feat!: Align path_view::render_* APIs with P1036R6
#140
Conversation
|
@ned14 can you investigate the CI failures?
|
| @@ -1568,21 +1568,19 @@ class LLFIO_DECL path_view_component | |||
| LLFIO_TREQUIRES(LLFIO_TPRED(is_source_acceptable<T>), | |||
There was a problem hiding this comment.
render(), immediately above this, was removed in P1036R6. Consistency dictates that either it should go away or also remove the first argument.
There was a problem hiding this comment.
Indeed, I missed the history entries
- R5:
render()function added as per LEWG request - R6:
render()function removed as per LEWG request
🙃
However, given that render() has existed for almost two years, I would first deprecate it for one "release" and then remove it. That should ease the transition for our user base considerably at a very low cost.
There was a problem hiding this comment.
I think the number of users of those functions is likely very low, as they were only added by LEWG and as you correctly point out, were clearly added wrong by me likely very late at night when my eyes weren't working.
P1030R7 is nearing completion after a month of working on it, so I'm going to go ahead and merge this. Thanks for the changeset.
|
Programs should now be fixed on Windows. Sorry about that. The macos installability tests take a very long time. I actually axed a bunch of stuff to make them go much faster. Even then, they are hideously slow. I assume it's an x64 box emulating a Mac M1 CPU. I do note github actions prints a bunch of warnings about the actions being used being too old, and therefore it will only run on them node X or whatever. Which could mean "emulated slow as mollasses" but I haven't investigated. |
Resolves #139