Cristian Le
Cristian Le
Was doing a review for Fedora packaging for this and noticed I did not actually properly connect the `.git_archival.txt`
The goal of this new mode is to keep all the artifacts in either the source-dir or the CMake build-dir. From there we reuse the `_editable_redirect.py` as having only `known_source_files`...
Mostly the cross-compilation works out-of-the-box with `crossenv` + `CMAKE_TOOLCHAIN_FILE`. There is only one adjustment needed for `Python_ROOT_DIR` to point to `sys.base_exec_prefix`, the rest is documentation of the change and adding...
This complements the discussion in #1050 that we should not allow some fields to be hard-coded. We should allow them in the context of overrides, so here is a rather...
@eirrgang sorry it took so long to get to this, but this should give you a more direct way to hook your cache-file as needed
In order both for the user and for us to keep track of experimental features, we should keep some issue trackers and have a dedicated doc pages for them. But...
We have a few fields that we still keep but are marked as deprecated like `cmake.targets`. We should remove those right before the release
### Description This is a follow-up to #2266. Although that provides a way to configure the CI for merge queue, this requires quite a lot of duplication of each test...
Follow-up to https://github.com/packit/packit-service/pull/2845. With this the parity with zuul should be all in place. TODO: - [x] Write new tests or update the old ones to cover new functionality. Merge...
### Description The comment in question is https://src.fedoraproject.org/rpms/python-scikit-build-core/pull-request/73#comment-287323 which triggered [this installability test](https://dashboard.packit.dev/jobs/testing-farm/1182711) ### Importance _No response_ ### Workaround - [ ] There is an existing workaround that can be...