BF: require argcomplete version at least 1.12.3 to test/operate correctly #6693
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
On debian stable bullseye and ubuntu 22.04 we have 1.8.1 and test_completion fails.
yarikoptic
added a commit
to yarikoptic/datalad
that referenced
this pull request
May 24, 2022
We keep getting odd fails on travis suggesting that not all tags fetched or smth like that for PRs against maint like datalad#6693 : ====================================================================== FAIL: datalad.tests.test_version.test__version__ ---------------------------------------------------------------------- Traceback (most recent call last): File "/tmp/dl-miniconda-lzitvwji/lib/python3.9/site-packages/nose/case.py", line 198, in runTest self.test(*self.arg) File "/home/travis/build/datalad/datalad/datalad/tests/test_version.py", line 61, in test__version__ ok_startswith(__version__, changelog_version) File "/home/travis/build/datalad/datalad/datalad/tests/utils.py", line 451, in ok_startswith ok_(s.startswith(prefix), AssertionError: String '0.16.2+26.g654fe57' doesn't start with '0.16.3' to troubleshoot this and future similar problems -- show tags and git describe output(s)
Codecov Report
@@ Coverage Diff @@
## maint #6693 +/- ##
==========================================
- Coverage 91.06% 90.68% -0.38%
==========================================
Files 353 353
Lines 44518 44518
==========================================
- Hits 40540 40372 -168
- Misses 3978 4146 +168
Continue to review full report at Codecov.
|
bpoldrack
reviewed
May 25, 2022
There was a problem hiding this comment.
Looks alright, but can you tell anything about those failures, @yarikoptic ? Just noticed them in #6704, too.
In principle needed only for builds against `maint`. Without it we might not have
the most recent git tag for a release in maint. Here is further details emailed to
Travis support team:
I think that the default git clone command used by travis with --depth=50
should also allow for the specification of the base branch .
Rationale:
in datalad project (paid plan) we were getting some periods
of time odd unittest fails suggesting that most recent git tag was
not fetched.
Reason:
1. we have two branches: master and maint. maint is the one getting
bugfixes releases cut off from, and which eventually get merged into
master.
2. travis, regardless of what is the base branch for the PR, just does a
shallow clone of the default branch, e.g. in our case:
git clone --depth=50 https://github.com/datalad/datalad.git datalad/datalad
and that in turn fetches tags reachable from that branch, not
necessarily all most recent tags. Then it fetches/checks out the
corresponding head or a merge head for the PR, e.g.
git fetch origin +refs/pull/6693/merge:
which doesn't fetch any extra tags. That results then in
git describe
not having access to the actual most recent git tag for that PR base branch (in
our case - maint).
The most obvious solution is to disable shallow cloning, but that would add 10
seconds to our already long runs. Ideally I should be able to adjust
aforementioned clone command to become
git clone --depth=50 --branch="{TRAVIS_BRANCH}" https://github.com/datalad/datalad.git datalad/datalad
which seems would resolve our issue.
yarikoptic
force-pushed
the
enh-argcomplete-dep
branch
from
a2168dd
to
4ce660d
Compare
|
yeap -- travis doesn't fetch all the tags!! and looking "above" I see so -- mystery is resolved! Now need to workout a solution for it... emailed travis support, force pushed here a workaround -- doing full clone (costs us ~10 seconds :-/) |
|
ok, mystery resolved and problems averted. Fails are AFAIK known on macs... one is the known test_ria_postclonecfg and another one is also something freshish git-annexy I whined about recently. Let's proceed. |
yarikoptic
changed the title
|
🚀 PR was released in |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
On debian stable bullseye and ubuntu 22.04 we have 1.8.1 and test_completion fails.
I will add corresponding version depend also for debian/control in debian branch.