alamb opened a new pull request #1127:
URL: https://github.com/apache/arrow-rs/pull/1127
Draft. Items left to do:
- [ ] PR for `impl Array` for `ArrayRef`, ete (and maybe remove
`downcast_array!`)
- [ ] Complete writing tests for Float arrays for comparison kernels
# Which issue does this PR close?
Re https://github.com/apache/arrow-rs/issues/1113
# Rationale for this change
The kernels added in #1115 and others don't support floats, as @liukun4515
noted in https://github.com/apache/arrow-rs/pull/1115#discussion_r777161312
Also, they currently require an array by owner (`ArrayRef`) rather than
`&dyn Array` which requires unnecessary cloning at the callsite. I discovered
this while trying to write tests for `Float32/64Array`
# What changes are included in this PR?
Thanks to @matthermturner's work (and tests!) this PR is fairly simple
1. Add support for Float32Array and Float64Array
2. Change the trait bound from `Into<i128>` to
[`num::ToPrimitive`](https://docs.rs/num/latest/num/trait.ToPrimitive.html)
which allows floats and avoids an unecessary intermediate cast
3. Remove redundant type checks
# Are there any user-facing changes?
API s changed
However, since this code / feature hasn't been released yet so this isn't an
API change to released code.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@arrow.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
