TST: Adjustments for my pcdswidgets PR#168
Conversation
tangkong
left a comment
There was a problem hiding this comment.
This looks reasonable to me
|
Is it worth keeping the build changes in if I end up removing the extra build step in pcdswidgets? I guess that's not a sure thing yet |
|
I suppose I don't expect this to really change our current CI processes and opens up more options for us, so it's still a net positive? We can decide when pcdswidgets is in a more stable place I suppose |
|
I'm mainly 50/50 because of the ergonomics of |
|
I'm going to merge this in and then if later I decide I don't like placehold_placehold_placehold we'll address that some other way- there must be a way to make conda use like 2 placeholds instead of 50 |
|
Geez apparently the long placeholder path is intentional to do some binary string replacement shenanigans, wild stuff |
|
The correct fix for placeholder spam is to actually install the built package instead of running from the env created for the build- but maybe this will never be needed |
REV: partially revert #168 changes
This makes my
pcdswidgetstests pass and helps the docs build properlySee pcdshub/pcdswidgets#96
Change 1: run conda test from installed version
pcdswidgetsThings to consider:
micromamba/conda-bld/pcdswidgets_1773337587302/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_pla/lib/python3.12/site-packages/pcdswidgets/tests/test_builder.py::test_it_was_built[ui_source0] PASSED [ 1%])Change 2: expose the docs extras argument
pcdswidgetsI need to keeppyqtout of the pypi requirements for smooth local environment generation