from transformers import pipeline
import gradio as gr
import requests

# Initialize the summarization pipeline (global initialization for efficiency)
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")

# Function to fetch content from Wikipedia with a limit on length
def fetch_wikipedia_content(search_term, max_chars=10000):
    url = "https://en.wikipedia.org/w/api.php"
    params = {
        "action": "query",
        "format": "json",
        "prop": "extracts",
        "explaintext": True,
        "titles": search_term,
        "redirects": 1  # Follow redirects
    }
    try:
        response = requests.get(url, headers=headers, params=params)
        response.raise_for_status()
        data = response.json()
        page = next(iter(data["query"]["pages"].values()))
        content = page.get("extract", "")
        return content[:max_chars] if content else "⚠️ No content found for this topic."
    except requests.RequestException as e:
        return f"⚠️ Error fetching article: {e}"

# Function to split text into manageable chunks
def split_text_into_chunks(text, max_words=500):
    words = text.split()
    return [" ".join(words[i:i + max_words]) for i in range(0, len(words), max_words)]

# Function to summarize text in chunks
def summarize_text(text):
    chunks = split_text_into_chunks(text)
    summaries = []
    for i, chunk in enumerate(chunks):
        try:
            chunk_word_count = len(chunk.split())
            max_summary_length = min(120, max(40, chunk_word_count // 2))  # Optimized for smaller summaries
            summary = summarizer(chunk, max_length=max_summary_length, min_length=30, do_sample=False)[0]['summary_text']
            summaries.append(summary)
        except Exception as e:
            summaries.append(f"⚠️ Error summarizing chunk {i + 1}: {str(e)}")
    return "\n\n".join(summaries)

# Main function for Gradio
def fetch_and_summarize(search_term):
    content = fetch_wikipedia_content(search_term)
    if content.startswith("⚠️"):
        return content
    return summarize_text(content)

# Gradio Interface
iface = gr.Interface(
    fn=fetch_and_summarize,
    inputs=gr.Textbox(lines=1, placeholder="Enter a Wikipedia topic", label="Wikipedia Topic"),
    outputs=gr.Textbox(label="Summarized Content"),
    title="Optimized Wikipedia Article Summarizer",
    description="Fetch and summarize Wikipedia articles efficiently. Optimized for lightweight summarization.",
    allow_flagging="never"  # Disable flagging for a lightweight deployment
)

# Launch Gradio with queuing for handling concurrent requests
if __name__ == "__main__":
    iface.queue(concurrency_count=4).launch(debug=False)