from PyPDF2 import PdfReader
from langchain.embeddings.openai import OpenAIEmbeddings
from langchain.text_splitter import CharacterTextSplitter
from langchain.vectorstores import FAISS
import os
from typing_extensions import Concatenate
import gradio as gr
from langchain.chains.question_answering import load_qa_chain
from langchain.llms import OpenAI

# Set OpenAI API key
os.environ["OPENAI_API_KEY"] = "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"

# Read text from PDF
pdfreader = PdfReader('schizophrenia.pdf')
raw_text = ''
for page in pdfreader.pages:
    content = page.extract_text()
    if content:
        raw_text += content

# Split text into chunks
text_splitter = CharacterTextSplitter(
    separator="\n",
    chunk_size=800,
    chunk_overlap=200,
    length_function=len,
)
texts = text_splitter.split_text(raw_text)

# Extract embeddings using OpenAI
embeddings = OpenAIEmbeddings()

# Perform similarity search
document_search = FAISS.from_texts(texts, embeddings)

# Load question-answering chain
chain = load_qa_chain(OpenAI(), chain_type="stuff")

# Define function to process user query
def process_query(query):
    # Perform similarity search to retrieve relevant documents
    docs = document_search.similarity_search(query)
    
    # Run question-answering process with retrieved documents and user query
    response = chain.run(input_documents=docs, question=query)
    
    # Return the response
    return response

# Define Gradio interface
iface = gr.Interface(
    fn=process_query,
    inputs="text",
    outputs="text",
    title="Serene",
    description="Hi! How do you feel today?"
)

# Launch the interface
iface.launch()