-
Notifications
You must be signed in to change notification settings - Fork 103
Release V5.2.0 #3176
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
Release V5.2.0 #3176
Conversation
Allows throwing more specific error in coordinates
Fix recompilation cascade
Caused linking issues. Also put used function in an anonymous namespace to prevent future linking issues
Also default to off
Usually installed under `usr/include/hypre` (or similar) for system packages
From that header: > PETSc "helpfully" defines macros for MPI functions that clobber the > real names, and short of `#undef`-ing all of them in every file > that includes any PETSc header, we can define the following macro > which should disable them, which I'm sure will work forever. This > means we _must_ `#include` this header _before_ any PETSc header! A recent commit reordered the PETSc headers in this file
Ensure PETSc headers are included after `bout/petsclib.hxx`
Avoid `#define` conflict with sundials
This works for 2D and 3D fields (and is also shorter code)
|
Note that there are some breaking changes in this release, but I don't think they should cause problems for very many people:
|
|
|
|
Thanks @ZedThree We probably should not run clang-format and clang-tidy-review on Will there also be a |
Yes, agreed!
Yeah, I was thinking we should maybe simplify things and just have |
We seem to have guards, but for some reason they are not working: RTFM :-) That will only prevent running the CI for PRs into -rc branches 😓
Unless we manage to hugely simplify the release workflow, I think it is not worth it. |
The big advantage is we would only have one branch to care about! |
|
We're almost warning clean again, just down to the deprecated |
|
We probably want #3166 so that the released package gets pushed to pypi. |
I merged the |
|
@bendudson @dschwoerer Unless there are any objections, I'll merge and release this on Friday |
|
For some reason the Fedora push build hung on I'm going to bump the DOI and make the release now |
This has been way too long!
Making a New Release of BOUT++
This is checklist of things to do (in order) when making a new
release. This applies equally to both major/minor releases and bugfix
releases
vX.Y.Z-rcX/Y) should be offnext. Bugfix releases (Z) should be offmastermaster"bugfixes" can include:
make check-allclang-tidy,clang-check,coverity, etc.Before merging PR:
make -C locale update-allreorder file paths in the .po and .pot files
CHANGELOG.md:bout-changelog-generator.py LAST_RELEASE NEXT_RELEASE][bin/bout-changelog-generator.py]update_citations.py][bin/update_citations.py] to add newauthors to
CITATION.cffCITATION.cffto new DOIREADME.mdto new DOICITATION.cffabidiffto see ifsonameneeds bumping inmakefile:update_version_number.py LAST_RELEASE NEXT_RELEASE][bin/update_version_number.py]After PR is merged:
make distfrom build directory. Ensure you are on a tag and correct version is used for archive and folder within.leading
vcmake -S . -B build && make dist -C buildmasterintonextconfigure.ac:AC_INITCITATION.cff:versionmanual/sphinx/conf.py:versionandreleasemanual/doxygen/Doxyfile_readthedocs:PROJECT_NUMBERmanual/doxygen/Doxyfile:PROJECT_NUMBERCMakeLists.txt:_bout_previous_version,_bout_next_versiontools/pylib/_boutpp_build/backend.py:_bout_previous_version,_bout_next_version