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
Is your feature request related to a problem? Please describe.
mlconjug3's dependencies use an exact version pin in requirements.txt. This makes it incompatible with other packages that use more recent versions of dependencies. For example, mlconjug3 is incompatible with instructor on the dependency rich due to mlconjug3 requiring rich==13.2.0 and instructor using ^13.7.0. (c.f. https://github.com/jxnl/instructor/blob/e8280bdfe9f20de9844324351659127528031cf0/pyproject.toml#L17).
Describe the solution you'd like
Allow the dependencies to have the pinned versions or newer.
Describe alternatives you've considered
My current solution is that I forked mlconjug3 and dropped the version pinning entirely.
The text was updated successfully, but these errors were encountered:
ReinderVosDeWael
changed the title
Unpin exact versions of dependencies
Feature Request: Unpin exact versions of dependencies
Apr 22, 2024
Is your feature request related to a problem? Please describe.
mlconjug3's dependencies use an exact version pin in
requirements.txt
. This makes it incompatible with other packages that use more recent versions of dependencies. For example,mlconjug3
is incompatible withinstructor
on the dependencyrich
due tomlconjug3
requiringrich==13.2.0
andinstructor
using^13.7.0.
(c.f. https://github.com/jxnl/instructor/blob/e8280bdfe9f20de9844324351659127528031cf0/pyproject.toml#L17).Describe the solution you'd like
Allow the dependencies to have the pinned versions or newer.
Describe alternatives you've considered
My current solution is that I forked mlconjug3 and dropped the version pinning entirely.
The text was updated successfully, but these errors were encountered: