import gradio as gr
from transformers import pipeline

text_generator = pipeline("text-generation", model="openai-community/gpt2")
translator = pipeline("translation_en_to_fr", model="Helsinki-NLP/opus-mt-en-fr")
summarizer = pipeline("summarization", model="facebook/bart-large-cnn")

def generate_text(prompt):
    try:
        result = text_generator(prompt, max_new_tokens=60, num_return_sequences=1)
        return result[0]["generated_text"]
    except Exception as e:
        return f"Error : {str(e)}"

def translate_text(text):
    try:
        result = translator(text)
        return result[0]["translation_text"]
    except Exception as e:
        return f"Error : {str(e)}"

def summarize_text(text):
    try:
        result = summarizer(text, max_length=80, min_length=20, do_sample=False)
        return result[0]["summary_text"]
    except Exception as e:
        return f"Error : {str(e)}"

with gr.Blocks() as demo:
    gr.Markdown("# Hugging Face Demo")

    with gr.Tab("Génération de texte"):
        text_input = gr.Textbox(label="Prompt", placeholder="Once upon a time...", lines=4, scale=2)
        text_output = gr.Textbox(label="Texte généré", lines=8, scale=2)
        text_btn = gr.Button("Générer")
        text_btn.click(generate_text, inputs=text_input, outputs=text_output)

    with gr.Tab("Traduction de texte (anglais -> français)"):
        translate_input = gr.Textbox(label="Texte en anglais", placeholder="Machine learning is fascinating!", lines=4, scale=2)
        translate_output = gr.Textbox(label="Traduction en français", lines=6, scale=2)
        translate_btn = gr.Button("Traduire")
        translate_btn.click(translate_text, inputs=translate_input, outputs=translate_output)

    with gr.Tab("Résumé de texte"):
        summarize_input = gr.Textbox(label="Texte long à résumer", lines=10, scale=2)
        summarize_output = gr.Textbox(label="Résumé", lines=6, scale=2)
        summarize_btn = gr.Button("Résumer")
        summarize_btn.click(summarize_text, inputs=summarize_input, outputs=summarize_output)

demo.launch()