app.py

import streamlit as st
import pandas as pd
import os
import traceback
from dotenv import load_dotenv
from llama_index.readers.file.paged_csv.base import PagedCSVReader
from llama_index.core import Settings, VectorStoreIndex
from llama_index.llms.openai import OpenAI
from llama_index.embeddings.openai import OpenAIEmbedding
from llama_index.vector_stores.faiss import FaissVectorStore
from llama_index.core.ingestion import IngestionPipeline
from langchain_community.vectorstores import FAISS as LangChainFAISS
from langchain_community.docstore.in_memory import InMemoryDocstore
from langchain.chains import create_retrieval_chain
from langchain.chains.combine_documents import create_stuff_documents_chain
from langchain_core.prompts import ChatPromptTemplate
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
from langchain_core.documents import Document  
import faiss
import tempfile

# Load environment variables
os.environ["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY")

# ✅ Check OpenAI API Key
if not os.getenv("OPENAI_API_KEY"):
    st.error("⚠️ OpenAI API Key is missing! Please check your .env file or environment variables.")

# ✅ Ensure OpenAI Embeddings match FAISS dimensions
embedding_function = OpenAIEmbeddings()
test_vector = embedding_function.embed_query("test")  # Sample embedding
faiss_dimension = len(test_vector)  # ✅ Dynamically detect correct dimension

# ✅ Update global settings for LlamaIndex
Settings.llm = OpenAI(model="gpt-4o")
Settings.embed_model = OpenAIEmbedding(model="text-embedding-3-small", dimensions=faiss_dimension)

# Streamlit app
st.title("Chat with CSV Files - LangChain vs LlamaIndex")

# File uploader
uploaded_file = st.file_uploader("Upload a CSV file", type=["csv"])
if uploaded_file:
    try:
        # Read and preview CSV data using pandas
        data = pd.read_csv(uploaded_file)
        st.write("Preview of uploaded data:")
        st.dataframe(data)

        # Save the uploaded file to a temporary location
        with tempfile.NamedTemporaryFile(delete=False, suffix=".csv", mode="w", encoding="utf-8") as temp_file:
            temp_file_path = temp_file.name
            data.to_csv(temp_file.name, index=False, encoding="utf-8")
            temp_file.flush()

        # Debugging: Verify the temporary file (Display partial content)
        st.write("Temporary file path:", temp_file_path)
        with open(temp_file_path, "r") as f:
            content = f.read()
        st.write("Partial file content (first 500 characters):")
        st.text(content[:500])

        # Tabs for LangChain and LlamaIndex
        tab1, tab2 = st.tabs(["LangChain", "LlamaIndex"])

        # ✅ LangChain Processing
        with tab1:
            st.subheader("LangChain Query")

            try:
                # ✅ Convert CSV rows into LangChain Document objects
                st.write("Processing CSV with a custom loader...")
                documents = []
                for _, row in data.iterrows():
                    content = "\n".join([f"{col}: {row[col]}" for col in data.columns])
                    doc = Document(page_content=content)  
                    documents.append(doc)


                # ✅ Create FAISS VectorStore with Correct Dimensions
                st.write(f"✅ Initializing FAISS with dimension: {faiss_dimension}")
                langchain_index = faiss.IndexFlatL2(faiss_dimension)

                docstore = InMemoryDocstore()
                index_to_docstore_id = {}

                langchain_vector_store = LangChainFAISS(
                    embedding_function=embedding_function,
                    index=langchain_index,
                    docstore=docstore,
                    index_to_docstore_id=index_to_docstore_id,
                )

                # ✅ Ensure documents are added correctly
                try:
                    langchain_vector_store.add_documents(documents)
                    st.write("✅ Documents successfully added to FAISS VectorStore.")
                except Exception as e:
                    st.error(f"Error adding documents to FAISS: {e}")

                # ✅ Create LangChain Query Execution Pipeline
                retriever = langchain_vector_store.as_retriever()
                system_prompt = (
                    "You are an assistant for question-answering tasks. "
                    "Use the following pieces of retrieved context to answer "
                    "the question. If you don't know the answer, say that you "
                    "don't know. Use three sentences maximum and keep the "
                    "answer concise.\n\n{context}"
                )
                prompt = ChatPromptTemplate.from_messages(
                    [("system", system_prompt), ("human", "{input}")]
                )
                question_answer_chain = create_stuff_documents_chain(ChatOpenAI(model="gpt-4o"), prompt)
                langchain_rag_chain = create_retrieval_chain(retriever, question_answer_chain)

                # ✅ Query Processing
                query = st.text_input("Ask a question about your data (LangChain):")

                if query:
                    try:
                        st.write("Processing your question...")
                        answer = langchain_rag_chain.invoke({"input": query})
                        st.write(f"**Answer:** {answer['answer']}")
                    except Exception as e:
                        error_message = traceback.format_exc()
                        st.error(f"Error processing query: {e}")
                        st.text(error_message)

            except Exception as e:
                error_message = traceback.format_exc()
                st.error(f"Error processing with LangChain: {e}")
                st.text(error_message)

    except Exception as e:
        error_message = traceback.format_exc()
        st.error(f"Error reading uploaded file: {e}")
        st.text(error_message)