app.py

import os
import gradio as gr
import google.generativeai as genai
import tempfile

# Ensure the API key is set
api_key = os.getenv("GOOGLE_API_KEY")
if not api_key:
    raise ValueError("GOOGLE_API_KEY not found. Please set it in your environment variables.")
genai.configure(api_key=api_key)

# Initialize the Gemini model
model = genai.GenerativeModel(model_name="gemini-2.0-flash-exp")

def process_input(text_input, files):
    """Process input text and files, send them to the Gemini API, and get a response."""
    contents = []
    if text_input:
        contents.append({"text": text_input})  # Add text content for Gemini

    # Handle files: read and prepare for Gemini API
    for file in files:
        # Temporary file storage to read content
        with tempfile.NamedTemporaryFile(delete=False) as tmp:
            tmp.write(file.read())
            tmp.seek(0)  # Rewind file to the beginning for reading
            
            # Depending on file type and Gemini requirements, you might adjust how you read the file
            contents.append({
                "file": tmp.name,
                "mime_type": file.type  # MIME type is needed if specific handling is required by Gemini
            })

    # Call Gemini API to process the collected contents
    try:
        response = model.generate_content(contents)
        response.resolve()
        return response.text
    except Exception as e:
        return f"Error communicating with Gemini API: {e}"

# Create the Gradio interface
def create_interface():
    with gr.Blocks() as app:
        with gr.Row():
            text_input = gr.Textbox(label="Enter your text:", placeholder="Type your query here...")
            file_input = gr.File(label="Upload files", file_types=["pdf", "png", "jpg", "mp3", "mp4"], multiple=True)
            submit_button = gr.Button("Process")

        output = gr.Textbox(placeholder="Response will appear here...")

        submit_button.click(
            fn=process_input,
            inputs=[text_input, file_input],
            outputs=output
        )

    return app

# Run the interface
if __name__ == "__main__":
    app = create_interface()
    app.launch()