Deprecate vectorized methods for trigonometric and hyperbolic functions in favor of compact broadcast syntax #18571
Conversation
Sacha0
changed the title
kshyatt
added
maths
|
Rebased. Best! |
| @@ -1003,6 +1003,7 @@ macro vectorize_2arg(S,f) | |||
| end | |||
| export @vectorize_1arg, @vectorize_2arg | |||
|
|
|||
| <<<<<<< HEAD | |||
There was a problem hiding this comment.
Looks like a bad rebase conflict resolution here.
…c functions in favor of compact broadcast syntax.
|
Absent objections or requests for time, I plan to begin sequentially rebasing/merging the uncontroversial pull requests in this series on Wednesday morning PDT. (Each merge will introduce conflicts with the remaining pull requests in this series.) Best! |
|
Please don't rebase all of them at once, that causes the better part of a day of CI backlog. Can you pick an order you'd like these merged in and just rebase a few at a time? |
Absolutely, my plan is to rebase and merge one at a time. (I had not considered how long that process will take if we wait for CI to complete after each rebase. Would you be on board with my starting that process now with the older pull requests in this series?) Thanks! |
|
It's quiet now, so no big deal to rebase a few now if they need it. It's doing any more than about 5 rebases in quick succession that starts to slow down other prs ability to get test results back. |
|
Sorry, I meant the sequence of rebases and merges, not solely rebases. (I rebased the PRs in this series over the last few days). Should I move ahead? Thanks! |
|
Yes, please do start merging the small ones of these in whatever order you see fit. |
|
Subsumed by #19791. |
This PR deprecates all (?) remaining vectorized methods for trigonometric and hyperbolic functions (less those for
SparseVectors, separate PR) in favor of compact broadcast syntax. Ref. #16285, #17302, #18495, #18512, #18513, #18558, #18564, and #18566. Best!(Unlike with
float,real, etc., the remaining vectorized methods for trigonometic and hyperbolic functions never alias their input. This PR should be less controversial than #18495, #18512, and #18513 as a result.)