedFile:
+                        def __init__(self, name, content):
+                            self.name = name
+                            self.content = content
 
-# Streamlit interface setup
-st.title("💬Chatbot")
-st.caption("🚀 A Streamlit Custom Chatbot")
+                        def getvalue(self):
+                            return self.content
 
+                    uploaded_file = UploadedFile(os.path.basename(file_path), f.read())
+                    upload_and_add_to_vector_store(uploaded_file)
+            except Exception as e:
+                print(f"File upload error: {str(e)}")
 
-with st.sidebar:
-    st.write("Upload PDF File")
-    uploaded_file = st.file_uploader("Choose a file", type=['pdf', 'docx'], key='file_uploader')
+        elif choice == "2":
+            list_all_files_in_vector_store()
+
+        elif choice == "3":
+            file_id = input("Enter File ID to delete: ")
+            delete_file_from_vector_store(file_id)
+
+        elif choice == "4":
+            if "messages" not in session_state:
+                session_state["messages"] = []
+            user_prompt = input("You: ")
+            session_state["messages"].append({"role": "user", "content": user_prompt})
+            thread_id = ensure_single_thread_id(session_state)
+            print("Assistant:", end=" ")
+            for part in stream_generator(user_prompt, thread_id):
+                print(part, end="", flush=True)
+            print()
+
+        elif choice == "5":
+            sys.exit()
 
-    if st.button('Upload File', key='process_file'):
-        if uploaded_file is not None:
-            upload_and_add_to_vector_store(uploaded_file)
-            st.success("File successfully uploaded and processed.")
         else:
-            st.error("Please upload a file to process.")
+            print("Invalid choice, please try again.")
+
+
+if __name__ == "__main__":
+    main()
+
 
-    # List all uploaded files
-    st.write("### Uploaded Files")
-    if 'uploaded_files' in st.session_state and st.session_state.uploaded_files:
-        for file_name, file_id in st.session_state.uploaded_files.items():
-            st.write(f"{file_name}: {file_id}")
+# import openai
+# import streamlit as st
+# from openai import OpenAI
+# import io
+# import time
+# import os
+# from dotenv import load_dotenv
+
+# # OpenAI keys
+# OPENAI_API_KEY = st.secrets["OPENAI_API_KEY"]
+# # VECTOR_STORE_ID = st.secrets["VECTOR_STORE_ID"]
+# # VECTOR_STORE_ID = os.getenv("VECTOR_STORE_ID")
+
+# client = OpenAI(api_key=OPENAI_API_KEY)
+# # vector_store_id =VECTOR_STORE_ID  # Vector Store ID to use
+# vector_store_id = "vs_sHLBqA88ng66RDZYMgaBmDuX"
+# st.write(vector_store_id)
+# # all_files = list(client.beta.vector_stores.files.list(vector_store_id))
+# # for file in all_files:
+# #   # print(file)
+# #   file_id = file.id
+# #   st.write(file_id)
+
+# # Set the assistant ID
+# assistant_id = "asst_UeIHpkpVejAgrIz3RnisxSLm"  # Replace with your own assistant ID
+
+# def ensure_single_thread_id():
+#     if "thread_id" not in st.session_state:
+#         thread = client.beta.threads.create()
+#         st.session_state.thread_id = thread.id
+#     return st.session_state.thread_id
+
+# def safe_message_send(prompt, thread_id):
+#     try:
+#         message = client.beta.threads.messages.create(
+#             thread_id=thread_id,
+#             role="user",
+#             content=prompt
+#         )
+#         return message
+#     except Exception as e:
+#         if "active" in str(e):
+#             print("Waiting for the current run to finish...")
+#             time.sleep(1)  # wait a bit before retrying
+#             return safe_message_send(prompt, thread_id)  # retry sending the message
+#         else:
+#             raise e
+
+# def stream_generator(prompt, thread_id):
+#     # print(f'First time thread in the function {thread_id}')
+#     message = safe_message_send(prompt, thread_id)  # use the new safe send function
+    
+#     with st.spinner("Wait... Generating response..."):
+#         try:
+#             stream = client.beta.threads.runs.create(
+#                 thread_id=thread_id,
+#                 assistant_id=assistant_id,
+#                 stream=True
+#             )
+
+#             for event in stream:
+#                 if event.data.object == "thread.message.delta":
+#                     for content in event.data.delta.content:
+#                         if content.type == 'text':
+#                             yield content.text.value
+#                             time.sleep(0.01)
+#                 elif event.data.object == "thread.run.stop":
+#                     break  # Break if the run stops
+#         except Exception as e:
+#             print(f"Error during streaming: {str(e)}")
+
+# def upload_and_add_to_vector_store(uploaded_file):
+#     """Upload a file to OpenAI and add it to the specified vector store."""
+#     try:
+#         # Convert the uploaded file to a BytesIO stream for uploading
+#         file_stream = io.BytesIO(uploaded_file.getvalue())
+#         file_stream.name = uploaded_file.name  # Preserve the file name    
+#         # Upload the file to the vector store
+#         file_batch = client.beta.vector_stores.file_batches.upload_and_poll(
+#             vector_store_id=vector_store_id,
+#             files=[file_stream]
+#         )
+
+
+#         st.success(f"File '{uploaded_file.name}' processed and added to vector store. Status: {file_batch.status}")
+#     except Exception as e:
+#         st.error(f"Failed to process file: {str(e)}")
+    
+
+# def list_all_files_in_vector_store():
+#     """List all files in the specified vector store."""
+#     try:
+#         all_files = list(client.beta.vector_stores.files.list(vector_store_id=vector_store_id))
+#         # st.write(all_files)
+#         for file in all_files:
+#             file_id = file.id
+#             st.write(file_id)   
+#     except Exception as e:
+#         st.error(f"Failed to list files: {str(e)}")
+#         return {}
+
+# def delete_file_from_vector_store(vector_store_id, file_id):
+#     """Delete a file from the specified vector store."""
+#     try:
+#         client.beta.vector_stores.files.delete(
+#             vector_store_id=vector_store_id,
+#             file_id=file_id
+#         )
+#         st.success(f"File with ID '{file_id}' deleted from vector store '{vector_store_id}'.")
+#     except Exception as e:
+#         st.error(f"Failed to delete file. File id is not Found.")
+
+# # Interface to delete files from vector store
+# st.sidebar.subheader("Delete File from Vector Store")
+# file_id_to_delete = st.sidebar.text_input("Enter File ID to Delete", "")
+# if st.sidebar.button("Delete File"):
+#     delete_file_from_vector_store(vector_store_id, file_id_to_delete)
+
+# # Streamlit interface setup
+# st.title("💬Chatbot")
+# st.caption("🚀 A Streamlit Custom Chatbot")
+
+
+# with st.sidebar:
+#     st.write("Upload PDF File")
+#     uploaded_file = st.file_uploader("Choose a file", type=['pdf', 'docx'], key='file_uploader')
+
+#     if st.button('Upload File', key='process_file'):
+#         if uploaded_file is not None:
+#             upload_and_add_to_vector_store(uploaded_file)
+#             st.success("File successfully uploaded and processed.")
+#         else:
+#             st.error("Please upload a file to process.")
+
+#     # List all uploaded files
+#     st.write("### Uploaded Files")
+#     if 'uploaded_files' in st.session_state and st.session_state.uploaded_files:
+#         for file_name, file_id in st.session_state.uploaded_files.items():
+#             st.write(f"{file_name}: {file_id}")
     
-    # List all files in the vector store
-    st.write("## All Files in Vector Store")
-    all_files = list_all_files_in_vector_store() 
-
-# Initialize session state for chat
-st.session_state.start_chat = True
-if 'start_chat' not in st.session_state:
-    st.session_state.start_chat = False
-
-# Main chat interface
-if st.session_state.start_chat:
-    if "messages" not in st.session_state:
-        st.session_state.messages = []
-
-    for message in st.session_state.messages:
-        with st.chat_message(message["role"]):
-            st.markdown(message["content"])
-
-    prompt = st.chat_input("Enter your message")
-    if prompt:
-        thread_id = ensure_single_thread_id()
-        with st.chat_message("user"):
-            st.markdown(prompt)
-        st.session_state.messages.append({"role": "user", "content": prompt})
+#     # List all files in the vector store
+#     st.write("## All Files in Vector Store")
+#     all_files = list_all_files_in_vector_store() 
+
+# # Initialize session state for chat
+# st.session_state.start_chat = True
+# if 'start_chat' not in st.session_state:
+#     st.session_state.start_chat = False
+
+# # Main chat interface
+# if st.session_state.start_chat:
+#     if "messages" not in st.session_state:
+#         st.session_state.messages = []
+
+#     for message in st.session_state.messages:
+#         with st.chat_message(message["role"]):
+#             st.markdown(message["content"])
+
+#     prompt = st.chat_input("Enter your message")
+#     if prompt:
+#         thread_id = ensure_single_thread_id()
+#         with st.chat_message("user"):
+#             st.markdown(prompt)
+#         st.session_state.messages.append({"role": "user", "content": prompt})
         
-        with st.chat_message("assistant"):
-            response = st.write_stream(stream_generator(prompt, thread_id))
-        st.session_state.messages.append({"role": "assistant", "content": response})
+#         with st.chat_message("assistant"):
+#             response = st.write_stream(stream_generator(prompt, thread_id))
+#         st.session_state.messages.append({"role": "assistant", "content": response})