# Copyright Volkan Sah! Do not steal code if its for free! Respect creators!
# You can use it for free a star or follow will be greate!
import gradio as gr
import numpy as np
import random
import os
import spaces
import time
from diffusers import AutoPipelineForText2Image
import torch
from huggingface_hub import login

login(os.environ.get("HF_TOKEN"))
device = "cuda" if torch.cuda.is_available() else "cpu"
torch_dtype = torch.float16 if torch.cuda.is_available() else torch.float32

pipe = AutoPipelineForText2Image.from_pretrained("black-forest-labs/FLUX.1-dev", torch_dtype=torch.bfloat16)
pipe.load_lora_weights('enhanceaiteam/Flux-uncensored', weight_name='lora.safetensors')
pipe = pipe.to(device)

MAX_SEED = np.iinfo(np.int32).max
MAX_IMAGE_SIZE = 2048

def dev_pipeline_test(prompt, width, height, steps):
    """Simulates pipeline tests without actual image generation"""
    time.sleep(0.1)  # Simulate processing
    estimated_memory = (width * height * 3 * 4) / (1024 * 1024)  # MB
    return {
        'success': True,
        'memory_required': f"{estimated_memory:.2f}MB",
        'compute_units': steps * (width * height) / 1024**2,
        'would_generate': True
    }

@spaces.GPU
def infer(
    prompt,
    seed,
    randomize_seed,
    width,
    height,
    guidance_scale,
    num_inference_steps,
    dev_mode,
    progress=gr.Progress(track_tqdm=True),
):
    if dev_mode:
        result = dev_pipeline_test(prompt, width, height, num_inference_steps)
        # Create a black test image with debug info
        debug_image = np.zeros((height, width, 3), dtype=np.uint8)
        return debug_image, seed, f"DEV MODE: {result}"
    
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    generator = torch.Generator().manual_seed(seed)
    image = pipe(
        prompt=prompt,
        guidance_scale=guidance_scale,
        num_inference_steps=num_inference_steps,
        width=width,
        height=height,
        generator=generator,
    ).images[0]
    return image, seed, "Production generation completed"

examples = [
    "Tiger in a jungle, cold color palette, muted colors, detailed, 8k",
    "An astronaut riding a pink horse",
    "A delicious ceviche cheesecake slice",
]

css = """
#col-container {
    margin: 0 auto;
    max-width: 640px;
}
"""

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="col-container"):
        gr.Markdown("""# [FLUX.1-dev](https://blackforestlabs.ai/)
Generate any type of image with Flux-Dev (Lora: Flux-uncensored). Note: This script works well, but please use min. ZeroGPU
""")
        
        with gr.Row():
            prompt = gr.Text(
                label="Prompt",
                show_label=False,
                max_lines=1,
                placeholder="Enter your prompt",
                container=False,
            )
            run_button = gr.Button("Run", scale=0, variant="primary")
        
        result = gr.Image(label="Result", show_label=False)
        status_text = gr.Text(label="Status", show_label=True)
        
        with gr.Accordion("Advanced Settings", open=False):
            dev_mode = gr.Checkbox(label="Developer Mode (No actual generation)", value=False)
            seed = gr.Slider(
                label="Seed",
                minimum=0,
                maximum=MAX_SEED,
                step=1,
                value=0,
            )
            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
            
            with gr.Row():
                width = gr.Slider(
                    label="Width",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=1024,
                )
                height = gr.Slider(
                    label="Height",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=1024,
                )
            
            with gr.Row():
                guidance_scale = gr.Slider(
                    label="Guidance scale",
                    minimum=0.0,
                    maximum=10.0,
                    step=0.1,
                    value=3.5,
                )
                num_inference_steps = gr.Slider(
                    label="Number of inference steps",
                    minimum=1,
                    maximum=50,
                    step=1,
                    value=2,
                )
        
        gr.Examples(examples=examples, inputs=[prompt])
        
    gr.on(
        triggers=[run_button.click, prompt.submit],
        fn=infer,
        inputs=[
            prompt,
            seed,
            randomize_seed,
            width,
            height,
            guidance_scale,
            num_inference_steps,
            dev_mode,
        ],
        outputs=[result, seed, status_text],
    )

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