Prepare 0.14.2 #182
Prepare 0.14.2 #182
Conversation
aseyboldt
force-pushed
the
prepare-0.14.2
branch
3 times, most recently
from
16eb15b to
cfecd6a
Compare
aseyboldt
force-pushed
the
prepare-0.14.2
branch
from
cfecd6a to
af6b690
Compare
aseyboldt
force-pushed
the
prepare-0.14.2
branch
from
af6b690 to
4bf7a8d
Compare
| ## [0.14.1] - 2025-03-05 | ||
|
|
||
| ### Ci | ||
|
|
||
| - Update run-on-arch to avoid segfault (Adrian Seyboldt) | ||
|
|
||
| - Repare 0.14.1 (Adrian Seyboldt) |
There was a problem hiding this comment.
ups. Unfortunately it's autogenerated based on the commit message, so pretty much too late to change now :-(
| if outer_kwargs is None: | ||
| outer_kwargs = {} | ||
|
|
||
| def make_adapter(*args, **kwargs): |
There was a problem hiding this comment.
Mostly just a stylistic suggestion, but I think using partial here (like previously) is quite a bit better for static code analysis. Is it possible to import make_transform_adapter at the top of with_data, or does it actually need to be imported within make_adapter?
There was a problem hiding this comment.
It needs to be imported late unfortunately.
I guess we could have a wrapper module that is empty if the optional dependency is not installed. Maybe that would be cleaner?
There was a problem hiding this comment.
That does feel significantly cleaner to my tastes. That way you get separation between the import logic and the inner workings.
| if outer_kwargs is None: | ||
| outer_kwargs = {} | ||
|
|
||
| def make_adapter(*args, **kwargs): |
There was a problem hiding this comment.
Other comment also applies here
No description provided.