import openai
import gradio as gr
from langchain.retriever import RetrievalQA
from langchain.chains.question_answering import load_qa_cha
from langchain.llms import OpenAI
from langchain.document_loaders import TextLoader
from langchain.indexes import VectorstoreIndexCreator
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.text_splitter import CharacterTextSplitter
from langchain.embeddings import OpenAIEmbeddings
from langchain.vectorstores import Chroma

# Initialize OpenAI API key
openai.api_key = "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"

# Load text from file
loader = TextLoader("Dropsheets.txt")
documents = loader.load()

# split the documents into chunks
text_splitter = RecursiveCharacterTextSplitter(chunk_size=1024, chunk_overlap=0)
texts = text_splitter.split_documents(documents)

# select embeddings 
embeddings = OpenAIEmbeddings()

# create the vectorestore to use as the index
db = Chroma.from_documents(texts, embeddings)

# expose this index in a retriever interface
retriever = db.as_retriever(search_type="similarity", search_kwargs={"k":2})


# Define OpenAI GPT-3.5 model function
## def generate_text(query):
#    response = openai.Completion.create(
 #       engine="text-davinci-002",
 #       temperature=0,
  #      max_tokens=7000,
   #     prompt=prompt
#    )
 #   return response.choices[0].text.strip()      

    
# Create Gradio interface
input_text = gr.Textbox(label="Enter prompt", type="text")
output_text = gr.Textbox(label="AI response", type="text")
demo = gr.Interface(
    fn = None,
    inputs=input_text,
    outputs=output_text,
    title="AI Chatbot for PlanetTogether Knowledge Base",
    description="Ask a question about the PlanetTogether APS:",
    examples=[["How do you create an Alternate Path?"]],
    theme="default"
)

# create a chain to answer questions 
qa = RetrievalQA.from_chain_type(
    llm=OpenAI(), chain_type="stuff", retriever=retriever)
result = qa({"query": query})
retriever.get_relevant_documents(query)


# Launch demo
demo.launch()