ooodc a7883dbed9 refactor(todo): 重构待办事项管理逻辑及更新状态规则

- 移除 TodoItem 中的 priority、created_at 和 updated_at 字段
- 强制每个任务都必须有唯一 id，且由用户负责生成
- 修改合并模式逻辑，merge=true 下保留未提及的旧任务
- 支持已完成和已取消任务重新激活（状态改回 pending 或 in_progress）
- 禁止 in_progress 状态退回到 pending，必须标记为 completed 或 cancelled
- 优化状态转换校验，允许特定状态间合法切换
- 简化任务变更消息，移除详细的新增/更新/移除统计
- 更新文档和示例，明确 id 必须由用户生成和使用
- 修复和补充测试，增强状态转换和合并模式验证
- 调整任务时间戳生成逻辑，统一使用当前时间及索引
- 该变更提供更合理的任务状态机械及管理模式，提升稳定性和易用性

2026-06-13 09:22:33 +08:00

2.6 KiB

Raw Blame History

+feed-group-list

Shortcut for lark-cli im +feed-group-list. List the caller's feed groups (tags) with auto-pagination that correctly merges both the live and soft-deleted lists.

+feed-group-list is the only CLI surface for listing feed groups — there is no raw feed.groups list command. The list response carries two parallel arrays — groups (live) and deleted_groups (soft-deleted). The shortcut paginates this dual-list response correctly: its --page-all merges both arrays across pages (a naive single-array pager would silently drop one list's later pages). It adds no enrichment.

Identity

User-only. Run with --as user.

Scopes

im:feed_group_v1:read

Usage

# First page
lark-cli im +feed-group-list --as user

# Auto-paginate through all your feed groups (both live and deleted)
lark-cli im +feed-group-list --as user --page-all

# Within an update-time window
lark-cli im +feed-group-list --as user --page-all \
  --start-time 1767196800000 --end-time 1767200000000

Flags

Flag	Required	Description
`--page-size`	No	Records per page, 1–50 (default 50). Caps the combined `groups` + `deleted_groups` count, so a page may hold fewer live groups than the size suggests
`--page-token`	No	Continuation token for a specific page
`--page-all`	No	Auto-paginate and merge all pages (both lists)
`--page-limit`	No	Max pages when `--page-all` is set, 1–1000 (default 20)
`--start-time`	No	Update-time window start (Unix milliseconds as a decimal string)
`--end-time`	No	Update-time window end (Unix milliseconds as a decimal string)

When --page-token is set explicitly, it wins over --page-all (you get exactly that page).

Output

JSON keeps the raw envelope; with --page-all both lists are returned fully merged:

{
  "groups": [
    { "group_id": "ofg_xxx", "type": "normal", "name": "Releases", "rules": { "rules": [] } }
  ],
  "deleted_groups": [
    { "group_id": "ofg_yyy", "type": "rule", "name": "Old", "rules": { "rules": [] } }
  ],
  "page_token": "",
  "has_more": false
}

page_size counts live and deleted groups together, and the per-page count can be smaller still when entries are filtered — so never infer completeness from counts. Pagination is governed solely by has_more.

2.6 KiB Raw Blame History Unescape Escape