feat: add book scraping (Douban via Playwright + Goodreads + Google Books)

- scrape.py: Douban scraping with Playwright (multi-strategy browser launch)
- scrape.py: Goodreads scraping with requests + BeautifulSoup
- scrape.py: Google Books API + Open Library as fallback
- bookshelf.py: add 'fetch' subcommand (ISBN / Douban URL / GR URL)
- Fix Douban #info parsing using span.pl tags
- Add requirements.txt (requests, beautifulsoup4, playwright)

bookshelf.py

@@ -5,6 +5,7 @@ import argparse
 import sys
 import db
 import web
+import scrape
 
 
 STATUS_CHOICES = ["to-read", "reading", "read"]
@@ -78,6 +79,72 @@ def cmd_delete(args):
     print(f"🗑️  已删除：《{book['title']}》(ID: {args.id})")
 
 
+def cmd_fetch(args):
+    """刮削图书元数据，可选择直接写入数据库。"""
+    identifier = args.identifier
+
+    # 选择刮削策略
+    if "douban.com" in identifier:
+        print(f"🔍 正在从豆瓣刮削：{identifier}")
+        data = scrape.fetch_douban(identifier)
+        source = "豆瓣"
+    elif "goodreads.com" in identifier:
+        print(f"🔍 正在从 Goodreads 刮削：{identifier}")
+        data = scrape.fetch_goodreads(identifier)
+        source = "Goodreads"
+    elif scrape._is_isbn(identifier):
+        print(f"🔍 正在刮削 ISBN：{identifier}")
+        data, source = scrape.fetch_by_isbn(identifier)
+    else:
+        print("❌ 请提供有效的 ISBN（10/13位）、豆瓣链接或 Goodreads 链接。")
+        sys.exit(1)
+
+    if not data:
+        print("❌ 刮削失败，未能获取到图书信息。")
+        print("   提示：豆瓣偶尔会拒绝请求（418/403），请稍后重试。")
+        sys.exit(1)
+
+    # 打印预览
+    print(scrape.preview(data, source))
+    print()
+
+    if args.update_id:
+        # 更新已有记录
+        db.init_db()
+        book = db.get_book(args.update_id)
+        if not book:
+            print(f"❌ 未找到 ID 为 {args.update_id} 的书目。")
+            sys.exit(1)
+        # 只覆盖刮削到的字段，保留用户已设置的字段
+        updates = {k: v for k, v in data.items() if v}
+        if args.format:
+            updates["format"] = args.format
+        if args.status:
+            updates["status"] = args.status
+        db.update_book(args.update_id, **updates)
+        print(f"✅ 已更新：《{book['title']}》→《{data.get('title', book['title'])}》(ID: {args.update_id})")
+
+    elif args.add:
+        # 写入新记录
+        db.init_db()
+        book_id = db.add_book(
+            data.get("title", ""),
+            author=data.get("author"),
+            translator=data.get("translator"),
+            publisher=data.get("publisher"),
+            pub_date=data.get("pub_date"),
+            cover_url=data.get("cover_url"),
+            format=args.format or "paper",
+            status=args.status or "to-read",
+            douban_score=data.get("douban_score"),
+            goodreads_score=data.get("goodreads_score"),
+            tags=data.get("tags", ""),
+        )
+        print(f"✅ 已添加：《{data.get('title')}》(ID: {book_id})")
+    else:
+        print("💡 加 --add 选项可直接写入数据库，或用 --update-id <ID> 更新已有记录。")
+
+
 def cmd_build(args):
     db.init_db()
     output_path = web.build()
@@ -142,6 +209,15 @@ def main():
     p_del.add_argument("id", type=int, help="书目 ID")
     p_del.set_defaults(func=cmd_delete)
 
+    # --- fetch ---
+    p_fetch = sub.add_parser("fetch", help="从豆瓣/Google Books 刮削书目信息")
+    p_fetch.add_argument("identifier", help="ISBN（10/13位）或豆瓣图书链接")
+    p_fetch.add_argument("--add", action="store_true", help="刮削后直接写入数据库")
+    p_fetch.add_argument("--update-id", type=int, metavar="ID", help="刮削后更新指定 ID 的书目")
+    p_fetch.add_argument("--format", choices=FORMAT_CHOICES, help="书籍格式（入库时使用）")
+    p_fetch.add_argument("--status", choices=STATUS_CHOICES, help="阅读状态（入库时使用）")
+    p_fetch.set_defaults(func=cmd_fetch)
+
     # --- build ---
     p_build = sub.add_parser("build", help="生成展示网页")
     p_build.set_defaults(func=cmd_build)