1#!/usr/bin/env python
2"""
3将小红书笔记按单独文件导出到 OpenClaw workspace
4 
5用法:
6    python export_to_workspace.py [db_path] [output_dir]
7 
8默认:
9    db_path: Volume/Download/ExploreData.db
10    output_dir: ~/.openclaw/workspace/xhs-memory
11 
12导出格式类似 gpt-history，每条笔记一个文件，文件名格式: YYYY-MM-标题.md
13"""
14import sqlite3
15import re
16import sys
17from pathlib import Path
18 
19 
20def sanitize_filename(name: str, max_len: int = 50) -> str:
21    """清理文件名，移除非法字符"""
22    name = re.sub(r'[<>:"/\\|?*\n\r\t]', '', name)
23    name = re.sub(r'\s+', '-', name.strip())
24    name = re.sub(r'-+', '-', name)
25    name = name.strip('-')
26    if len(name) > max_len:
27        name = name[:max_len].rstrip('-')
28    return name or "无标题"
29 
30 
31def export_to_workspace(db_path: Path = None, output_dir: Path = None):
32    db_path = db_path or Path("Volume/Download/ExploreData.db")
33    output_dir = output_dir or Path.home() / ".openclaw/workspace/xhs-memory"
34    output_dir.mkdir(parents=True, exist_ok=True)
35 
36    if not db_path.exists():
37        print(f"错误: 数据库不存在: {db_path}")
38        return False
39 
40    conn = sqlite3.connect(db_path)
41    cursor = conn.cursor()
42 
43    cursor.execute("""
44        SELECT 作品标题, 发布时间, 作品链接, 作品描述, 作者昵称, 作品标签
45        FROM explore_data
46        ORDER BY 发布时间 DESC
47    """)
48 
49    rows = cursor.fetchall()
50    conn.close()
51 
52    if not rows:
53        print("数据库为空")
54        return False
55 
56    count = 0
57    for title, time, link, desc, author, tags in rows:
58        # 解析时间: 2026-01-25_18:17:43 -> 2026-01
59        if time:
60            date_prefix = time[:7]  # YYYY-MM
61            full_date = time.replace('_', ' ')
62        else:
63            date_prefix = "unknown"
64            full_date = "未知"
65 
66        # 生成文件名
67        safe_title = sanitize_filename(title or "无标题")
68        filename = f"{date_prefix}-{safe_title}.md"
69        filepath = output_dir / filename
70 
71        # 避免重复文件名
72        counter = 1
73        while filepath.exists():
74            filename = f"{date_prefix}-{safe_title}-{counter}.md"
75            filepath = output_dir / filename
76            counter += 1
77 
78        # 生成内容
79        content = f"# {title or '无标题'}\n\n"
80        content += f"**来源**: 小红书收藏/点赞\n\n"
81        content += f"**日期**: {full_date}\n\n"
82        content += f"**作者**: {author or '未知'}\n\n"
83        content += f"**链接**: {link or '无'}\n\n"
84        if tags:
85            content += f"**标签**: {tags}\n\n"
86        content += "---\n\n"
87        content += "## 内容\n\n"
88        content += f"{desc or '无内容'}\n"
89 
90        filepath.write_text(content, encoding="utf-8")
91        count += 1
92 
93    print(f"导出完成: {output_dir}")
94    print(f"共生成 {count} 个文件")
95    return True
96 
97 
98if __name__ == "__main__":
99    db_path = Path(sys.argv[1]) if len(sys.argv) > 1 else None
100    output_dir = Path(sys.argv[2]) if len(sys.argv) > 2 else None
101    export_to_workspace(db_path, output_dir)
102