from ailab_crs import NLP_tasks_crs, Prompt_engineering_crs
import gradio as gr
from googletrans import LANGUAGES 


# CSS to hide the entire footer, including "Made with Gradio", settings, and API info
hide_footer_css = """
<style>
footer, footer * {
    display: none !important;
}
</style>
"""

# Mobile-friendly CSS
mobile_css = """
<style>
/* Make all inputs, textboxes, and buttons full width on small screens */
@media (max-width: 768px) {
    input, textarea, button, .gr-textbox, .gr-button {
        width: 100% !important;
        margin: 5px 0 !important;
    }
}
</style>
"""
    
nlp_tasks = NLP_tasks_crs()
supported_langs = list(LANGUAGES.values())


with gr.Blocks() as demo:
    # Inject the mobile CSS
    gr.HTML(mobile_css)

    # App title
    gr.Markdown("# 🧠 NaanhAI 💡")

    # Inject the CSS at the top
    gr.HTML(hide_footer_css)
        
    with gr.Row():
        with gr.Column():
            text = gr.Textbox(label="Your Query", lines=8)
        with gr.Column():
            with gr.Accordion("Translation & Summarization Parameters   ", open= True):
                language = gr.Dropdown(choices=supported_langs, label="Select Target Language", value="french")
                style = gr.Textbox(label="Choose Your Style", value="polite")
    
    with gr.Row(scale=5):
    # with gr.Column(scale=1, min_width=1):
        btn = gr.Button("Q&A")
    # with gr.Column(scale=2, min_width=1):
        btn1 = gr.Button("Translator")
    # with gr.Column(scale=2, min_width=1):
        btn2 = gr.Button("Summarizer")
    # with gr.Column(scale=2, min_width=1):
        btn3 = gr.Button("Translator_Summarizer")

    answer = gr.Textbox(label="AI Answer", lines=2)

    btn.click(
        fn= nlp_tasks.question_answer,
        inputs= text,
        outputs=answer
    )
    btn1.click(
        fn= nlp_tasks.translator,
        inputs= [text, language, style],
        outputs=answer
    )
    btn2.click(
        fn= nlp_tasks.summarization,
        inputs= text,
        outputs=answer
    )
    btn3.click(
        fn= nlp_tasks.translator_summarization,
        inputs= [text, language, style],
        outputs=answer
    )


if __name__ == "__main__":
    demo.launch()