Richard Chien
Richard Chien
> 相较于nonebot很多强大的功能，我似乎更希望nonebot仅仅作为一个基础框架，把用户发来的消息直接转交给我自己的插件程序来处理。 @needhourger 其实如果真的是这个需求的话，可以使用 [python-aiocqhttp](https://github.com/richardchien/python-aiocqhttp)，这个 SDK 的定位就是仅仅提供一个最基础的 CQHTTP 插件的 Python binding
> 一个踩过一年坑的人建议还是把精力放在有灵魂的功能上，不要做这么繁琐又没有意义的事情（嘤嘤嘤 23333，适配确实是过于繁琐了，反正先讨论着，到时候再决定怎么写 > 游戏 ❌ 感觉游戏算是个蛮实际的需求，不过我有点不太清楚现在那些游戏都是什么个逻辑，自己没写过
@huaji0353 你这个功能必然是要保存每个潜在用户的最近几条消息的
> 据我的观察权限控制仅仅靠那几个 `int` 遠遠不夠，而且這些常量現在的實現只能取聯合不能取交集，導致更鷄肋。我建議弄成類似策略模式： > > ```python > my_perm_setting = Permission( > lambda perm, event: (perm.ISSUPERUSER or perm.ISGROUP_MEMBER) and event.sender_id != 123456789 > ) > > @on_command("weather", permission=my_perm_setting) > async...
Any more information?
During a previous refactor to source parser, I found a possible cause of cdc offset rewind, which according to @StrikeW is possibly a reason of this issue. The bug case...
> So we're actually not relying on the `Future` to be `Send`? It looks so, though I don't clearly know why and why we need `Send` before.
> @stdrc could you check if it can be closed when you're free? thanks. It seems that `pk_indices` may be able to be removed from `ExecutorInfo`, but some executor still...
After extracting `ExecutorInfo` out of `XxxExecutor` (#15167), the idea proposed in this PR is still valid. It's still true that most executors don't access their input nodes' `pk_indices` (and also...
> kindly cc @stdrc in case you are interested and don't know this feature. 😄 I knew it but I didn't know we have this issue🤣