-
Notifications
You must be signed in to change notification settings - Fork 1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Apptainer environment override, Python drop/bump, pyhit-wasp recipe marriage, versioner QOL #28847
base: next
Are you sure you want to change the base?
Conversation
Job Precheck, step Conda build config check on 1ecb994 wanted to post the following: A change of the following file(s) triggered this check: conda/peacock The following file(s) are unchanged: conda/mpi/conda_build_config.yaml The conda build config was not changed but one or more of its dependencies have changed |
1ecb994
to
2f892f4
Compare
Job Precheck, step Conda build config check on 2f892f4 wanted to post the following: A change of the following file(s) triggered this check: conda/peacock The following file(s) are unchanged: conda/mpi/conda_build_config.yaml The conda build config was not changed but one or more of its dependencies have changed |
2f892f4
to
6a3f65b
Compare
Job Precheck, step Conda build config check on 6a3f65b wanted to post the following: A change of the following file(s) triggered this check: conda/peacock The following file(s) are unchanged: conda/mpi/conda_build_config.yaml The conda build config was not changed but one or more of its dependencies have changed |
6a3f65b
to
41d7ffe
Compare
Job Precheck, step Conda build config check on 41d7ffe wanted to post the following: A change of the following file(s) triggered this check: conda/peacock The following file(s) are unchanged: conda/mpi/conda_build_config.yaml The conda build config was not changed but one or more of its dependencies have changed |
41d7ffe
to
770787a
Compare
Job Precheck, step Conda build config check on 770787a wanted to post the following: A change of the following file(s) triggered this check: conda/peacock The following file(s) are unchanged: conda/mpi/conda_build_config.yaml The conda build config was not changed but one or more of its dependencies have changed |
770787a
to
68e3cfa
Compare
Job Precheck, step Conda build config check on 68e3cfa wanted to post the following: A change of the following file(s) triggered this check: conda/peacock The following file(s) are unchanged: conda/mpi/conda_build_config.yaml The conda build config was not changed but one or more of its dependencies have changed |
68e3cfa
to
19ef3c1
Compare
Job Precheck, step Conda build config check on 19ef3c1 wanted to post the following: A change of the following file(s) triggered this check: conda/peacock The following file(s) are unchanged: conda/mpi/conda_build_config.yaml The conda build config was not changed but one or more of its dependencies have changed |
19ef3c1
to
21c813f
Compare
Job Precheck, step Conda build config check on 21c813f wanted to post the following: A change of the following file(s) triggered this check: conda/peacock The following file(s) are unchanged: conda/mpi/conda_build_config.yaml The conda build config was not changed but one or more of its dependencies have changed |
Job Documentation, step Docs: sync website on 0e89a6f wanted to post the following: View the site here This comment will be updated on new commits. |
Job Coverage, step Generate coverage on 6e0a8a2 wanted to post the following: Framework coverageCoverage did not change Modules coverageCoverage did not change Full coverage reportsReports
This comment will be updated on new commits. |
21c813f
to
b121b5f
Compare
'peacock', | ||
'pprof', | ||
'pyhit', | ||
'seacas'] |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@loganharbour I am not familiar with these things not having a 'conda version'. Especially the new tracking additions. But without these, I do get errors. I want to make sure I am doing this right.
error:
Traceback (most recent call last):
File "/Users/milljm/projects/moose/python/MooseDocs/test/extensions/test_versioner.py", line 80, in testCodeRender
conda_version = self.version_meta[package]['conda']['version_and_build']
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^
KeyError: 'version_and_build'
I suppose I need to see how we're building this dictionary...
b121b5f
to
a67c80a
Compare
1ad08df
to
eb82471
Compare
b4adf12
to
08fe173
Compare
8e39fa1
to
f46569f
Compare
55d54af
to
d9e1699
Compare
Adjust versioner.yaml as suggested in next update. Bump everything. Again. pin xmltodict below 0.14 (issues discovered). Closes idaholab#28813
setuptools <60 is not available for Python >=3.12. We pinned setuptools long ago. Lets see if we don't have to anymore.
Move away from using my own private pyhit repo. Needed now that pyhit must be built, to conform to our new Python version stack. Because wasp/pyhit are closely intwined, it makes sense to have pyhit as an output variant of wasp. Modify versioner.yaml influential files to match changes. Closes idaholab#27690
Consolidate the compilers used for building Wasp and MPI. Looks like these can now diverge on conda-forge. Add additional packages to be ignored due to not having a conda version... meh, I am not familiar with the why of this, as the libraries added, are being tracked, and have a conda version. TODO: ask Logan!
Add a retry function for all packages that can support it.
I cannot fathom what I might be doing wrong. upstream/deve builds seacas just fine. Try to make the way I am building seacas more like it was, each step of the way.
Everywhere we were allowing use of jinja2 compilers from conda-forge, supply a pinned default instead. This insures our packages are compatible with each other. When moving away from jinja templated compilers, I found we are not including enough. *compilers*_arch package is necessary. For example: ```pre conda install gcc cmake dosomething ``` ...may fail with Cmake not able to discover where/how to use GCC. SEACAS contrib example: ```pre -- The C compiler identification is unknown CMake Error at cmake/tribits/core/package_arch/TribitsGlobalMacros.cmake:1462 (enable_language): No CMAKE_C_COMPILER could be found. ``` `conda install gcc gcc_linux-64`, is the proper way (or rather, more proper than without it). As with all things conda-build, I am still learning. SEACAS build.sh change: Supply lib directory when building SEACAS. There seems to now be an issue with contribs being installed in `lib64` while Conda still expects things in `lib`. This is the bit I am referring to when I say "or rather, more proper than without it" above. Before this change, telling Cmake were libs go was not necessary...
Seems tests fail now without it.
Bump versions. Fix VTK_PREFIX being declared outside environment. We are getting lucky with the other builds. TODO: clean those up. Remove exports where possible. Clean up the retry_build function.
Allow the printing of a summary of what to append to versioner_hashes.yaml ```bash ./versioner.py -s||--summary ```
Forgot to add the 99-user-overrides.sh file... chmod +rx on *.sh remove deprecated/TODO entry in versioner.py
Drop Python 3.8, add 3.12. Remove what seems no longer used: conda_version
885ecf9
to
262eb0f
Compare
Restrict Python veresion in peacock. Bring unittests inline with Python 3.12 requirements. Fix non-raw strings.
6e0a8a2
to
0e89a6f
Compare
This PR adds the following:
${HOME}/.bashrc_apptainer
user override to be applied as they enter our containers if the file exists.versioner_hashes.yaml
when bumping an influential file:$ ./versioner.py --summary 8e39fa181b8abba62f117006603d3a1dddea642d: #PR tools: 0c9ab93 mpi: 80648d3 petsc: 9664f0c libmesh: cf37955 wasp: 98d4d49 moose-dev: 5410b50
Closes #28813 #27690