from youtube_transcript_api import YouTubeTranscriptApi
from youtube_transcript_api.formatters import TextFormatter
import re
import torch
import gradio as gr

import os

os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'

from transformers import pipeline


text_summary = pipeline("summarization", model="sshleifer/distilbart-cnn-12-6", torch_dtype=torch.bfloat16)

def summary (input):
    output = text_summary(input)
    return output[0]

def extract_video_id(youtube_url):
    """
    Extracts the video ID from a YouTube URL.
    """
    video_id = re.search(r'(?<=v=)[\w-]+', youtube_url)
    if video_id:
        return video_id.group(0)
    video_id = re.search(r'(?<=be/)[\w-]+', youtube_url)
    if video_id:
        return video_id.group(0)
    return None

def get_transcript(youtube_url):
    """
    Fetches the transcript of a YouTube video given its URL.
    """
    video_id = extract_video_id(youtube_url)
    if not video_id:
        return "Invalid YouTube URL"

    try:
        transcript_list = YouTubeTranscriptApi.get_transcript(video_id)
        formatter = TextFormatter()
        transcript = formatter.format_transcript(transcript_list)
        sum_text = summary(transcript)
        print(sum_text)

        return sum_text
    except Exception as e:
        return str(e)


gr.close_all()
op = gr.Interface(fn=get_transcript,
                  inputs=  [gr.Textbox(label="Enter YouTube URL:",lines=1)],
                  outputs= [gr.Textbox(label="Summarized Transcript:", lines=4)],
                  title = "YouTube Video Transcript Summarizer")
op.launch()
gr.close_all()