You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have a private package repo with pre-release versions of internal packages that are in-development.
I use pipx to install packages, including for testing in some cases.
Using --pre seems to apply to all dependencies, as well as the target package. In some cases, this causes things to break (e.g. "AttributeError: cython_sources" with Cython 3.0.0a10 yaml/pyyaml#601) when attempting to install my pre-release packages like this: pipx install <internal package> --pip-args="--index-url=<internal host:port> --trusted-host=<internal host> --pre" --python <path to pypy>
Describe the solution you'd like
It seems like it could help if there was a more specific option, maybe something like --pre-no-deps that would only allow for installing pre-release versions of the top level packages that are specified.
Guess I posted this too quickly.. For future reference, it makes sense to just specify the exact version to install directly and not use the --pre flag at all.
E.g. pipx install <internal package>==1.0.0a3 --pip-args="--index-url=<internal host:port> --trusted-host=<internal host>" --python <path to pypy>
What's the problem this feature will solve?
pipx
to install packages, including for testing in some cases.--pre
seems to apply to all dependencies, as well as the target package. In some cases, this causes things to break (e.g. "AttributeError: cython_sources" with Cython 3.0.0a10 yaml/pyyaml#601) when attempting to install my pre-release packages like this:pipx install <internal package> --pip-args="--index-url=<internal host:port> --trusted-host=<internal host> --pre" --python <path to pypy>
Describe the solution you'd like
It seems like it could help if there was a more specific option, maybe something like
--pre-no-deps
that would only allow for installing pre-release versions of the top level packages that are specified.Alternative Solutions
N/A
Additional context
pypy
.Code of Conduct
The text was updated successfully, but these errors were encountered: