Closed Jazzmanpw closed 3 years ago
And maybe I need to create a new fork just for the last commits, don't I?
And maybe I need to create a new fork just for the last commits, don't I?
Maybe pull the latest master or next branch.
I believe I screwed up this pull request a little bit. I found how to pull the changes from the original repo, but they're now all in this pull request. I've added a couple of commits before the pull and one that you've requested. Is it convenient to merge these changes? The Files changes
tab looks a little bit messy
The original question about any
is still actual.
I've added a couple of commits before the pull and one that you've requested
You should pull the upstream with rebase.
is it OK to have the any type there, or should I try and find some way for a better structure for type inferring?
It's important to have correct type for created reducer. I know what is the problem. The data type for HandlerMap is wrong.
You should pull the upstream with rebase.
I fixed the branching problem. Sorry, but I should force push to fix it.
Please let me know your opinion about the changes I have made.
I think this PR is ready to merge and release. Hope you think so @Jazzmanpw.
I'm sorry for being away for so long. Thank you for the changes. I still don't feel comfortable with TS. And I'm not sure if I could do it better. As well as I don't see any remaining bugs.
Thank you for your contribution @Jazzmanpw. If you are interested in type-level programming in statically typed languages such as TS which inspired their type system from ML languages especially Haskell, I suggest you learning Haskell. It's the best teacher I ever had.
Thank you for the advice. And for the opportunity to become a part of the open-source community. =)
I believe the work is mostly finished. I only have one typing issue, but I'm not sure if it's a problem. When I run dts tests, one for createReducer falis (it worked previously, but some other types were not correct). In this line it determines the action type as
any
.I don't know if there's any use of an exact action type for a reducer and assume that such behavior shouldn't be dangerous. As well as it affects only the new
handle.others
feature, so it's back compatible.My question is: is it OK to have the
any
type there, or should I try and find some way for a better structure for type inferring?