app.py

import gradio as gr
import requests
from concurrent.futures import ThreadPoolExecutor
import os
import re
import zipfile
import tempfile
import shutil
from urllib.parse import urlparse
import time
from selenium import webdriver
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.chrome.service import Service
from selenium.webdriver.common.by import By
from selenium.common.exceptions import TimeoutException, WebDriverException
import subprocess
import sys

def setup_chrome_driver():
    """設置Chrome WebDriver，適用於Hugging Face Spaces"""
    chrome_options = Options()
    chrome_options.add_argument("--headless")
    chrome_options.add_argument("--no-sandbox")
    chrome_options.add_argument("--disable-dev-shm-usage")
    chrome_options.add_argument("--disable-gpu")
    chrome_options.add_argument("--window-size=1920,1080")
    chrome_options.add_argument("--user-agent=Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36")
    
    try:
        # 嘗試使用系統中的Chrome
        driver = webdriver.Chrome(options=chrome_options)
        return driver
    except Exception as e:
        print(f"Chrome WebDriver setup failed: {e}")
        return None

def extract_images_from_pinterest(url, max_images, progress_callback=None):
    """從Pinterest頁面提取圖片URL"""
    driver = setup_chrome_driver()
    if not driver:
        return [], "無法啟動Chrome WebDriver"
    
    try:
        driver.get(url)
        time.sleep(3)
        
        imgList = []
        scroll = 0
        no_new_images_count = 0
        
        while len(imgList) < max_images:
            # 滾動頁面
            scroll += 800
            driver.execute_script(f'window.scrollTo(0, {scroll})')
            time.sleep(1)
            
            # 獲取圖片元素
            imgs = driver.find_elements(By.CSS_SELECTOR, 'div[data-test-id="pin"] img')
            new_images = 0
            
            for img in imgs:
                try:
                    img_url = img.get_attribute('src')
                    if img_url and img_url not in imgList and len(imgList) < max_images:
                        imgList.append(img_url)
                        new_images += 1
                except:
                    continue
            
            if progress_callback:
                progress_callback(f"已找到 {len(imgList)} 張圖片")
            
            # 檢查是否沒有新圖片
            if new_images == 0:
                no_new_images_count += 1
            else:
                no_new_images_count = 0
                
            if no_new_images_count >= 5:
                break
                
        return imgList, None
        
    except Exception as e:
        return [], f"提取圖片時出錯: {str(e)}"
    finally:
        driver.quit()

def download_image(args):
    """下載單張圖片"""
    index, url, temp_dir = args
    try:
        # 轉換為高清圖片URL
        if '236x' in url:
            url = url.replace('236x', 'originals')
        elif '474x' in url:
            url = url.replace('474x', 'originals')
        
        # 生成文件名
        filename = f"pinterest_img_{index+1:04d}"
        
        # 從URL提取原始文件名
        url_parts = url.split('/')
        if len(url_parts) > 0:
            original_name = url_parts[-1].split('?')[0]
            if '.' in original_name and len(original_name) < 100:
                clean_name = re.sub(r'[^\w\-_\.]', '_', original_name)
                filename = f"pinterest_img_{index+1:04d}_{clean_name}"
        
        # 確保文件擴展名
        if not filename.lower().endswith(('.jpg', '.jpeg', '.png', '.gif', '.webp')):
            filename += '.jpg'
        
        filepath = os.path.join(temp_dir, filename)
        
        # 下載圖片
        headers = {
            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36'
        }
        
        response = requests.get(url, headers=headers, timeout=30)
        response.raise_for_status()
        
        with open(filepath, 'wb') as f:
            f.write(response.content)
        
        return True, filename
        
    except Exception as e:
        return False, f"下載失敗: {str(e)}"

def scrape_pinterest_images(pinterest_url, num_images, progress=gr.Progress()):
    """主要的爬蟲函數"""
    if not pinterest_url:
        return None, "請輸入Pinterest URL"
    
    if num_images <= 0:
        return None, "圖片數量必須大於0"
    
    if num_images > 500:  # 限制最大數量
        return None, "圖片數量不能超過500張"
    
    # 驗證URL
    try:
        parsed_url = urlparse(pinterest_url)
        if 'pinterest.com' not in parsed_url.netloc:
            return None, "請輸入有效的Pinterest URL"
    except:
        return None, "URL格式無效"
    
    progress(0, desc="開始提取圖片URL...")
    
    # 提取圖片URL
    def update_progress(msg):
        progress(0.3, desc=msg)
    
    img_urls, error = extract_images_from_pinterest(pinterest_url, num_images, update_progress)
    
    if error:
        return None, error
    
    if not img_urls:
        return None, "未找到任何圖片"
    
    progress(0.5, desc=f"開始下載 {len(img_urls)} 張圖片...")
    
    # 創建臨時目錄
    temp_dir = tempfile.mkdtemp()
    
    try:
        # 準備下載參數
        download_args = [(i, url, temp_dir) for i, url in enumerate(img_urls)]
        
        # 多線程下載
        successful_downloads = []
        failed_downloads = []
        
        with ThreadPoolExecutor(max_workers=3) as executor:  # 降低並發數
            results = list(executor.map(download_image, download_args))
        
        for i, (success, info) in enumerate(results):
            if success:
                successful_downloads.append(info)
            else:
                failed_downloads.append(f"圖片 {i+1}: {info}")
            
            # 更新進度
            progress((0.5 + 0.4 * (i + 1) / len(results)), 
                    desc=f"已下載 {len(successful_downloads)} / {len(img_urls)} 張圖片")
        
        if not successful_downloads:
            return None, "所有圖片下載失敗"
        
        progress(0.9, desc="創建ZIP文件...")
        
        # 創建ZIP文件
        zip_path = os.path.join(tempfile.gettempdir(), "pinterest_images.zip")
        with zipfile.ZipFile(zip_path, 'w') as zipf:
            for filename in os.listdir(temp_dir):
                file_path = os.path.join(temp_dir, filename)
                if os.path.isfile(file_path):
                    zipf.write(file_path, filename)
        
        progress(1.0, desc="完成!")
        
        # 準備結果信息
        result_info = f"""
下載完成！

成功下載: {len(successful_downloads)} 張圖片
失敗: {len(failed_downloads)} 張圖片
總計: {len(img_urls)} 張圖片

請點擊下方鏈接下載ZIP文件。
        """
        
        if failed_downloads:
            result_info += f"\n\n失敗詳情:\n" + "\n".join(failed_downloads[:10])  # 只顯示前10個錯誤
        
        return zip_path, result_info
        
    except Exception as e:
        return None, f"處理過程中出錯: {str(e)}"
    finally:
        # 清理臨時目錄
        try:
            shutil.rmtree(temp_dir)
        except:
            pass

# 創建Gradio界面
def create_interface():
    with gr.Blocks(title="Pinterest圖片下載器", theme=gr.themes.Soft()) as interface:
        gr.Markdown("""
        # 🖼️ Pinterest 圖片下載器
        
        輸入Pinterest搜索頁面的URL，批量下載圖片。
        
        **使用說明:**
        1. 輸入Pinterest搜索頁面或板塊的完整URL
        2. 設置要下載的圖片數量（建議不超過100張）
        3. 點擊"開始下載"按鈕
        4. 等待處理完成後下載ZIP文件
        
        **注意事項:**
        - 請確保輸入的是有效的Pinterest URL
        - 下載速度取決於網絡狀況和圖片大小
        - 建議單次下載不超過100張圖片
        """)
        
        with gr.Row():
            with gr.Column():
                pinterest_url = gr.Textbox(
                    label="Pinterest URL",
                    placeholder="https://www.pinterest.com/search/pins/?q=your-search-term",
                    lines=2,
                    info="輸入Pinterest搜索頁面或板塊的完整URL"
                )
                
                num_images = gr.Slider(
                    minimum=1,
                    maximum=500,
                    value=20,
                    step=1,
                    label="圖片數量",
                    info="要下載的圖片數量（建議不超過100張）"
                )
                
                download_btn = gr.Button("🚀 開始下載", variant="primary", size="lg")
            
            with gr.Column():
                result_info = gr.Textbox(
                    label="下載結果",
                    lines=10,
                    interactive=False,
                    info="顯示下載進度和結果信息"
                )
                
                download_file = gr.File(
                    label="下載文件",
                    interactive=False,
                    visible=False
                )
        
        # 示例URL
        gr.Markdown("""
        ### 示例URL:
        ```
        https://www.pinterest.com/search/pins/?q=landscape%20photography
        https://www.pinterest.com/search/pins/?q=interior%20design
        https://www.pinterest.com/search/pins/?q=food%20photography
        ```
        """)
        
        # 綁定事件
        def handle_download(url, num):
            zip_path, info = scrape_pinterest_images(url, int(num))
            if zip_path:
                return info, gr.File(value=zip_path, visible=True)
            else:
                return info, gr.File(visible=False)
        
        download_btn.click(
            fn=handle_download,
            inputs=[pinterest_url, num_images],
            outputs=[result_info, download_file],
            show_progress=True
        )
    
    return interface

# 啟動應用
if __name__ == "__main__":
    interface = create_interface()
    interface.launch(
        server_name="0.0.0.0",
        server_port=7860,
        share=True
    )