from gpt_index import GPTSimpleVectorIndex
from langchain import OpenAI
import gradio as gr
import sys
import os
import datetime
import huggingface_hub
from huggingface_hub import Repository
from datetime import datetime
import csv

os.environ["OPENAI_API_KEY"] = os.environ['SECRET_CODE']

# Need to write to persistent dataset because cannot store temp data on spaces
DATASET_REPO_URL = "https://huggingface.co/datasets/peterpull/MediatorBot"
DATA_FILENAME = "data.csv"
DATA_FILE = os.path.join("data", DATA_FILENAME)

# I am guessing we need a write access token.
HF_TOKEN = os.environ.get("HF_TOKEN")
print("HF TOKEN is none?", HF_TOKEN is None)
print("HF hub ver", huggingface_hub.__version__)

repo = Repository(
    local_dir="data", 
    clone_from=DATASET_REPO_URL, 
    use_auth_token=HF_TOKEN)


def generate_html() -> str:
    with open(DATA_FILE) as csvfile:
        reader = csv.DictReader(csvfile)
        rows = []
        for row in reader:
            rows.append(row)
        rows.reverse()
        if len(rows) == 0:
            return "no messages yet"
        else:
            html = "<div class='chatbot'>"
            for row in rows:
                html += "<div>"
                html += f"<span>{row['chatinput']}</span>"
                html += f"<span class='message'>{row['chatresponse']}</span>"
                html += f"<span class='time'>{row['time']}</span>"
                html += "</div>"
            html += "</div>"
            return html

def store_message(chatinput: str, chatresponse: str):
    if chatinput and chatresponse:
        with open(DATA_FILE, "a") as csvfile:
            writer = csv.writer(csvfile, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
            if csvfile.tell() == 0:
                writer.writerow(['User input', 'Chatbot response', 'Date and Time'])
            writer.writerow([chatinput, chatresponse, datetime.now().strftime('%Y-%m-%d %H:%M:%S')])
        commit_url = repo.push_to_hub()
        print(commit_url)

    return generate_html()

            
#gets the index file which is the context data
def get_index(index_file_path):
    if os.path.exists(index_file_path):
        return GPTSimpleVectorIndex.load_from_disk(index_file_path)
    else:
        print(f"Error: '{index_file_path}' does not exist.")
        sys.exit()

# passes the prompt to the chatbot
def chatbot(input_text, mentioned_person='Mediator John Haynes'):
    index = get_index('./index/indexsmall.json')
    prompt = f"You are {mentioned_person}: {input_text}\n\n At the end of your answer  ask a provocative question."
    response = index.query(prompt, response_mode="compact")

    store_message(input_text,response)
    
    # return the response
    return response.response



iface = gr.Interface(fn=chatbot, 
                     inputs=gr.inputs.Textbox("Enter your question"), 
                     outputs="text",
                     title="AI Chatbot trained on J. Haynes mediation material, v0.1",
                     description="test")
iface.launch()