-
Couldn't load subscription status.
- Fork 45
Redundant dimension wastes space and allows disagreement #178
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
Codecov ReportPatch coverage is
📢 Thoughts on this report? Let us know!. |
Interesting, tests passed in #175 end of July, apart from a bug in ArrayLayouts that I fixed upstream. Seems I had also fixed the BandedMatrices tests in that PR. Edit: Seems tests in that PR still pass on all Julia versions apart from the nightlies. |
03c210a to
801eb29
Compare
|
Good to merge? |
|
Looks good but I would like the test errors and inconsistencies on the master branch to be fixed first since tests on Julia 1.9 (and the nightlies) in this PR are aborted due to these problems. I opened #180. |
|
Looks good, and tests pass as well with the recent changes on the master branch 🙂 Speaking about tests, I guess the only thing we should check before merging the PR: Are there existing tests of the |
|
Re pre-existing > rg "\.d"
pdmtypes.jl
52: @test convert(PDiagMat{Float64}, PDiagMat(m)).diag == PDiagMat(convert(Array{Float64}, m)).diag
53: @test convert(AbstractArray{Float64}, PDiagMat(m)).diag == PDiagMat(convert(Array{Float64}, m)).diag
84: @test d.diag === vWhile Re deprecated constructors: tests still pass if we delete them, so they aren't used anywhere that is tested. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you @chriselrod! I added a missing deprecation and tests (of the deprecations, properties, and dimension checks).
Tests also fail on Julia 1.9 on the master branch.
I fixed one of the failures, but the BandedMatrices failure is obviously unrelated so I figured someone else could handle it in a separate PR.
This PR should be non-breaking. I left the old constructors, and added
getpropertyandpropertynamesmethods so that the old behaviors are preserved.Thus, I think bumping the patch version to
0.11.18should be fine, but I could make it0.12.0instead.