Phil Chen

> Return prefecture level division of current division. If current division is a province, return none/null/nil. https://github.com/cn/GB2260/blob/develop/spec.md#prefecture

我的意思是，既然这个方法的未找到是这样定义的，那这里是不是也应该保持一致

@flupke Thanks for your suggestions that makes sense. @ejpcmac Thanks for your reply, and I’ve updated the PR according to the [CONTRIBUTING](https://github.com/ejpcmac/typed_struct/blob/main/CONTRIBUTING.md#workflow).

> The intent here matters a lot, and what do you mean by "the aggregate does not exist" If an aggregate cannot be reconstructed from events, can we assume that...

The changes impact solely the `aggregate_state` function and do not influence the internal logic, such as your own aggregates.

> Because if you use nil, you will need to read source code files in production/environment to know the value of status. The worst situation is if two deployments have...

@yordis Let's separate things apart, I found that problems always seem to be discussed together. **1. Why is it necessary to differentiate the initial state using the `nil` value** Determining...

Here is my situation: My service could handle requests from outside, from users or other systems, which are handled by aggregate_state function. I have to tell them the aggregate’s existence...

> @fahchen For what it's worth, this is why I tend to have my enums default to an `UNKNOWN` value. It helps distinguish the zero value from a fresh but...

> This isn't really necessary. If someone wants the `:aggregate_state` along with other information, they can ask for an `:execution_result` and pull it out of there. @jvantuyl In fact, you...