app.py

from langchain.llms import OpenAI
from langchain.chains.qa_with_sources import load_qa_with_sources_chain
from langchain.docstore.document import Document
import requests
import pathlib
import subprocess
import tempfile
import os
import gradio as gr
import pickle

# using a vector space for our search
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.vectorstores.faiss import FAISS
from langchain.text_splitter import CharacterTextSplitter

#To get markdowns from github fo Gradio (/or your) repo
def get_github_docs(repo_owner, repo_name):
    with tempfile.TemporaryDirectory() as d:
        subprocess.check_call(
            f"git clone https://github.com/{repo_owner}/{repo_name}.git .",
            cwd=d,
            shell=True,
        )
        git_sha = (
            subprocess.check_output("git rev-parse HEAD", shell=True, cwd=d)
            .decode("utf-8")
            .strip()
        )
        repo_path = pathlib.Path(d)
        markdown_files = list(repo_path.rglob("*.md")) + list(
            repo_path.rglob("*.mdx")
        )
        for markdown_file in markdown_files:
            try:
                with open(markdown_file, "r") as f:
                    relative_path = markdown_file.relative_to(repo_path)
                    github_url = f"https://github.com/{repo_owner}/{repo_name}/blob/{git_sha}/{relative_path}"
                    yield Document(page_content=f.read(), metadata={"source": github_url})
            except FileNotFoundError:
                print(f"Could not open file: {markdown_file}")

#sources = get_github_docs("gradio-app", "gradio")
#source_chunks = []
#splitter = CharacterTextSplitter(separator=" ", chunk_size=1024, chunk_overlap=0)

#for source in sources:
#    for chunk in splitter.split_text(source.page_content):
#        source_chunks.append(Document(page_content=chunk, metadata=source.metadata))
        
#search_index = FAISS.from_documents(source_chunks, OpenAIEmbeddings()) #(source_chunks, OpenAIEmbeddings()) # <------
#chain = load_qa_with_sources_chain(OpenAI(temperature=0))  ## <<---------

                
#loading FAISS search index from disk
with open("search_index.pickle", "rb") as f:
    search_index = pickle.load(f)


def print_answer(question, openai):   #openai_embeddings
    #search_index = get_search_index()
    chain = load_qa_with_sources_chain(openai) #(OpenAI(temperature=0)) 
    response = (
        chain(
            {
                "input_documents": search_index.similarity_search(question, k=4),
                "question": question,
            },
            return_only_outputs=True,
        )["output_text"]
    )
    #print(response)
    if len(response.split('\n')[-1].split())>2:
        response = response.split('\n')[0] + ', '.join([' <a href="' + response.split('\n')[-1].split()[i] + '" target="_blank"><u>Click Link' + str(i) + '</u></a>' for i in range(1,len(response.split('\n')[-1].split()))])
    else: 
        response = response.split('\n')[0] + ' <a href="' + response.split('\n')[-1].split()[-1] + '" target="_blank"><u>Click Link</u></a>'
    return response



def chat(message, history, openai_api_key):
    #openai_embeddings = OpenAIEmbeddings(openai_api_key=openai_api_key)
    openai = OpenAI(temperature=0, openai_api_key=openai_api_key )
    #os.environ["OPENAI_API_KEY"] = openai_api_key
    history = history or []
    message = message.lower()
    response = print_answer(message, openai)   #openai_embeddings
    history.append((message, response))
    return history, history

#chatbot = gr.Chatbot().style(color_map=("green", "orange"))

with gr.Blocks() as demo:
  #gr.Markdown("""<h1><centre>LangChain - powered - Gradio-Helper-Bot </h1></centre>  """)
  gr.HTML("""<div style="text-align: center; max-width: 700px; margin: 0 auto;">
        <div
        style="
            display: inline-flex;
            align-items: center;
            gap: 0.8rem;
            font-size: 1.75rem;
        "
        >
        <h1 style="font-weight: 900; margin-bottom: 7px; margin-top: 5px;">
            Gradio QandA - LangChain Bot
        </h1>
        </div>
        <p style="margin-bottom: 10px; font-size: 94%">
        Hi, I'm a Q and A Gradio expert bot, start by typing in your OpenAI API key, questions/issues you are facing in your Gradio implementations and then press enter.<br>
        <a href="https://huggingface.co/spaces/ysharma/InstructPix2Pix_Chatbot?duplicate=true"><img src="https://bit.ly/3gLdBN6" alt="Duplicate Space"></a>Duplicate Space with GPU Upgrade for fast Inference & no queue<br> 
        Built using <a href="https://langchain.readthedocs.io/en/latest/" target="_blank">LangChain</a> and <a href="https://github.com/gradio-app/gradio" target="_blank">Gradio Github repo</a>
        </p>
    </div>""")  
  with gr.Row():
    question = gr.Textbox(label = 'Type in your questions about Gradio here and press Enter!', placeholder = 'What is the role of "every" argument in a component')
    openai_api_key = gr.Textbox(type='password', label="Enter your OpenAI API key here")
  state = gr.State()
  chatbot = gr.Chatbot()
  question.submit(chat, [question, state, openai_api_key], [chatbot, state])

if __name__ == "__main__":
    demo.launch()