BUG: Ensure that same-kind casting works for uints (mostly) #27802
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This is the most minimal fix I could think off... If we have uints still use the out-of-bounds value, but make it uint. This is because the code uses copyto with the default same-kind casting in places... This doesn't remove all the other quirks, and surely, it is a behavior change in some awkward situations (since you got a uint, someone might mix the newly uint value with an integer and get float64 results, etc.) But... it is by far the most minimal fix I could think of after a longer time. A more thorough fix may be to just always store the exact dtype, but it would propagate differently e.g. when casting. A start for this is currently in numpygh-27596. Closes numpygh-27269, numpygh-27580
|
Thanks Sebastian. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is the most minimal fix I could think off... If we have uints still use the out-of-bounds value, but make it uint. This is because the code uses copyto with the default same-kind casting in places...
This doesn't remove all the other quirks, and surely, it is a behavior change in some awkward situations (since you got a uint, someone might mix the newly uint value with an integer and get float64 results, etc.)
But... it is by far the most minimal fix I could think of after a longer time.
A more thorough fix may be to just always store the exact dtype, but it would propagate differently e.g. when casting. A start for this is currently in gh-27596.
Closes gh-27269, gh-27580
I am not convinced we should backport this or not. It is the most minimal fix I could think of (besides adding unsafe casts everywhere and hoping for the best/that we found them all). But it clearly has the potential to change behavior in some contexts...