This is an archive of the discontinued Mercurial Phabricator instance.

Differential D8963

test-run-tests.t: remove a sub-test since we are done with it
ClosedPublic

Authored by khanchi97 on Aug 27 2020, 3:03 AM.

Details

Reviewers
pulkit
Group Reviewers
hg-reviewers
Commits
rHG601ec3faeb6b: tests: remove a sub-test since we are done with it in test-run-tests.t
Summary

removing test-failure.t to make sure that in next patch when
we re-run failed tests using --retest, it won't get considered
unnecessarily.

Diff Detail

Repository
rHG Mercurial
Branch
default
Lint
No Linters Available
Unit
No Unit Test Coverage

Event Timeline

khanchi97 created this revision.Aug 27 2020, 3:03 AM
pulkit added a subscriber: pulkit.Sep 2 2020, 12:30 PM
pulkit added inline comments.
tests/test-run-tests.t
847

Hm, how does it cause issues with --retest. IIUC that depends on .err file presence right?

khanchi97 added inline comments.Sep 2 2020, 1:11 PM
tests/test-run-tests.t
847

Yes, the file I deleted has .err path present as well. And to make sure that it won't reconsidered at least one of them (.t, .err) should be absent IIUC. So I deleted .t file because that's the pattern I see here in this test file, that we remove a test if we don't need it anymore. Maybe we should also cleanup .err files left behind for the tests we have run above (which I can see, are still present)?

pulkit accepted this revision.Sep 3 2020, 2:13 AM
This revision is now accepted and ready to land.Sep 3 2020, 2:13 AM
pulkit added a comment.Sep 3 2020, 2:57 AM

Edited commit message of this and next one in flight as test-check-commit.t was failing.

Revision Contents
Changeset List

PathPackages
Mtests/test-run-tests.t (2 lines)

Diff 22465

tests/test-run-tests.t

pad pad pad pad............................................................ pad pad pad pad............................................................
$ echo 'saved backup bundle to $TESTTMP/foo.hg' $ echo 'saved backup bundle to $TESTTMP/foo.hg'
saved backup bundle to $TESTTMP/foo.hg saved backup bundle to $TESTTMP/foo.hg
$ echo 'saved backup bundle to $TESTTMP/foo.hg' $ echo 'saved backup bundle to $TESTTMP/foo.hg'
saved backup bundle to $TESTTMP/foo.hg saved backup bundle to $TESTTMP/foo.hg
$ echo 'saved backup bundle to $TESTTMP/foo.hg' $ echo 'saved backup bundle to $TESTTMP/foo.hg'
saved backup bundle to $TESTTMP/*.hg (glob)< saved backup bundle to $TESTTMP/*.hg (glob)<
$ rm test-failure.t
pulkitUnsubmitted
Not Done

Hm, how does it cause issues with --retest. IIUC that depends on .err file presence right?

pulkit: Hm, how does it cause issues with `--retest`. IIUC that depends on `.err` file presence right?
khanchi97AuthorUnsubmitted
Done

Yes, the file I deleted has .err path present as well. And to make sure that it won't reconsidered at least one of them (.t, .err) should be absent IIUC. So I deleted .t file because that's the pattern I see here in this test file, that we remove a test if we don't need it anymore. Maybe we should also cleanup .err files left behind for the tests we have run above (which I can see, are still present)?

khanchi97: Yes, the file I deleted has `.err` path present as well. And to make sure that it won't…
Race condition - test file was modified when test is running Race condition - test file was modified when test is running
$ TESTRACEDIR=`pwd` $ TESTRACEDIR=`pwd`
$ export TESTRACEDIR $ export TESTRACEDIR
$ cat > test-race.t <<EOF $ cat > test-race.t <<EOF
> $ echo 1 > $ echo 1
> $ echo "# a new line" >> $TESTRACEDIR/test-race.t > $ echo "# a new line" >> $TESTRACEDIR/test-race.t
> EOF > EOF