Bump v0.16 + Remove old deprecations - #321
lkdvos merged 16 commits into
Conversation
Sorry, something went wrong.
Codecov Report✅ All modified and coverable lines are covered by tests.
... and 1 file with indirect coverage changes 🚀 New features to boost your workflow:
|
Sorry, something went wrong.
|
|
||
| ### Changed | ||
|
|
||
| - Tensors now no longer print their data by default, only their spaces. Use `blocks(t)` or `subblocks(t)` to inspect data ([#304](https://github.com/QuantumKitHub/TensorKit.jl/pull/304)) |
There was a problem hiding this comment.
I forgot about the status of that. I do think people are not too happy with no output being shown; would it be so bad to shows blocks(t) by default?
Sorry, something went wrong.
There was a problem hiding this comment.
Or is this having a lot of repercussions again on MPSKit etc?
Sorry, something went wrong.
There was a problem hiding this comment.
We definitely used to show the output of subblocks(t), which we could add back here (even though I strongly prefer blocks(t), I think that is what people wanted?).
I think MPSKit wouldn't be affected since that would have to use the compact = true setting anyways, which shouldn't print this.
If we were to do this, I would however want to limit the output of that by default.
While I agree that for small tensors this can be useful, this is a nuisance for larger tensors...
Do you think this constitutes breaking changes though?
Sorry, something went wrong.
|
After the build completes, the updated documentation will be available here |
Sorry, something went wrong.
|
I tried to further stabilize the tests, but it does seem like playing whack-a-mole with the random number generator again... |
Sorry, something went wrong.
|
Looks good to me, but there is a failing AD test on LTS, so that would preferably be fixed. I am to tired at this point to investigate myself but can do so tomorrow. |
Sorry, something went wrong.
|
I think I kind of ran out of time for this week to work on this, so if you have some time tomorrow, feel free to do so! |
Sorry, something went wrong.
This PR removes the deprecate.jl file, which mostly contained factorization code.
The goal is to free up these names again, which could then be reused in further updates without breaking changes, if we wish to do so.
There were some older deprecated functions in that file as well, that had been there forever (Some even 2+ years), which I think is fine to also delete.
Finally, I removed the PackageExtensionCompat dependency, since that was used for supporting package extensions in Julia <v1.9, which we no longer support anyways.
(I also bumped the version and took the liberty of adding myself to the author list, if you don't mind)
This would probably best be merged after #291 and #316, and can then be used to register v0.16, for which I'll try to write some release notes/changelog and add this to the PR.