change download fifter count

Readme.md

@@ -19,7 +19,16 @@
 ## 安装依赖
 
 ```bash
+# 安装 bilibili-api
 pip install bilibili-api-python
+
+# 安装 you-get（用于下载视频）
+pip install you-get
+
+# 安装其他依赖库
+pip3 install aiohttp
+pip3 install httpx
+pip3 install "curl_cffi"
 ```
 
 ## 安装 ffmpeg

spider.py

@@ -13,6 +13,7 @@ from bilibili_api import HEADERS, get_client, video  # pip install bilibili-api-
 BASE_DIR = Path(__file__).resolve().parent
 OUTPUT_DIR = BASE_DIR / "downloads"
 DOWNLOADED_BVID_FILE = BASE_DIR / "downloaded_bvids.txt"  # 已下载 bvid 记录文件
+SKIPPED_BVID_FILE = BASE_DIR / "skipped_bvids.txt"  # 跳过的 bvid 记录文件
 FFMPEG_PATH = None  # Will be set by find_ffmpeg()
 POLL_SECONDS = 2
 FETCH_RELATED_LIMIT = 20
@@ -73,6 +74,15 @@ def save_downloaded_bvid(bvid: str):
         print(f"failed to save bvid {bvid}: {e}")
 
 
+def save_skipped_bvid(bvid: str, view_count: int, title: str):
+    """将跳过的 bvid 及其信息追加保存到文件"""
+    try:
+        with open(SKIPPED_BVID_FILE, "a", encoding="utf-8") as f:
+            f.write(f"{bvid}\t{view_count}\t{title}\n")
+    except Exception as e:
+        print(f"failed to save skipped bvid {bvid}: {e}")
+
+
 def get_bvid_from_url(url: str) -> str:
     match = re.search(r"BV[0-9A-Za-z]{10}", url)
     if not match:
@@ -81,13 +91,14 @@ def get_bvid_from_url(url: str) -> str:
 
 
 def sanitize_title(title: str, max_length: int = 80) -> str:
-    cleaned = re.sub(r'[\\/:*?"<>|]+', "_", title).strip()
-    cleaned = re.sub(r"\s+", "_", cleaned)
-    cleaned = re.sub(r"_+", "_", cleaned)
-    cleaned = cleaned.strip("._")
+    # 只保留字母、数字和中文，其他全部移除
+    cleaned = re.sub(r'[^\w\u4e00-\u9fff]', '', title)
+    # 移除连续下划线
+    cleaned = re.sub(r'_+', '_', cleaned)
+    cleaned = cleaned.strip('_')
     if not cleaned:
         cleaned = "video"
-    return cleaned[:max_length].rstrip("._")
+    return cleaned[:max_length].rstrip('_')
 
 
 def build_output_file_name(bvid: str, title: str) -> str:
@@ -217,15 +228,54 @@ async def download_video_file(bvid: str, output_file_name: str):
     return final_path
 
 
+def download_with_youget(bvid: str, title: str, output_dir: Path) -> bool:
+    """使用 you-get 下载视频"""
+    url = f"https://www.bilibili.com/video/{bvid}"
+    # 生成自定义文件名：bvid_清理后的标题
+    custom_filename = f"{bvid}_{sanitize_title(title)}"
+    try:
+        # 使用 you-get 下载视频，指定文件名，禁用字幕
+        result = subprocess.run(
+            ["you-get", "-o", str(output_dir), "-O", custom_filename, "--no-caption", url],
+            capture_output=True,
+            text=True,
+            check=True
+        )
+        print(f"you-get output: {result.stdout}")
+        return True
+    except subprocess.CalledProcessError as e:
+        print(f"you-get download failed: {e.stderr}")
+        return False
+    except FileNotFoundError:
+        print("you-get not found, please install it: pip install you-get")
+        return False
+
+
 def process_download(bvid: str):
     """处理单个视频的下载"""
     try:
         # 获取视频信息
         info, related_items = asyncio.run(fetch_video_info_and_related(bvid))
         title = info.get("title", bvid)
-        file_name = build_output_file_name(bvid, title)
         
-        print(f"fetching info done: {bvid}, title: {title}")
+        # 获取观看量
+        view_count = info.get("stat", {}).get("view", 0)
+        
+        print(f"fetching info done: {bvid}, title: {title}, views: {view_count}")
+        
+        # 判断观看量是否超过 50w (500000)
+        if view_count < 500000:
+            print(f"skipped: {bvid}, view count {view_count} < 500000")
+            # 记录跳过的视频信息
+            save_skipped_bvid(bvid, view_count, title)
+            # 仍然添加相关视频到队列
+            inserted, skipped = save_related_bvids(bvid, related_items)
+            print(f"related videos: inserted={inserted}, skipped={skipped}")
+            return
+        
+        print(f"view count {view_count} >= 500000, downloading...")
+        
+        file_name = build_output_file_name(bvid, title)
         
         # 检查文件是否已存在
         final_path = get_output_path(file_name)
@@ -235,14 +285,18 @@ def process_download(bvid: str):
                 downloaded_set.add(bvid)
             return
         
-        # 下载视频
-        final_path = asyncio.run(download_video_file(bvid, file_name))
-        print(f"download done: {bvid} -> {final_path}")
+        # 使用 you-get 下载视频
+        success = download_with_youget(bvid, title, OUTPUT_DIR)
+        
+        if success:
+            print(f"download done: {bvid}")
             
             # 标记为已下载并保存到文件
             with queue_lock:
                 downloaded_set.add(bvid)
             save_downloaded_bvid(bvid)
+        else:
+            print(f"download failed: {bvid}")
         
         # 添加相关视频到队列
         inserted, skipped = save_related_bvids(bvid, related_items)