Upload hf_backend/test_slow_download.py with huggingface_hub

hf_backend/test_slow_download.py

@@ -0,0 +1,274 @@
+#!/usr/bin/env python3
+"""
+Test multiple strategies to access slow_download and get final EPUB link.
+"""
+
+from playwright.sync_api import sync_playwright
+import time
+import sys
+import os
+
+MD5 = "d94c20d1364af9b484949659398c4062"
+SLOW_URL = f"https://annas-archive.gl/slow_download/{MD5}/0/3"
+
+def strategy_1_regular_browser():
+    """Try with a regular non-headless browser - less likely to be flagged."""
+    print("\n[策略1] 使用非无头模式浏览器...")
+    with sync_playwright() as p:
+        browser = p.chromium.launch(headless=False)
+        context = browser.new_context(
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36",
+        )
+        page = context.new_page()
+
+        try:
+            response = page.goto(SLOW_URL, timeout=120000, wait_until="domcontentloaded")
+            print(f"   状态码: {response.status if response else 'None'}")
+
+            # 等待最多60秒
+            for i in range(60):
+                time.sleep(1)
+                url = page.url
+                if url.lower().endswith('.epub'):
+                    print(f"   成功! 在第 {i+1} 秒 URL 变成 EPUB")
+                    browser.close()
+                    return url
+
+                if (i + 1) % 10 == 0:
+                    print(f"   {i+1}秒... URL: {url[:80]}...")
+
+        except Exception as e:
+            print(f"   错误: {e}")
+
+        browser.close()
+    return None
+
+def strategy_2_stealth_with_webdriver_hidden():
+    """Try with webdriver property properly hidden."""
+    print("\n[策略2] 隐藏 webdriver 属性...")
+    with sync_playwright() as p:
+        browser = p.chromium.launch(headless=True)
+        context = browser.new_context(
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36",
+            viewport={"width": 1920, "height": 1080},
+            locale="en-US",
+        )
+        page = context.new_page()
+
+        # Inject script to hide webdriver
+        page.add_init_script("""
+            Object.defineProperty(navigator, 'webdriver', {
+                get: () => undefined,
+                configurable: true
+            });
+            delete navigator.__webdriver;
+            delete navigator.__proto__?.webdriver;
+        """)
+
+        try:
+            response = page.goto(SLOW_URL, timeout=120000, wait_until="domcontentloaded")
+            print(f"   状态码: {response.status if response else 'None'}")
+
+            for i in range(60):
+                time.sleep(1)
+                url = page.url
+                if url.lower().endswith('.epub'):
+                    print(f"   成功! 在第 {i+1} 秒 URL 变成 EPUB")
+                    browser.close()
+                    return url
+
+                if (i + 1) % 10 == 0:
+                    print(f"   {i+1}秒... URL: {url[:80]}...")
+
+        except Exception as e:
+            print(f"   错误: {e}")
+
+        browser.close()
+    return None
+
+def strategy_3_wait_for_network():
+    """Try waiting for network idle instead of fixed timeout."""
+    print("\n[策略3] 等待网络空闲...")
+    with sync_playwright() as p:
+        browser = p.chromium.launch(headless=True)
+        context = browser.new_context(
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36",
+        )
+        page = context.new_page()
+
+        try:
+            page.goto(SLOW_URL, timeout=120000, wait_until="networkidle", network_idle_timeout=30)
+
+            for i in range(30):
+                time.sleep(1)
+                url = page.url
+                if url.lower().endswith('.epub'):
+                    print(f"   成功! 在第 {i+1} 秒 URL 变成 EPUB")
+                    browser.close()
+                    return url
+
+            print(f"   最终 URL: {page.url}")
+
+        except Exception as e:
+            print(f"   错误: {e}")
+
+        browser.close()
+    return None
+
+def strategy_4_check_for_download_event():
+    """Listen for download event instead of checking URL."""
+    print("\n[策略4] 监听下载事件...")
+    download_info = {"url": None}
+
+    with sync_playwright() as p:
+        browser = p.chromium.launch(headless=True)
+        context = browser.new_context(
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36",
+        )
+        page = context.new_page()
+
+        def on_download(download):
+            download_info["url"] = download.url
+            print(f"   检测到下载: {download.url}")
+
+        page.on("download", on_download)
+
+        try:
+            page.goto(SLOW_URL, timeout=120000, wait_until="domcontentloaded")
+
+            for i in range(60):
+                time.sleep(1)
+                if download_info["url"]:
+                    print(f"   在第 {i+1} 秒检测到下载")
+                    browser.close()
+                    return download_info["url"]
+
+                if (i + 1) % 10 == 0:
+                    print(f"   {i+1}秒...")
+
+        except Exception as e:
+            print(f"   错误: {e}")
+
+        browser.close()
+    return None
+
+def strategy_5_check_page_content():
+    """Check if page content eventually reveals the EPUB URL."""
+    print("\n[策略5] 检查页面内容中的 EPUB URL...")
+    with sync_playwright() as p:
+        browser = p.chromium.launch(headless=True)
+        context = browser.new_context(
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/123.0.0.0 Safari/537.36",
+        )
+        page = context.new_page()
+
+        try:
+            page.goto(SLOW_URL, timeout=120000, wait_until="domcontentloaded")
+
+            for i in range(60):
+                time.sleep(1)
+
+                # Check URL
+                url = page.url
+                if url.lower().endswith('.epub'):
+                    print(f"   URL 变成 EPUB: {url}")
+                    browser.close()
+                    return url
+
+                # Check page content for EPUB links
+                content = page.content()
+                if '.epub' in content.lower():
+                    import re
+                    # Find EPUB URLs in page
+                    matches = re.findall(r'href=["\']([^"\']*\.epub[^"\']*)["\']', content, re.IGNORECASE)
+                    if matches:
+                        print(f"   从页面内容找到 EPUB: {matches[0]}")
+                        browser.close()
+                        return matches[0]
+
+                    # Also try meta refresh or javascript redirects
+                    matches = re.findall(r'url\s*[=:]\s*["\']([^"\']*\.epub[^"\']*)["\']', content, re.IGNORECASE)
+                    if matches:
+                        print(f"   从 JS 找到 EPUB: {matches[0]}")
+                        browser.close()
+                        return matches[0]
+
+                if (i + 1) % 10 == 0:
+                    title = page.title()
+                    print(f"   {i+1}秒... 标题: {title}")
+
+        except Exception as e:
+            print(f"   错误: {e}")
+
+        browser.close()
+    return None
+
+def strategy_6_firefox_instead():
+    """Try with Firefox instead of Chromium."""
+    print("\n[策略6] 使用 Firefox 浏览器...")
+    with sync_playwright() as p:
+        try:
+            browser = p.firefox.launch(headless=True)
+        except Exception as e:
+            print(f"   Firefox 启动失败: {e}")
+            return None
+
+        context = browser.new_context(
+            user_agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:123.0) Gecko/20100101 Firefox/123.0",
+        )
+        page = context.new_page()
+
+        try:
+            response = page.goto(SLOW_URL, timeout=120000, wait_until="domcontentloaded")
+            print(f"   状态码: {response.status if response else 'None'}")
+
+            for i in range(60):
+                time.sleep(1)
+                url = page.url
+                if url.lower().endswith('.epub'):
+                    print(f"   成功! 在第 {i+1} 秒 URL 变成 EPUB")
+                    browser.close()
+                    return url
+
+                if (i + 1) % 10 == 0:
+                    print(f"   {i+1}秒... URL: {url[:80]}...")
+
+        except Exception as e:
+            print(f"   错误: {e}")
+
+        browser.close()
+    return None
+
+def main():
+    print("=" * 60)
+    print("Anna's Archive slow_download 访问测试")
+    print("=" * 60)
+    print(f"目标: {SLOW_URL}")
+    print("测试多种策略看哪种能绕过 DDoS-Guard...\n")
+
+    strategies = [
+        ("策略1-非无头浏览器", strategy_1_regular_browser),
+        ("策略2-Stealth模式", strategy_2_stealth_with_webdriver_hidden),
+        ("策略3-等待网络空闲", strategy_3_wait_for_network),
+        ("策略4-监听下载事件", strategy_4_check_for_download_event),
+        ("策略5-检查页面内容", strategy_5_check_page_content),
+        ("策略6-Firefox", strategy_6_firefox_instead),
+    ]
+
+    for name, func in strategies:
+        print(f"\n{'='*60}")
+        print(f"尝试: {name}")
+        print("=" * 60)
+        result = func()
+        if result:
+            print(f"\n*** 成功! 最终链接: {result} ***")
+            return result
+        print(f"   {name} 失败")
+
+    print("\n" + "=" * 60)
+    print("所有策略都失败了")
+    print("=" * 60)
+    return None
+
+if __name__ == "__main__":
+    main()