app.py

import base64
import dash
from dash import dcc, html, Input, Output, State
import dash_bootstrap_components as dbc
from dash.exceptions import PreventUpdate
import google.generativeai as genai
import requests
import logging
import threading
import time
import os

# Set up logging
logging.basicConfig(level=logging.DEBUG, format='%(asctime)s - %(levelname)s - %(message)s')

# Updated STYLES list
STYLES = [
    "photographic", "3d-model", "analog-film", "anime", "cinematic", "comic-book", 
    "digital-art", "enhance", "fantasy-art", "isometric", "line-art", "low-poly", 
    "modeling-compound", "neon-punk", "origami", "pixel-art", "tile-texture"
]

# Default negative prompt (hidden from UI)
DEFAULT_NEGATIVE_PROMPT = """
ugly, tiling, poorly drawn hands, poorly drawn feet, poorly drawn face, out of frame, 
extra limbs, disfigured, deformed, body out of frame, bad anatomy, watermark, signature, 
cut off, low contrast, underexposed, overexposed, bad art, beginner, amateur, distorted face,
plastic, cartoonish, artificial, fake, unnatural, blurry, smooth, lack of detail, low quality
"""

app = dash.Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP])

app.layout = dbc.Container([
    html.H1("ImaGen", className="my-4"),
    dbc.Row([
        # Left column: Form entry
        dbc.Col([
            dbc.Card([
                dbc.CardBody([
                    dbc.Textarea(id="prompt", placeholder="Enter your prompt", className="mb-3"),
                    dcc.Dropdown(
                        id="style",
                        options=[{"label": s.replace("-", " ").title(), "value": s} for s in STYLES],
                        value="photographic",
                        placeholder="Select style",
                        className="mb-3"
                    ),
                    dbc.Button("Generate Image", id="submit-btn", color="primary", className="mb-3"),
                    dbc.Accordion([
                        dbc.AccordionItem(
                            [
                                dbc.Label("Aspect Ratio"),
                                dcc.Dropdown(
                                    id="aspect-ratio",
                                    options=[
                                        {"label": ar, "value": ar} for ar in 
                                        ["16:9", "1:1", "21:9", "2:3", "3:2", "4:5", "5:4", "9:16", "9:21"]
                                    ],
                                    value="1:1"
                                ),
                                dbc.Label("Steps"),
                                dcc.Slider(id="steps", min=4, max=50, step=1, value=30, marks={4: '4', 25: '25', 50: '50'}),
                            ],
                            title="Advanced Settings",
                        ),
                    ], start_collapsed=True, className="mb-3"),
                ])
            ], className="mb-4"),
        ], width=6),
        # Right column: Image preview
        dbc.Col([
            dbc.Card([
                dbc.CardBody([
                    dcc.Loading(
                        id="loading",
                        type="circle",
                        children=[
                            html.Div(id="status-message", className="mb-3"),
                            html.Img(id="image-output", className="img-fluid mb-3"),
                            html.Div(id="enhanced-prompt-output", className="mb-3"),
                            dbc.Button("Download Image", id="download-btn", color="secondary", className="mb-3", disabled=True),
                            dcc.Download(id="download-image")
                        ]
                    ),
                ])
            ]),
        ], width=6),
    ]),
], fluid=True)

def enhance_prompt(google_api_key, prompt, style):
    genai.configure(api_key=google_api_key)
    model = genai.GenerativeModel("gemini-2.0-flash-lite")
    enhanced_prompt_request = f"""
    Task: Enhance the following prompt with details to match the specified style
    Style: {style}
    Original prompt: '{prompt}'
    
    Instructions:
    1. Expand the prompt to be more detailed, vivid, and realistic with camera used and the setting for that camera like ISO etc.
    2. Incorporate elements of the specified style.
    3. Add details that enhance the scene to the specified style
    4. Emphasize natural lighting and enhance the realism of textures and colors based on the specified style.
    5. Avoid terms that might result in artificial or cartoonish appearance unless specified by user.
    6. Maintain the original intent of the prompt while significantly improving its descriptive quality with details.
    7. Provide ONLY the enhanced prompt, without any explanations or options.
    8. Keep the enhanced prompt concise, ideally under 100 words.

    Enhanced prompt:
    """
    
    try:
        response = model.generate_content(enhanced_prompt_request)
        
        enhanced_prompt = response.text.strip()
        
        prefixes_to_remove = ["Enhanced prompt:", "Here's the enhanced prompt:", "The enhanced prompt is:"]
        for prefix in prefixes_to_remove:
            if enhanced_prompt.lower().startswith(prefix.lower()):
                enhanced_prompt = enhanced_prompt[len(prefix):].strip()
        
        logging.info(f"Enhanced prompt: {enhanced_prompt}")
        return enhanced_prompt
    except Exception as e:
        logging.error(f"Error in enhance_prompt: {str(e)}")
        raise

def generate_image(stability_api_key, enhanced_prompt, style, negative_prompt, steps, aspect_ratio):
    url = "https://api.stability.ai/v2beta/stable-image/generate/sd3"
    
    headers = {
        "Accept": "image/*",
        "Authorization": f"Bearer {stability_api_key}"
    }
    
    data = {
        "prompt": f"{enhanced_prompt}, Style: {style}, highly detailed, high quality, descriptive, sharp focus, intricate details",
        "negative_prompt": negative_prompt,
        "model": "sd3.5-large-turbo",
        "output_format": "jpeg",
        "num_images": 1,
        "steps": steps,
        "style_preset": style,
        "aspect_ratio": aspect_ratio,
    }
    
    try:
        response = requests.post(url, headers=headers, files={"none": ''}, data=data, timeout=60)
        response.raise_for_status()
        
        logging.debug(f"Response headers: {response.headers}")
        logging.debug(f"Response content type: {response.headers.get('content-type')}")
        
        if response.headers.get('content-type').startswith('image/'):
            image_data = response.content
            if len(image_data) < 1000:
                raise Exception("Received incomplete image data")
            return image_data
        else:
            error_message = response.text
            logging.error(f"Unexpected content type: {response.headers.get('content-type')}. Response: {error_message}")
            raise Exception(f"Unexpected content type: {response.headers.get('content-type')}. Response: {error_message}")
    
    except requests.exceptions.RequestException as e:
        logging.error(f"Request failed: {str(e)}")
        raise Exception(f"Request failed: {str(e)}")

def process_and_generate(google_api_key, stability_api_key, prompt, style, steps, aspect_ratio, set_status):
    try:
        set_status("Enhancing prompt...")
        enhanced_prompt = enhance_prompt(google_api_key, prompt, style)
        
        set_status("Generating image...")
        max_attempts = 3
        for attempt in range(max_attempts):
            try:
                image_bytes = generate_image(stability_api_key, enhanced_prompt, style, DEFAULT_NEGATIVE_PROMPT, steps, aspect_ratio)
                set_status("Image generated successfully!")
                return image_bytes, enhanced_prompt
            except Exception as e:
                if attempt < max_attempts - 1:
                    set_status(f"Attempt {attempt + 1} failed. Retrying...")
                    time.sleep(2)
                else:
                    raise e
    except Exception as e:
        logging.error(f"Error in process_and_generate: {str(e)}")
        set_status(f"Error: {str(e)}")
        return None, str(e)

@app.callback(
    [Output("image-output", "src"),
     Output("enhanced-prompt-output", "children"),
     Output("status-message", "children"),
     Output("download-btn", "disabled")],
    [Input("submit-btn", "n_clicks")],
    [State("prompt", "value"),
     State("style", "value"),
     State("steps", "value"),
     State("aspect-ratio", "value")],
    prevent_initial_call=True
)
def update_output(n_clicks, prompt, style, steps, aspect_ratio):
    if n_clicks is None:
        raise PreventUpdate
    
    google_api_key = os.getenv('GOOGLE_API_KEY')
    stability_api_key = os.getenv('STABILITY_API_KEY')
    
    if not google_api_key or not stability_api_key:
        return "", "Error: API keys not found in environment variables", "API keys missing", True
    
    logging.debug(f"Stability API Key (first 4 chars): {stability_api_key[:4]}...")
    
    status = {"message": "Starting process..."}
    
    def set_status(message):
        status["message"] = message
    
    def run_process():
        image_bytes, enhanced_prompt = process_and_generate(google_api_key, stability_api_key, prompt, style, steps, aspect_ratio, set_status)
        if image_bytes:
            encoded_image = base64.b64encode(image_bytes).decode('ascii')
            return f"data:image/jpeg;base64,{encoded_image}", f"Enhanced Prompt: {enhanced_prompt}", status["message"], False
        else:
            return "", f"Error: {enhanced_prompt}", status["message"], True
    
    try:
        thread = threading.Thread(target=run_process)
        thread.start()
        thread.join(timeout=90)
        
        if thread.is_alive():
            return "", "Error: Image generation timed out", "Process timed out", True
        
        return run_process()
    except Exception as e:
        logging.error(f"Unexpected error in update_output: {str(e)}")
        return "", f"Unexpected error: {str(e)}", "An unexpected error occurred", True

@app.callback(
    Output("download-image", "data"),
    Input("download-btn", "n_clicks"),
    State("image-output", "src"),
    prevent_initial_call=True
)
def download_image(n_clicks, image_src):
    if n_clicks is None:
        raise PreventUpdate
    
    image_data = image_src.split(",")[1]
    image_bytes = base64.b64decode(image_data)
    
    return dcc.send_bytes(image_bytes, "generated_image.jpeg")

if __name__ == '__main__':
    print("Starting the Dash application...")
    app.run(debug=False, host='0.0.0.0', port=7860)
    print("Dash application has finished running.")