Small logging update to java executor and fix undefined var in release.sh #7370
Conversation
There was a problem hiding this comment.
Cool! Would you be willing to add a test for this new branch please? I'm thinking you could either modify or emulate our util function sample_jarfile https://github.com/pantsbuild/pants/blob/master/tests/python/pants_test/backend/jvm/tasks/test_unpack_jars.py#L30 to test that this works for .tar.gz. We're highly unlikely to ever work with those files while doing Pants devs, so a test would be quite helpful to ensure we don't break its support.
| @@ -469,7 +469,7 @@ function fetch_and_check_prebuilt_wheels() { | |||
| do | |||
| packages=($(find_pkg "${PACKAGE}" "${PANTS_UNSTABLE_VERSION}" "${check_dir}")) | |||
| if [ ${#packages[@]} -eq 0 ]; then | |||
| missing+=("${NAME}") | |||
| missing+=("${PACKAGE}") | |||
There was a problem hiding this comment.
Is this change breaking anything for you at the moment, or you only noticed it and are being a good citizen? I can roll it into #7197 if you prefer Stu to keep this PR more focused.
There was a problem hiding this comment.
It's not breaking anything at the moment, just needed to be cleaned up.
There was a problem hiding this comment.
Sounds good! I'll add it to my PR that makes a bunch of releases.sh changes already, meaning it can be removed here and the PR title can be shortened.
There was a problem hiding this comment.
So looking more into it, there are 5 distinct places we use ${NAME} in release.sh. I'm not entirely sure where it gets defined, but it seems like they're getting resolved as we run these functions when doing a manual release, and no one has seemed to notice.
I'm going to leave it out of my PR after all. If we still want to address it, I think this change would be best broken out into a dedicated PR.
| if jar_path.endswith('.jar'): | ||
| ZIP.extract(jar_path, unpack_dir, filter_func=unpack_filter) | ||
| else: | ||
| archiver_for_path(jar_path).extract(jar_path, unpack_dir) |
There was a problem hiding this comment.
This will not use the unpack_filter, but probably still should.
| @@ -62,4 +62,7 @@ def unpack_target(self, unpacked_jars, unpack_dir): | |||
| if not unpacked_jars.payload.intransitive or coordinate in direct_coords: | |||
| self.context.log.info('Unpacking jar {coordinate} from {jar_path} to {unpack_dir}.'.format( | |||
| coordinate=coordinate, jar_path=jar_path, unpack_dir=unpack_dir)) | |||
| ZIP.extract(jar_path, unpack_dir, filter_func=unpack_filter) | |||
| if jar_path.endswith('.jar'): | |||
There was a problem hiding this comment.
So, assuming that the unpack_filter can be applied to archiver_for_path (I think it needs to be...) you shouldn't need this conditional, and can just always use archiver_for_path.
cheister
force-pushed
the
unpack-non-jars-and-small-cleanup
branch
from
6384e0c
to
748b3f2
Compare
cheister
changed the title
|
Removed the changes to unpack_jars.py |
|
I think that this broke CI due to the changes to It looks like this PR was merged on red CI, which is why we didn't catch this. @cheister could you please put up a PR to revert the changes to |
…n release.sh (pantsbuild#7370)" This reverts commit db391a8.
This reverts commit 6437a91.
Fixing undefined vars in release and better log messaging