✨ 更新数据库模型 #39

GuGuMur · 2023-12-17T09:15:36Z

No description provided.

AzideCupric · 2023-12-31T05:56:08Z

.github/workflows/testing.yml

+      - uses: actions/checkout@v4
+      - name: Set up PDM
+        uses: pdm-project/setup-pdm@v3


为什么要改这里，不就是上面的./.github/actions/setup-python离得内容吗

~~不知道为什么action全报错了，完整抄下一份发现能用~~

感觉是你修改了哪个地方的问题(

这个pr之前commit的action test报错记录出错位置

~~不是很懂导入怎么出错了，反正是把他展开写进去以后能用了~~

AzideCupric · 2023-12-31T05:56:38Z

.github/workflows/testing.yml

+      - name: Install dependencies
+        run: |
+          pdm sync -d -G test
+      - name: Run Tests
+        run: |
+          pdm run -v pytest tests


有什么区别吗，为什么要sync一次？以前不是没问题

抄的 pdm 文档里的（

然后之前的 pr 没问题但是这个 pr 里每次 commit 以后的 preview action 就全炸了

AzideCupric · 2023-12-31T05:57:57Z

.pre-commit-config.yaml

-  - repo: https://github.com/pre-commit/mirrors-prettier
-    rev: v4.0.0-alpha.3
-    hooks:
-      - id: prettier
-        types_or: [javascript, jsx, ts, tsx, markdown, yaml, json]
-        exclude: "admin-frontend/"
-        stages: [commit]
-


这里保留是考虑到还可以留着格式化yaml和md，就没删，删掉exclude那段就行

最近的一次错误报告：对flag怎么写没什么头猪，直接删了（逃

怪事也没写这两个参数啊

AzideCupric · 2023-12-31T05:59:09Z

README.md

-> 注意：非[超级用户](https://nonebot.dev/docs/appendices/config#superusers)只可删除自己绑定的账号，超级用户可以删除bot数据库内所有账号
+> 超级用户可以删除bot数据库内的任意账号
+>
+> [OneBot V11](https://onebot.adapters.nonebot.dev/) 适配器中的 QQ 群管理员以上权限的用户可以在群内删除本群内的任意账号


我认为在一个好的权限插件出来之前还是保留原状？这样也太平台特化了

AzideCupric · 2023-12-31T06:01:59Z

nonebot_plugin_skland_arksign/model.py

    cred: Mapped[str] = mapped_column(String, doc="森空岛账号CRED")
    token: Mapped[str] = mapped_column(String, doc="森空岛账号TOKEN", nullable=True)
-    note: Mapped[str] = mapped_column(String, doc="备注", nullable=True, unique=True)
+    note: Mapped[str] = mapped_column(String, doc="备注", nullable=True)


备注得保证唯一性吧，不然有个人备注abc，另一个人也备注abc，那 signin abc 签到谁的

signin abc以后bot会返回当前用户能控制的所有备注为abc的账户，通过回复序号来确定签到哪个

AzideCupric · 2023-12-31T06:02:54Z

nonebot_plugin_skland_arksign/utils.py

+    if not token:
+        return "未绑定"
+    else:
+        if is_group:
+            return "已绑定"
+        return token


Suggested change

if not token:

return "未绑定"

else:

if is_group:

return "已绑定"

return token

if not token:

return "未绑定"

elif is_group:

return "已绑定"

return token

AzideCupric · 2023-12-31T06:05:29Z

往一个pr里塞进太多不同方向的更改了，显得很杂，得考虑拆分
但拆分成若干pr的话现在的发版操作会导致版号疯涨，我先改改发版操作

AzideCupric · 2024-01-04T07:56:51Z

这个pr里糅合了多个不同方向的更改，应该拆分开来

数据库字段添加，以及添加导致的命令修改
旧命令的改动
~~CI 变动（已经修过了）~~

GuGuMur · 2024-01-06T13:43:01Z

一开始这个 pr 的灵感来源就是群友想 update 备注发现没权限然后转过来~~要烤我~~，新数据库和新指令实际都是为了这个服务的，CI 更新其实是看靠前的 commit 出错不顺眼的（

~~简单来说就是感觉没必要拆这个 pr~~

AzideCupric · 2024-01-06T13:52:17Z

这明显两个不重叠的更改吧(
发现卡了一些reviews

AzideCupric · 2024-01-04T06:44:37Z

nonebot_plugin_skland_arksign/alc_parser.py

+        help_text="在私聊绑定一个在群聊中添加的签到账号",
+    ),
+    Subcommand(
+        "rebind",


不应该叫 rebind，其实就是添加了一个 send_to 字段，直接叫 sendto 就行

其实指令叫 send_to 感觉更容易混（

AzideCupric · 2024-01-04T06:45:31Z

nonebot_plugin_skland_arksign/alc_parser.py

    Subcommand(
        "list",
        help_text="列出所有签到账号",
    ),
    Subcommand(
        "del",
-        Args["identifier", str],
+        Args["identifier", str]["position?", int],


增加position容易导致误删吧，万一在增删账号之后原来那个账号的位置改变了呢，我觉得不应该加

发送指令以后能处理的列表就固定在 state 里了，position也只对这部分起效

发送指令以后能处理的列表就固定在 state 里了，position也只对这部分起效

何意没太看懂（

AzideCupric · 2024-01-04T06:45:48Z

nonebot_plugin_skland_arksign/alc_parser.py

        help_text="使用uid或者备注删除一个签到账号",
    ),
    Subcommand(
        "update",
-        Args["identifier", str],
+        Args["identifier", str]["position?", int],


这里的position也是

AzideCupric · 2024-01-04T06:46:11Z

nonebot_plugin_skland_arksign/alc_parser.py

    Subcommand(
        "signin",
-        Args["identifier", str],
+        Args["identifier", str]["position?", int],


还有这里的position

AzideCupric · 2024-01-04T06:47:29Z

nonebot_plugin_skland_arksign/command.py

-        await skland.finish("该UID已经被注册，请检查")
-    new_record = SklandSubscribe(user=user_account.dict(), uid=uid, token=token, cred="", note=note)
+    # 1. 检查UID是否被注册
+    async with db_session.begin():


一定需要这个 async with 吗

好像是直接用过一次之后就不能再直接用了，所以套了几层begin

AzideCupric · 2024-01-04T06:55:45Z

nonebot_plugin_skland_arksign/command.py

+            result: list[SklandSubscribe] = (await db_session.scalars(stmt)).all()
+            result = [i for i in result if compare_user_info(i.user, event_session_dict)]
+            if result:
+                await skland.finish("该note已经被您注册，请检查")


Suggested change

result: list[SklandSubscribe] = (await db_session.scalars(stmt)).all()

result = [i for i in result if compare_user_info(i.user, event_session_dict)]

if result:

await skland.finish("该note已经被您注册，请检查")

result: list[SklandSubscribe] = (await db_session.scalars(stmt)).all()

if any(i for i in result if compare_user_info(i.user, event_session_dict)):

await skland.finish("该note已经被您注册，请检查")

不过我觉得 compare_user_info 的实现很怪，能不能直接比较没有dict()之前的值

试过很多种从数据库里带函数筛的方法，全用不了

AzideCupric · 2024-01-04T06:58:52Z

nonebot_plugin_skland_arksign/command.py

+    db_session: AsyncSession = Depends(get_session),
+):
+    is_group = skland_is_group(bot, event)
+    all_subscribes: list[SklandSubscribe] = await skland_list_subscribes(bot, event, matcher, db_session)


skland_list_subscribes为什么不直接依赖注入呢

并且在内部判断权限来决定返回的数据

呜呜呜
前面的commit有试过这种的，但是prompt值就写不到state里了，所以拿了这么一种很丑的写法

？感觉不对劲哪个commit

能想到的一个commit

没记错的话，执行到询问position那一步会报错 keyerror 找不到 prompt

GuGuMur · 2024-01-20T13:23:45Z

这回不拆分是不行咯（

AzideCupric · 2024-01-20T13:25:05Z

乐

AzideCupric · 2024-01-20T13:35:08Z

动工之前先写个 Feature Issue 吧

GuGuMur and others added 13 commits December 17, 2023 17:13

✨ 更新数据库

8de332b

💄 auto fix by pre-commit hooks

581f598

tweak

25ee373

💄 auto fix by pre-commit hooks

77d6451

🐛 fix: compare_user_info

74e65fb

tweak

9810441

fix: 禁用set_path_arg

a860a25

💄 auto fix by pre-commit hooks

9c7186c

run: 先这样算了

e3dee8a

💄 auto fix by pre-commit hooks

4695feb

🐛 fix: can work

699f5d5

💄 auto fix by pre-commit hooks

3341927

💚 ci: test

9b517b2

GuGuMur marked this pull request as ready for review December 31, 2023 01:47

GuGuMur requested a review from AzideCupric December 31, 2023 01:47

AzideCupric reviewed Dec 31, 2023

View reviewed changes

✨ feat: rebind func

bab33de

AzideCupric reviewed Jan 6, 2024

View reviewed changes

GuGuMur and others added 3 commits January 7, 2024 12:29

🎨 tweak

1eb30ff

💄 auto fix by pre-commit hooks

8478b25

⬇️ downgrade version

bd31d4e

AzideCupric force-pushed the database_update branch 2 times, most recently from 5a2088d to bd31d4e Compare January 12, 2024 14:36

💚 还原CI更改

23f44e9

GuGuMur closed this Jan 20, 2024

GuGuMur mentioned this pull request Jan 20, 2024

[Feature] 数据库新字段添加 & 指令系统更改 #53

Open

3 tasks

✨ 更新数据库模型 #39

✨ 更新数据库模型 #39

Conversation

GuGuMur commented Dec 17, 2023

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

AzideCupric commented Dec 31, 2023

AzideCupric commented Jan 4, 2024 • edited by GuGuMur Loading

GuGuMur commented Jan 6, 2024

AzideCupric commented Jan 6, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

GuGuMur commented Jan 20, 2024

AzideCupric commented Jan 20, 2024

AzideCupric commented Jan 20, 2024

AzideCupric commented Jan 4, 2024 •

edited by GuGuMur

Loading

AzideCupric commented Jan 6, 2024 •

edited

Loading