Conversation
|
I would consider making direct usage of transformers a private API, to be honest. |
Oh, this also kinda applies to the question of whether or not to do the major bump after #1094, I do personally agree that considering transformers private makes more sense. I originally even thought they were, didn't actually realize they were documented as public. I don't think there's many people using them anyways. I'm fine with removing them from the docs. |
|
I removed transformers from the docs, so we consider them private now. I think we have somewhere a policy, "if it is in the docs, it is public API. Unless it is on the |
|
BTW, do we consider removing transformers from public API a breaking change? This doesn't technically break any usage of them right now, but future changes might, and unless we remember not to make such changes to transformers and essentially consider them public at least until the next major release, which I'm not that confident we will, that could pose problems. Though maybe it's enough to just leave a comment somewhere close to the top of the file about it. Also, if we want, we could also literally deprecate direct usage of transformers, we could do this by adding a kw only, |
When did that directive even got here?
Hmm, I like the idea of deprecating that |
Misspelled it
ItsDrike
left a comment
There was a problem hiding this comment.
will need resolving merge conflicts too after the new deprecation handling
A warning is emitted anyway if user calls the classes directly
ItsDrike
left a comment
There was a problem hiding this comment.
Just one minor wording detail, otherwise it can be merged
I hope my grammar is better this time, but please suggest any nitpicks. I will appreciate it
The docs for transformers are on the "MOTD parsing" page