app.py

import gradio as gr
from huggingface_hub import InferenceClient
from PIL import Image
import time
import os

# Get the Hugging Face token from the environment variable, or a secret if available.
HF_TOKEN = os.environ.get("HF_TOKEN")

# Check if HF_TOKEN is set; if not, raise a configuration error (handled later)
if not HF_TOKEN:
    HF_TOKEN_ERROR = "Hugging Face API token (HF_TOKEN) not found.  Please set it as an environment variable or Gradio secret."
else:
    HF_TOKEN_ERROR = None  # No error if the token is found


client = InferenceClient(token=HF_TOKEN)  # Use token instead of provider and api_key

def generate_image(prompt, progress=gr.Progress()):
    """Generates an image using the InferenceClient and provides progress updates."""

    if HF_TOKEN_ERROR:
        raise gr.Error(HF_TOKEN_ERROR)

    progress(0, desc="Sending request to Hugging Face...")
    try:
        # Use the client.text_to_image method.  Assume xylaria-iris is valid here.
        image = client.text_to_image(prompt, model="black-forest-labs/FLUX.1-schnell")

        if not isinstance(image, Image.Image): # Basic type checking.
             raise Exception(f"Expected a PIL Image, but got: {type(image)}")

        progress(0.8, desc="Processing image...")
        time.sleep(0.5)  # Simulate some processing
        progress(1.0, desc="Done!")
        return image
    except Exception as e:  # Catch all exceptions from the API call
        # Check for rate limit errors (different with InferenceClient).  This is a best-effort check.
        if "rate limit" in str(e).lower(): # Check message, case-insensitively.
            error_message = f"Rate limit exceeded. Please try again later.  Error: {e}"
        else:
            error_message = f"An error occurred: {e}"  # Generic error message
        raise gr.Error(error_message)



# Gradio Interface (same CSS as before, for consistency)
css = """
.container {
    max-width: 800px;
    margin: auto;
    padding: 20px;
    border: 1px solid #ddd;
    border-radius: 10px;
    box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1);
}
.title {
    text-align: center;
    font-size: 2.5em;
    margin-bottom: 0.5em;
    color: #333;
    font-family: 'Arial', sans-serif; /* More readable font */
}
.description {
    text-align: center;
    font-size: 1.1em;
    margin-bottom: 1.5em;
    color: #555;
}
.input-section, .output-section {
    margin-bottom: 1.5em;
}


/* Animation for the image appearance - subtle fade-in */
@keyframes fadeIn {
    from { opacity: 0; transform: translateY(20px); }
    to   { opacity: 1; transform: translateY(0); }
}


/* Improve button style */
.submit-button {
    display: block;
    margin: auto;
    padding: 10px 20px;
    font-size: 1.1em;
    color: white;
    background-color: #4CAF50;
    border: none;
    border-radius: 5px;
    cursor: pointer;
    transition: background-color 0.3s ease;
}
.submit-button:hover {
    background-color: #367c39;
}

/* Style the error messages */
.error-message {
    color: red;
    text-align: center;
    margin-top: 1em;
    font-weight: bold;
}
label{
    font-weight: bold; /* Make labels bold */
    display: block;      /* Each label on its own line */
    margin-bottom: 0.5em; /* Space between label and input */
}
"""


with gr.Blocks(css=css) as demo:
    gr.Markdown(
        """
        # Xylaria Iris Image Generator
        Enter a text prompt and generate an image using the Xylaria Iris model!
        """,
        elem_classes="title"
    )


    with gr.Row():
        with gr.Column():
            with gr.Group(elem_classes="input-section"):
                prompt_input = gr.Textbox(label="Enter your prompt", placeholder="e.g., A beautiful landscape with a magical tree", lines=3)
                generate_button = gr.Button("Generate Image", elem_classes="submit-button")
        with gr.Column():
            with gr.Group(elem_classes="output-section") as output_group:
                image_output = gr.Image(label="Generated Image") # Removed width and height


    def on_generate_click(prompt):
        output_group.elem_classes = ["output-section", "animate"]
        image = generate_image(prompt)
        output_group.elem_classes = ["output-section"]
        return image


    generate_button.click(on_generate_click, inputs=prompt_input, outputs=image_output)
    prompt_input.submit(on_generate_click, inputs=prompt_input, outputs=image_output)

    gr.Examples(
        [["A futuristic cityscape at night"],
         ["A mystical forest with glowing mushrooms"],
         ["An astronaut exploring a new planet"],
         ["A cat wearing a top hat"]],
        inputs=prompt_input
    )

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