Henry Schreiner
Henry Schreiner
Maybe it would be better (cross-platform) to handle these sorts of signals and just exit with the right error code instead of exposing the Python traceback? (Well, at least that...
And you’d need to duplicate the executable, I think, since the module can’t reliably locate the scripts install location.
No version of METADATA (up to and including 2.3) has License-File: https://packaging.python.org/en/latest/specifications/core-metadata/ I hadn't realized I was adding something that's only in a draft PEP and not part of METADATA...
(If [this](https://peps.python.org/pep-0639/) does get standardized, it will become METADATA 2.4, and the License field will be deprecated - it also provides a way for the user to specify the values...
You certainly can't define `license` as a string. Also to compute/normalize an SPDX license expression you either need a heavy third party dependency, or introduce potentially unsafe eval's, from what...
Currently the canonical license specifiers are in the trove classifiers - you can get the SPDX license expression from that (https://peps.python.org/pep-0639/appendix-mapping-classifiers/#spec-mapping-classifiers-identifiers). I believe both setuptools and hatchling fully adopt PEP...
Correction: setuptools adopts an earlier version of the draft.
PEP provisionally accepted (waiting on backends) so starting work here: https://github.com/pypa/pyproject-metadata/pull/132!
Does this still happen? Latest scikit-build's should support this.
@jcfr, @thewtex, should we add Zenodo?