app.py

import requests
import os
import gradio as gr
from gradio_client import Client, handle_file
import random


HF_TOKEN = os.environ.get("girlToken")
TARGET_SPACE_URL = "https://prithivmlmods-qwen-image-edit-2511-loras-fast.hf.space"

def upload_file_to_space(local_path):
    """手动上传文件到目标 Space，返回远端 URL"""
    upload_url = f"{TARGET_SPACE_URL}/upload"
    
    headers = {}
    if HF_TOKEN:
        headers["Authorization"] = f"Bearer {HF_TOKEN}"
    
    with open(local_path, "rb") as f:
        response = requests.post(
            upload_url,
            headers=headers,
            files={"files": (os.path.basename(local_path), f, "image/jpeg")},
        )
    
    print(f"上传状态码: {response.status_code}")
    print(f"上传响应: {response.text}")
    
    if response.status_code == 200:
        result = response.json()
        # 返回的是文件路径列表，取第一个
        remote_path = result[0] if isinstance(result, list) else result
        return remote_path
    else:
        raise Exception(f"上传失败: {response.status_code} {response.text}")


def infer(image, prompt, lora_adapter, seed, randomize_seed, guidance_scale, steps, progress=gr.Progress(track_tqdm=True)):
    if image is None or not os.path.exists(image):
        return None, seed

    if randomize_seed:
        seed = random.randint(0, MAX_SEED)

    # 手动上传，拿到远端路径
    remote_path = upload_file_to_space(image)
    print(f"远端路径: {remote_path}")

    # 用远端路径构造 images 参数
    images_input = [{
        "image": {
            "path": remote_path,
            "url": f"{TARGET_SPACE_URL}/file={remote_path}",
            "size": os.path.getsize(image),
            "orig_name": os.path.basename(image),
            "mime_type": "image/jpeg",
            "is_stream": False,
            "meta": {}
        },
        "caption": None
    }]

    print(f"images_input: {images_input}")

    try:
        result = space_client.predict(
            images=images_input,
            prompt=prompt,
            lora_adapter=lora_adapter,
            seed=float(seed),
            randomize_seed=bool(randomize_seed),
            guidance_scale=float(guidance_scale),
            steps=float(steps),
            api_name="/infer",
        )

        image_info, seed_used = result
        img_out = image_info.get("path") or image_info.get("url")
        return img_out, int(seed_used)

    except Exception as e:
        import traceback
        traceback.print_exc()
        print(f"[调用API] 异常: {e}")
        return None, seed