feat: 重构图书墙交互及多源高阶信息抓取注入系统

1. UI 层： - 抛弃网格卡片结构，改版为沉浸式的“图书墙”(1:1.43 原比例包含模式)。 - 新增悬浮深色遮罩系统(hover-details)，解绑评分点击阻塞。 - 过滤系统默认进入“在读”状态页，彻底移除冗余无效的全选按钮。 2. 抓取与刮削层 (scrape.py)： - 抛弃对 ISBN 查询 Google Books 的严重强依赖。 - 重构抓取链路，首选 Goodreads 与特供的 Books.com.tw(博客来) 高精爬取港台原版资料。 - 建立 Amazon/Goodreads 图片去码净化正则，告别低分辨率与畸形图片。 - 修复 bs4 (.select_first) 解析选择器在旧版环境的兼容性异常。 3. Agent AI 数据管线设计 (bookshelf.py)： - 开放基于 --json 与 --cover 的强力直接注入功能。 - 升级 add 命令作为最优先级的录入手段，免去繁琐的 enrich 阶段，直接连带高清图和满载元数据强插。 - 建立入库反重名智能拦截 (title + author 交叉校验)，形成完美的防呆与自动化壁垒。
2026-03-26 11:42:43 +08:00
parent 69c34d4faf
commit 0801f7c7df
9 changed files with 684 additions and 133 deletions
@@ -3,6 +3,7 @@

 import argparse
 import sys
+import sqlite3
 import db
 import web
 import scrape
@@ -11,27 +12,53 @@ import scrape
 STATUS_CHOICES = ["to-read", "reading", "read"]
 FORMAT_CHOICES = ["paper", "ebook"]

-
 def cmd_add(args):
    db.init_db()
-    book_id = db.add_book(
-        args.title,
-        author=args.author,
-        translator=args.translator,
-        publisher=args.publisher,
-        pub_date=args.pub_date,
-        cover_url=args.cover_url,
-        format=args.format,
-        status=args.status,
-        rating=args.rating,
-        douban_score=args.douban_score,
-        goodreads_score=args.goodreads_score,
-        tags=args.tags or "",
-        notes=args.notes,
-        start_date=args.start_date,
-        finish_date=args.finish_date,
-    )
-    print(f"✅ 已添加：《{args.title}》(ID: {book_id})")
+    
+    kwargs = {
+        "author": args.author, "translator": args.translator,
+        "publisher": args.publisher, "pub_date": args.pub_date,
+        "cover_url": args.cover_url, "format": args.format,
+        "status": args.status, "rating": args.rating,
+        "douban_score": args.douban_score, "goodreads_score": args.goodreads_score,
+        "tags": args.tags or "", "notes": args.notes,
+        "start_date": args.start_date, "finish_date": args.finish_date,
+        "title_en": args.title_en or "", "author_en": args.author_en or "",
+        "douban_url": args.douban_url or "", "goodreads_url": args.goodreads_url or ""
+    }
+    
+    if getattr(args, "json", None):
+        import json
+        try:
+            extra = json.loads(args.json)
+            kwargs.update(extra)
+        except Exception as e:
+            print(f"❌ JSON 解析失败: {e}")
+            return
+            
+    title = kwargs.pop("title", args.title)
+    if not title:
+        print("❌ 必须提供书名 (title)。")
+        return
+        
+    all_books = db.list_books()
+    for b in all_books:
+        db_title = b['title'].lower()
+        t_lower = title.lower()
+        if t_lower == db_title or (len(t_lower) > 3 and (t_lower in db_title or db_title in t_lower)):
+            a_in = (kwargs.get("author") or "").lower()
+            a_db = (b.get("author") or "").lower()
+            if not a_in or not a_db or a_in in a_db or a_db in a_in:
+                print(f"⚠️  库中可能已存在同名记录：[ID: {b['id']}] 《{b['title']}》 (作者: {b.get('author', '未知')})。")
+                print("   已取消添加。若确认并非同一本请修改书名/作者，若需更新请使用 `update` 命令。")
+                return
+
+    book_id = db.add_book(title)
+    updates = {k: v for k, v in kwargs.items() if v}
+    if updates:
+        db.update_book(book_id, **updates)
+        
+    print(f"✅ 已添加并录入数据：《{title}》(ID: {book_id})")


 def cmd_list(args):
@@ -48,27 +75,55 @@ def cmd_list(args):
        print(f"  {status_icon} [{b['id']:>3}] 《{b['title']}》— {b['author'] or '未知'}"
              f"  {fmt_icon}{rating_str}{tags_str}")

-
 def cmd_update(args):
    db.init_db()
    book = db.get_book(args.id)
    if not book:
        print(f"❌ 未找到 ID 为 {args.id} 的书目。")
        sys.exit(1)
+        
    updates = {}
+    
+    # 获取命令行常规参数
    for field in ["title", "author", "translator", "publisher", "pub_date",
                   "cover_url", "format", "status", "rating", "douban_score",
-                   "goodreads_score", "tags", "notes", "start_date", "finish_date"]:
+                   "goodreads_score", "tags", "notes", "start_date", "finish_date",
+                   "title_en", "author_en", "douban_url", "goodreads_url"]:
        val = getattr(args, field.replace("-", "_"), None)
        if val is not None:
            updates[field] = val
+            
+    # 从 JSON 批量注入参数（AI优先）
+    if getattr(args, "json", None):
+        import json
+        try:
+            updates.update(json.loads(args.json))
+        except Exception as e:
+            print(f"❌ JSON 解析失败: {e}")
+            return
+            
+    # 从本地图片注入封面（AI优先）
+    if getattr(args, "cover", None):
+         import os
+         if not os.path.exists(args.cover):
+             print(f"❌ 找不到指定的本封面: {args.cover}")
+             return
+         try:
+             with open(args.cover, "rb") as f:
+                 updates["cover_blob"] = f.read()
+                 updates["cover_local"] = 1
+             print(f"🖼️  已应用本地封面: {args.cover}")
+         except Exception as e:
+             print(f"❌ 读取封面文件失败: {e}")
+             return
+
    if not updates:
        print("⚠️  未指定任何要更新的字段。")
        return
+        
    db.update_book(args.id, **updates)
    print(f"✅ 已更新：《{book['title']}》(ID: {args.id})")

-
 def cmd_delete(args):
    db.init_db()
    book = db.get_book(args.id)
@@ -139,12 +194,187 @@ def cmd_fetch(args):
            douban_score=data.get("douban_score"),
            goodreads_score=data.get("goodreads_score"),
            tags=data.get("tags", ""),
+            title_en=data.get("title_en", ""),
+            author_en=data.get("author_en", ""),
+            douban_url=data.get("douban_url", ""),
+            goodreads_url=data.get("goodreads_url", ""),
        )
        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 _is_match(b_title, b_author, s_title, s_title_en, s_author):
+    import re
+    def canon(s):
+        return re.sub(r'[^\w\u4e00-\u9fff]', '', str(s).lower()) if s else ""
+    bt = canon(b_title)
+    st = canon(s_title)
+    ste = canon(s_title_en)
+    title_match = (bt in st or st in bt) if (bt and st) else False
+    if not title_match and ste:
+        title_match = (bt in ste or ste in bt) if bt else False
+    ba = canon(b_author)
+    sa = canon(s_author)
+    author_match = (ba in sa or sa in ba) if (ba and sa) else False
+    # 书名匹配或作者匹配之一满足即可认为是同一个记录
+    return title_match or author_match
+
+
+def cmd_enrich(args):
+    """自动补全数据库中缺失元数据的书目。"""
+    db.init_db()
+    books = db.list_books()
+    to_enrich = []
+    for b in books:
+        # 只要没有豆瓣或 GR 的链接，或者没有封面，就尝试 enrich
+        if not b.get("cover_url") or not b.get("douban_url") or not b.get("goodreads_url"):
+            to_enrich.append(b)
+
+    if args.id:
+        to_enrich = [b for b in to_enrich if b["id"] == args.id]
+
+    if not to_enrich:
+        print("🎉 所有书目数据已相对完整，无需 enrich。")
+        return
+
+    print(f"🔍 共有 {len(to_enrich)} 本书需补全或验证信息...")
+    import urllib.parse
+    
+    for b in to_enrich:
+        print(f"\n=> 正在处理 [{b['id']}]: 《{b['title']}》 作者: {b.get('author','')} ...")
+        
+        scraped = None
+        src = ""
+        
+        search_term = b['title']
+        if b.get('author'): search_term += f" {b['author']}"
+        
+        # 策略 1: 直接向 Goodreads 询问（它对原版图书封面最标准、非促销封套）
+        print("   🔍 尝试通过 Goodreads 直搜...")
+        gr_data = scrape.fetch_goodreads(search_term)
+        if gr_data and _is_match(b['title'], b.get('author'), gr_data.get('title'), gr_data.get('title_en'), gr_data.get('author')):
+            scraped = gr_data
+            src = "Goodreads"
+
+        # 策略 2: 如果 GR 失败或缺信息，对于繁体中文，尝试博客来
+        if not scraped or not scraped.get("cover_url"):
+            print("   🔍 尝试通过博客来(Books.com.tw)直搜...")
+            tw_data = scrape.fetch_books_tw(search_term)
+            if tw_data and _is_match(b['title'], b.get('author'), tw_data.get('title'), None, tw_data.get('author')):
+                if not scraped: scraped = tw_data
+                else:
+                    for k, v in tw_data.items():
+                        if not scraped.get(k): scraped[k] = v
+                scraped["format"] = "paper" # 博客来主要是实体书
+                src = "博客来" + ((" / " + src) if src else "")
+        
+        # 策略 3: Google Books 仅作为属性补充，**决不拿它的 ISBN 去瞎关联豆瓣**
+        query = f"intitle:{b['title']}"
+        if b.get('author'): query += f"+inauthor:{b['author']}"
+        q = urllib.parse.quote(query)
+        url = f"https://www.googleapis.com/books/v1/volumes?q={q}"
+        try:
+            html = scrape._get(url)
+            if html:
+                import json
+                data = json.loads(html)
+                if data.get("totalItems", 0) > 0:
+                    vol = data["items"][0].get("volumeInfo", {})
+                    # 只要匹配度过关，就拿它的 publisher, pub_date, ISBN
+                    if _is_match(b['title'], b.get('author'), vol.get("title"), None, ",".join(vol.get("authors", []))):
+                        isbn = next((ids.get("identifier") for ids in vol.get("industryIdentifiers", []) if ids.get("type") in ("ISBN_13", "ISBN_10")), None)
+                        if not scraped: scraped = {"title": vol.get("title")}
+                        if vol.get("publisher") and not scraped.get("publisher"): scraped["publisher"] = vol["publisher"]
+                        if vol.get("publishedDate") and not scraped.get("pub_date"): scraped["pub_date"] = vol["publishedDate"][:7]
+                        if isbn and not scraped.get("isbn"): scraped["isbn"] = isbn
+                        src = src or "Google Books"
+        except Exception:
+            pass
+
+        # 策略 4: 如果有 ISBN，去豆瓣白嫖评分和出版信息（但为了防止封面被覆盖为促销版，不轻易覆写封面）
+        if scraped and scraped.get("isbn"):
+             douban_data = scrape.fetch_douban(scraped["isbn"])
+             if douban_data and _is_match(b['title'], b.get('author'), douban_data.get('title'), douban_data.get('title_en'), douban_data.get('author')):
+                 for k, v in douban_data.items():
+                     if k == "cover_url" and scraped.get("cover_url"): continue # 优先信任前面外站的原版封面
+                     if not scraped.get(k): scraped[k] = v
+                 src += " / 豆瓣"
+        
+        if scraped:
+             updates = {}
+             # 新刮削的标题如果是外文但不等于原书名，放到 title_en 保护原中文书名
+             if scraped.get("title") and not _is_match(b["title"], None, scraped["title"], None, None):
+                 if not b.get("title_en"):
+                     updates["title_en"] = scraped["title"]
+             if scraped.get("author") and not _is_match(b["author"], None, scraped["author"], None, None):
+                 if not b.get("author_en"):
+                     updates["author_en"] = scraped["author"]
+                 
+             for k in ["author", "translator", "publisher", "pub_date", "cover_url", "tags", 
+                       "douban_score", "goodreads_score", "isbn", "title_en", "author_en", "douban_url", "goodreads_url"]:
+                 if scraped.get(k) and not b.get(k):
+                     updates[k] = scraped[k]
+             if updates:
+                 db.update_book(b["id"], **updates)
+                 print(f"   ✅ 已补全字段: {', '.join(updates.keys())} (来源: {src})")
+             else:
+                 print(f"   👍 已匹配到信息，但无需补充新字段 (来源: {src})")
+        else:
+             print("   🚫 彻底失败，保留原样。")
+             
+        # 延时避免被反爬
+        scrape._sleep()
+
+
+def cmd_covers(args):
+    """批量下载封面存入数据库 BLOB。"""
+    db.init_db()
+    books = db.list_books()
+    to_download = [b for b in books if b.get("cover_url") and not b.get("cover_blob")]
+    
+    if args.id:
+        to_download = [b for b in to_download if b["id"] == args.id]
+        
+    print(f"🖼️  共有 {len(to_download)} 本书需要下载封面到数据库...")
+    
+    conn = db._connect()
+    try:
+        import requests
+    except ImportError:
+        import urllib.request
+    
+    count = 0
+    for b in to_download:
+        url = b["cover_url"]
+        print(f"下载 [{b['id']}] 《{b['title']}》封面: {url}")
+        try:
+            # 豆瓣防盗链需要 Referer，而且可能需要 Playwright 那个 UA
+            headers = {"User-Agent": scrape._UA, "Referer": "https://book.douban.com/"}
+            if scrape._HAS_REQUESTS:
+                resp = requests.get(url, headers=headers, timeout=15)
+                if resp.status_code == 200:
+                    blob = resp.content
+                else:
+                    blob = None
+            else:
+                req = urllib.request.Request(url, headers=headers)
+                with urllib.request.urlopen(req, timeout=15) as resp:
+                    blob = resp.read()
+            
+            if blob:
+                conn.execute("UPDATE books SET cover_blob = ? WHERE id = ?", (sqlite3.Binary(blob), b["id"]))
+                conn.commit()
+                print("   ✅ 成功")
+                count += 1
+            else:
+                print("   ❌ 获取失败")
+            scrape._sleep()
+        except Exception as e:
+             print(f"   ❌ 网络错误: {e}")
+             
+    conn.close()
+    print(f"🎉 封面下载完成，成功 {count} 本。")
 def cmd_build(args):
    db.init_db()
    output_path = web.build()
@@ -175,6 +405,10 @@ def main():
    p_add.add_argument("--notes", help="笔记/简评")
    p_add.add_argument("--start-date", help="开始阅读日期")
    p_add.add_argument("--finish-date", help="完成阅读日期")
+    p_add.add_argument("--title-en", help="外文原名")
+    p_add.add_argument("--author-en", help="外文作者名")
+    p_add.add_argument("--douban-url", help="豆瓣链接")
+    p_add.add_argument("--goodreads-url", help="Goodreads 链接")
    p_add.set_defaults(func=cmd_add)

    # --- list ---
@@ -202,6 +436,10 @@ def main():
    p_upd.add_argument("--notes", help="笔记/简评")
    p_upd.add_argument("--start-date", help="开始阅读日期")
    p_upd.add_argument("--finish-date", help="完成阅读日期")
+    p_upd.add_argument("--title-en", help="外文原名")
+    p_upd.add_argument("--author-en", help="外文作者名")
+    p_upd.add_argument("--douban-url", help="豆瓣链接")
+    p_upd.add_argument("--goodreads-url", help="Goodreads 链接")
    p_upd.set_defaults(func=cmd_update)

    # --- delete ---
@@ -218,10 +456,27 @@ def main():
    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)

+    # --- enrich ---
+    p_enrich = sub.add_parser("enrich", help="批量刮削补全缺失信息")
+    p_enrich.add_argument("--id", type=int, help="只补全特定 ID 的书")
+    p_enrich.set_defaults(func=cmd_enrich)
+
+    # --- covers ---
+    p_covers = sub.add_parser("covers", help="批量下载并保存封面到数据库")
+    p_covers.add_argument("--id", type=int, help="只下载特定 ID 的书")
+    p_covers.set_defaults(func=cmd_covers)
+
    # --- build ---
    p_build = sub.add_parser("build", help="生成展示网页")
    p_build.set_defaults(func=cmd_build)

+    # --- update ---
+    p_update = sub.add_parser("update", help="(高级/AI专用)通过 JSON 直接更新元数据或本地封面")
+    p_update.add_argument("id", type=int, help="书籍记录的数字 ID")
+    p_update.add_argument("--json", type=str, help="包含需更新字段的 JSON 字符串 (例: '{\"douban_score\": 9.5}')")
+    p_update.add_argument("--cover", type=str, help="本地图片文件路径，直接注入 cover_blob")
+    p_update.set_defaults(func=cmd_update)
+
    args = parser.parse_args()
    args.func(args)