import gradio as gr
import pytube
from transformers import pipeline
from textblob import TextBlob

# Initialize sentiment analysis pipeline
sentiment_analyzer = pipeline("sentiment-analysis")

def analyze_youtube_content(youtube_url, transcript_text=""):
    """Analyze YouTube content"""
    results = {}
    
    # Get video info
    if youtube_url:
        try:
            yt = pytube.YouTube(youtube_url)
            results["video_info"] = {
                "title": yt.title,
                "status": "success"
            }
        except Exception as e:
            results["video_info"] = {
                "status": "error",
                "message": str(e)
            }
    
    # Analyze transcript
    if transcript_text:
        # TextBlob sentiment
        blob = TextBlob(transcript_text)
        sentiment = blob.sentiment
        
        # Hugging Face sentiment
        hf_result = sentiment_analyzer(transcript_text[:512])[0]
        
        results["sentiment"] = {
            "polarity": round(sentiment.polarity, 2),
            "assessment": "positive" if sentiment.polarity > 0 else "negative" if sentiment.polarity < 0 else "neutral",
            "huggingface": hf_result["label"]
        }
    
    return results

# Create Gradio interface
demo = gr.Interface(
    fn=analyze_youtube_content,
    inputs=[
        gr.Textbox(label="YouTube URL"),
        gr.Textbox(label="Transcript Text", lines=10)
    ],
    outputs=gr.JSON(label="Analysis Results"),
    title="YouTube Viral Moment Analyzer",
    description="Analyze viral moments from YouTube videos using ML models"
)

# Launch with MCP server enabled
if __name__ == "__main__":
    demo.launch(mcp_server=True)