On 29. 12. 23 20:41, Mattia Verga wrote:
I think I would just do something like this in %prep:
sed -r -i 's/^file:/# &/' tests/requirements.in
Thanks, that did it. Is this something broken in upstream code, or is it our
parser that needs to be adjusted?
Our thing wants to generate dependencies on packages via their names. It cannot
properly generate dependencies for file:, https:, git, etc. packages. What
should the parser do?
We could generate a dependency on python3dist(rpds-py), but it woulds be wrong,
this package does not BuildRequire self, it just wants to be installed.
It isn't broken in upstream, it's just a tad weird. What upstream does here is
using the tests/requirements.in file for a slightly different purpose than we do.
Their purpose: describe everything we want installed in the test environment.
Our purpose: list the dependencies for tests.
Similar or not, the self-dependency is the difference. And unlike other types
of BuildRequires we can generate, generating them from a file is
non-standardized, so we cannot go to upstream and say "your text file is
wrong", because it's just that -- a text file.
I couldn't find the relevant pip documentation about 'file:.#egg=' format.
file: -> this will be a local path on the filesystem
. -> the actual path
#egg=rpds-py -> this package is called rpds-py (a hint to pip to know this
information before actually going to the path and building a package)
--
Miro Hrončok
--
Phone: +420777974800
IRC: mhroncok
--
_______________________________________________
python-devel mailing list -- python-devel@lists.fedoraproject.org
To unsubscribe send an email to python-devel-le...@lists.fedoraproject.org
Fedora Code of Conduct:
https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives:
https://lists.fedoraproject.org/archives/list/python-devel@lists.fedoraproject.org
Do not reply to spam, report it:
https://pagure.io/fedora-infrastructure/new_issue
