Mercurial > hgsubversion
view tests/fixtures/subprojects.sh @ 1539:0ebcc5bbf692
tests: when making a `memctx`, make sure to use a single repo instance
The way self.repo is dynamic produces bad lock behavior because the
`context.memctx` ends up with a different instance than self in
`localrepo.commitctx`, which means the callbacks in the `memctx` get
an unlocked repo instance. This causes lock warning failures. When
it's not a code freeze for core, we should probably:
* Make lock failures hard, not just warnings
* Stop holding a repo reference in memctx, or otherwise check it's the same
instance as `self` during `localrepo.commitctx`
That's my best guess based on the (very hard to debug) test failures here.
author | Augie Fackler <raf@durin42.com> |
---|---|
date | Sat, 28 Oct 2017 21:34:52 -0400 |
parents | 9a7e3dbd0f6e |
children |
line wrap: on
line source
#!/usr/bin/env bash set -e mkdir temp cd temp svnadmin create testrepo svn checkout file://`pwd`/testrepo client cd client mkdir trunk mkdir -p branches mkdir -p tags svn add trunk branches tags svn commit -m "Initial commit" mkdir trunk/project trunk/other echo "project trunk" > trunk/project/file echo "other trunk" > trunk/other/phile svn add trunk/project trunk/other svn commit -m "Added file and phile in trunk" svn up svn cp trunk tags/tag_from_trunk svn ci -m 'created tag from trunk' svn up svn cp trunk branches/branch svn ci -m 'created branch from trunk' svn up echo "project branch" > branches/branch/project/file svn ci -m "committed to the project branch" svn up echo "trunk2" > trunk/project/file svn ci -m "committed to trunk again" svn up echo "other branch" > branches/branch/other/phile svn ci -m "committed to the other branch" svn up svn cp branches/branch tags/tag_from_branch svn ci -m "create tag from branch" cd .. svnadmin dump testrepo > ../subprojects.svndump echo "Created subprojects.svndump" echo "You might want to clean up ${PWD} now"