cli_##############
#####################

openclaw channels add

"feishu": {
// 全局默认设置（可被单账号覆盖）
"enabled": true,
"domain": "feishu", // 或 "lark"

"accounts": {
// 第一个机器人（主账号）
"main": {
"appId": "cli_a1b2c3d4",
"appSecret": "wkEb...123",
"botName": "主助手",
"groupPolicy": "open"
},
// 第二个机器人（例如用于特定测试或部门）
"test_bot": {
"appId": "cli_x9y8z7w6",
"appSecret": "abCd...456",
"botName": "测试号",
"dmPolicy": "allowlist", // 不同的权限策略
"allowFrom": ["ou_..."]
}

然后用openclaw channels list命令：
? OpenClaw 2026.2.14 (c1feda1) — I'm not magic—I'm just extremely persistent with retries and coping strategies.

03:36:47 [plugins] feishu_doc: Registered feishu_doc, feishu_app_scopes
03:36:47 [plugins] feishu_wiki: Registered feishu_wiki tool
03:36:47 [plugins] feishu_drive: Registered feishu_drive tool
Chat channels:
- Telegram default: configured, token=config, enabled
- Feishu test_bot: configured, enabled
- Feishu main: configured, enabled

Auth providers (OAuth + API keys):
- google-gemini-cli:chedong@gmail.com (oauth)
- google:default (api_key)
- minimax:default (api_key)
- qwen-portal:default (oauth)
- moonshot:default (api_key)
- minimax-portal:default (oauth)
- minimax-cn:default (api_key)

下载安装后：会有这样一个界面，按照安装OpenCore ==> 创建USB安装盘 ==> 安装后的驱动更新这几个步骤进行系统升级：

安装主要以下几个步骤：

brew install cvs2svn
Warning: No available formula with the name "cvs2svn".
==> Searching for similarly named formulae and casks...
Error: No formulae or casks found for cvs2svn.

install cvs2svn Defaulting to user installation because normal site-packages is not writeable ERROR: Could not find a version that satisfies the requirement cvs2svn (from versions: none) ERROR: No matching distribution found for cvs2svn

错误: cvs2git 无法运行
请检查 Python 3 是否已安装: python3 --version

ERROR: cvs2git requires Python 2, version 2.4 or later; it does not
work with Python 3. You are currently using version 3.14.2.
Please restart cvs2git using a different version of the Python
interpreter. Visit http://www.python.org or consult your local system
administrator if you need help.

站点运行于 Movable Type 3.3，历史跨度较长（十年以上）。 在升级服务器环境（Perl 5.3 + 新版 MySQL / MariaDB）后，发现：

• 文章标题、分类、评论、TrackBack 出现乱码（如：安装笔记）
• 数据库表字符集显示为 utf8 / utf8mb3，但内容明显异常
• 后台管理界面：界面菜单正常，但博客各种内容部分都受中文乱码影响

二、问题本质分析（核心结论）

这是一个经典的历史字符集问题：

• 早期 MT 3.x + MySQL 常以 latin1 方式写入数据
• 实际内容是 UTF-8（甚至混有 GBK / GB2312）
• 后期 MySQL / Perl 按 UTF-8 读取，导致“UTF-8 被当 latin1 再显示”

修复原则：

• 不修改程序逻辑：尝试修改mt-config.cgi中配置增加

  DBIParams mysql_enable_utf8=0

  是没有用的，改用ObjectDriver DBI::MariaDB 但Movable Type 3.3根本不支持MariaDB驱动。

  Got an error: Unsupported driver MT::ObjectDriver::DBI::MariaDB: Can't locate MT/ObjectDriver/DBI/MariaDB.pm in @INC (you may need to install the MT::ObjectDriver::DBI::MariaDB module) (@INC contains: /home/path/to/cgi-bin/mt/extlib lib /etc/perl /usr/local/lib/x86_64-linux-gnu/perl/5.34.0 /usr/local/share/perl/5.34.0 /usr/lib/x86_64-linux-gnu/perl5/5.34 /usr/share/perl5 /usr/lib/x86_64-linux-gnu/perl-base /usr/lib/x86_64-linux-gnu/perl/5.34 /usr/share/perl/5.34 /usr/local/lib/site_perl) at (eval 8) line 1. BEGIN failed--compilation aborted at (eval 8) line 1.
  



• 不整体强制转换数据库字符集


• 仅针对“人类可读字段”做字节级修复


• 修复前使用mysqldump或者phpmyadmin的工具备份数据库：mysqldump database_of_mt > mt_entry.bak.sql

四、通用修复方案（核心 SQL 模式）

统一采用以下安全转换模式：

CONVERT(
  CAST(CONVERT(column_name USING latin1) AS BINARY)
  USING utf8mb4
)

说明：

• latin1：还原历史错误解码
• BINARY：保留原始字节
• utf8mb4：以现代 UTF-8 重新解释

五、各主要数据表修复脚本

1. mt_entry（文章）

UPDATE mt_entry SET
  entry_title = CONVERT(CAST(CONVERT(entry_title USING latin1) AS BINARY) USING utf8mb4),
  entry_text = CONVERT(CAST(CONVERT(entry_text USING latin1) AS BINARY) USING utf8mb4),
  entry_text_more = CONVERT(CAST(CONVERT(entry_text_more USING latin1) AS BINARY) USING utf8mb4),
  entry_excerpt = CONVERT(CAST(CONVERT(entry_excerpt USING latin1) AS BINARY) USING utf8mb4),
  entry_keywords = CONVERT(CAST(CONVERT(entry_keywords USING latin1) AS BINARY) USING utf8mb4);

2. mt_comment（评论）

UPDATE mt_comment SET
  comment_author = CONVERT(CAST(CONVERT(comment_author USING latin1) AS BINARY) USING utf8mb4),
  comment_text = CONVERT(CAST(CONVERT(comment_text USING latin1) AS BINARY) USING utf8mb4);

3. mt_category （分类目录） / mt_tag （标签）

UPDATE mt_category SET
  category_label = CONVERT(CAST(CONVERT(category_label USING latin1) AS BINARY) USING utf8mb4),
  category_description = CONVERT(CAST(CONVERT(category_description USING latin1) AS BINARY) USING utf8mb4);

UPDATE mt_tag SET

  tag_name = CONVERT(CAST(CONVERT(tag_name USING latin1) AS BINARY) USING utf8mb4);

4. mt_template（模板）

UPDATE mt_template SET
  template_name = CONVERT(CAST(CONVERT(template_name USING latin1) AS BINARY) USING utf8mb4),
  template_text = CONVERT(CAST(CONVERT(template_text USING latin1) AS BINARY) USING utf8mb4);

5. mt_tbping（TrackBack ping）

UPDATE mt_tbping SET
  tbping_blog_name = CONVERT(CAST(CONVERT(tbping_blog_name USING latin1) AS BINARY) USING utf8mb4),
  tbping_title = CONVERT(CAST(CONVERT(tbping_title USING latin1) AS BINARY) USING utf8mb4),
  tbping_excerpt = CONVERT(CAST(CONVERT(tbping_excerpt USING latin1) AS BINARY) USING utf8mb4),
  tbping_junk_log = CONVERT(CAST(CONVERT(tbping_junk_log USING latin1) AS BINARY) USING utf8mb4);

6. mt_blog（博客信息）

UPDATE mt_blog SET
  blog_name = CONVERT(CAST(CONVERT(blog_name USING latin1) AS BINARY) USING utf8mb4),
  blog_description = CONVERT(CAST(CONVERT(blog_description USING latin1) AS BINARY) USING utf8mb4),
  blog_welcome_msg = CONVERT(CAST(CONVERT(blog_welcome_msg USING latin1) AS BINARY) USING utf8mb4);

7. mt_author（作者）

仅修以下字段，严禁修改密码 / token 字段：

UPDATE mt_author SET
  author_name = CONVERT(CAST(CONVERT(author_name USING latin1) AS BINARY) USING utf8mb4),
  author_nickname = CONVERT(CAST(CONVERT(author_nickname USING latin1) AS BINARY) USING utf8mb4),
  author_hint = CONVERT(CAST(CONVERT(author_hint USING latin1) AS BINARY) USING utf8mb4);

六、验证与验收

1. 乱码扫描

SELECT entry_id FROM mt_entry
WHERE entry_title REGEXP '[åæçéï]';

返回 0 行表示修复完成。

2. 随机抽样

SELECT entry_id, entry_title FROM mt_entry
ORDER BY RAND() LIMIT 10;

相对来说： 语雀的效果就不错：结合目录管理和协同的文档，表格管理等，而且还支持从confluence备份整体导入。

Notion 估计是商业组合软件里面模板做得最好的。