app.py

import os
from typing import List
from langchain_community.vectorstores import FAISS
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.chains import (
    ConversationalRetrievalChain,
)
from langchain.document_loaders import PyPDFLoader
from langchain.chat_models import ChatOpenAI
from langchain.prompts.chat import (
    ChatPromptTemplate,
    SystemMessagePromptTemplate,
    HumanMessagePromptTemplate,
)
from langchain.docstore.document import Document
from langchain.memory import ChatMessageHistory, ConversationBufferMemory
from chainlit.types import AskFileResponse
from langchain.document_loaders import PyMuPDFLoader  # Added import for PyMuPDFLoader

import chainlit as cl

text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=100)

system_template = """Use the following pieces of context to answer the users question.
If you don't know the answer, just say that you don't know, don't try to make up an answer.
ALWAYS return a "SOURCES" part in your answer.
The "SOURCES" part should be a reference to the source of the document from which you got your answer.
And if the user greets with greetings like Hi, hello, How are you, etc reply accordingly as well.
Example of your response should be:
The answer is foo
SOURCES: xyz
Begin!
----------------
{summaries}"""
messages = [
    SystemMessagePromptTemplate.from_template(system_template),
    HumanMessagePromptTemplate.from_template("{question}"),
]
prompt = ChatPromptTemplate.from_messages(messages)
chain_type_kwargs = {"prompt": prompt}


def process_pdf_from_link(link: str):
    # Download the PDF from the link
    pdf_loader = PyMuPDFLoader(link)
    docs = pdf_loader.load()
    texts = [doc.page_content for doc in docs]
    return texts


@cl.on_chat_start
async def on_chat_start():
    # Process the PDF from the link
    link = "https://d18rn0p25nwr6d.cloudfront.net/CIK-0001045810/1cbe8fe7-e08a-46e3-8dcc-b429fc06c1a4.pdf"
    texts = process_pdf_from_link(link)

    # Create metadata for each chunk
    metadatas = [{"source": f"{i}-pl"} for i in range(len(texts))]

    # Create a Chroma vector store
    embeddings = OpenAIEmbeddings()
    docsearch = await cl.make_async(FAISS.from_texts)(
        texts, embeddings, metadatas=metadatas
    )

    message_history = ChatMessageHistory()

    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(
        ChatOpenAI(model_name="gpt-3.5-turbo", temperature=0, streaming=True),
        chain_type="stuff",
        retriever=docsearch.as_retriever(),
        memory=memory,
        return_source_documents=True,
    )

    # Let the user know that the system is ready
    await cl.Message(content="PDF processing complete. You can now ask questions!").send()

    cl.user_session.set("chain", chain)


@cl.on_message
async def main(message):
    chain = cl.user_session.get("chain")  # type: ConversationalRetrievalChain
    cb = cl.AsyncLangchainCallbackHandler()

    res = await chain.acall(message.content, callbacks=[cb])
    answer = res["answer"]
    source_documents = res["source_documents"]  # type: List[Document]

    text_elements = []  # type: List[cl.Text]

    if source_documents:
        for source_idx, source_doc in enumerate(source_documents):
            source_name = f"source_{source_idx}"
            # Create the text element referenced in the message
            text_elements.append(
                cl.Text(content=source_doc.page_content, name=source_name)
            )
        source_names = [text_el.name for text_el in text_elements]

        if source_names:
            answer += f"\nSources: {', '.join(source_names)}"
        else:
            answer += "\nNo sources found"

    await cl.Message(content=answer, elements=text_elements).send()