- Jun 11, 2020
-
-
Georges Racinet authored
It's been introduced by py-heptapod#3
-
- Jun 09, 2020
-
-
Georges Racinet authored
It's similar to the existing web hooks tests: we're really listening to the POSTed data. One reason that is possible is that integration/service has a full URL field, instead of just project name and id on Read the docs.
-
- May 18, 2020
-
-
Georges Racinet authored
This reproduces heptapod#260
-
- May 05, 2020
-
-
Georges Racinet authored
-
- Apr 29, 2020
-
-
Georges Racinet authored
- the link should be to create a MR for the natural target branch - the prompt should not fail if the natural target named branch doesn't exist – no prompt in that case
-
- Apr 06, 2020
-
-
Georges Racinet authored
it's need to seed the topic being there, even though we disable the evolve extension.
-
- Mar 30, 2020
-
-
Georges Racinet authored
It seems that the topic aware client must have hg-evolve 9.3.0 in order to push the topic. For the time being, besides asserting that clients without the `topic` extension don't get topics, we check that the `topic` extension *alone* is enough to pull by revision, which is what the runners do.
-
- Mar 29, 2020
-
-
Georges Racinet authored
That way, we won't miss any fix. Obviously, gitlab-ci.yml has to stay onto the `heptapod:lastest` image.
-
Georges Racinet authored
In Heptapod 0.8, it's just impossible to trigger protected branch errors by Mercurial pushes: we rely on Mercurial internal checks for soundness. That test may be interesting in the future for Heptapod 0.12 or 0.13, so we'll push it in a topic against the default branch.
-
Georges Racinet authored
This had to be adapted for Heptapod 0.8, and is about the kind of corruption that could have occurred in older versions. Marked as Docker only because of the `chpst`, that has no equivalent for source installs. Could be improved later.
-
- Mar 24, 2020
-
-
Georges Racinet authored
This could have been done before: these assertions reproduce heptapod#223.
-
- Mar 14, 2020
-
-
Georges Racinet authored
By default, projects are in the hashed path, which is derived from the SHA256 of the decimal representation of their `id`. In `load_tarball` we introduce a new convention that the tarball should make two directories, `hg` and `git` that will become PATH_TO_PROJECT.hg and PATH_TO_PROJECT.git respectively, and update the `issue-3.tar` accordingly.
-
- Mar 13, 2020
-
-
Georges Racinet authored
at the time of our writing, this fails on our current GitLab 12 code, because we only prune GitLab branches for published topics if there is a corresponding MR. as anything related to GitLab post-receive treatment, this has to take asynchronicity into account
-
- Feb 19, 2020
-
-
Georges Racinet authored
The various problems with teardown of projects, that can't always be fixed remotely, have been a pain for a long time. They are worse in constrained environments, as in CI contexts and are always a pain point while bumping upstream GitLab versions, because we always have to pay the price to make project deletion work first. By doing this, we lose the ability to easily find the current test project in logs etc, but it's comparatively acceptable
-
- Feb 15, 2020
-
-
Georges Racinet authored
This means they don't need file system access anymore, and we greatly extend our coverage in remote server tests.
-
Georges Racinet authored
This is first test for heptapod#196
-
- Feb 16, 2020
-
-
Georges Racinet authored
The intent of 2b6ab4a10a1a was to move them aside, but it's been only half committed, resulting in copying them. This is the second half
-
- Feb 15, 2020
-
-
Georges Racinet authored
-
- Jan 24, 2020
-
-
Georges Racinet authored
The code paths are indeed quite different for this, and it's been demonstrated with the last remnaining bug of heptapod#72 before Heptapod 0.8.1 being for HTTP only.
-
Georges Racinet authored
It was convenient at first, but now that we also need to run the tests without file system access, it's been a liability to conflate basic push and hook environment aspects.
-
Georges Racinet authored
-
- Jan 19, 2020
-
-
Georges Racinet authored
For the time being (see heptapod#165), all tests that tweak server-side HGRCs need filesystem access. Those that load repos directly from tarballs need it as well
-
- Jan 15, 2020
-
-
Georges Racinet authored
In other words, for heptapod#129
-
Georges Racinet authored
It's been there for a while, but existing tests had not been converted. This makes them way more readable.
-
Georges Racinet authored
This was duplicated in a few places that now make use of that cleaner call.
-
- Dec 21, 2019
-
-
Georges Racinet authored
Once all tasks related to heptapod#144 are done, inner push over ssh just can't work.
-
- Dec 23, 2019
-
-
Georges Racinet authored
This helper method will make many of our tests more readable. Just demonstrating for one, as there are too many to convert.
-
Georges Racinet authored
This will make tests entailing permission changes more readable
-
- Dec 02, 2019
-
-
Georges Racinet authored
This reproduces heptapod#134, adding the case of the retargetting (i.e., branch of topic changeset has changed) and a sanity check for the fix.
-
Georges Racinet authored
-
Georges Racinet authored
not essential, but still useful
-
- Nov 25, 2019
-
-
Georges Racinet authored
With hg-git!27 being merged, we have much better guarantees that topics get pushed before named branches. Hence we need to avoid explicitely that topics end up being the Gitlab default branches (see also heptapod#126)
-
- Nov 15, 2019
-
-
Georges Racinet authored
Related to direct hooks (heptapod#73) we'll be add User and Project ids to the (WSGI) environment variables.
-
- Nov 18, 2019
-
-
Georges Racinet authored
This is the functional test for heptapod#122
-
Georges Racinet authored
These are the tests for heptapod#119
-
- Nov 10, 2019
-
-
Georges Racinet authored
This led us to introduce a class for user namespaces (not really groups in GitLab terminology). We had to update all URL forming code paths for the fact that now Project.group is an object instead of a namespace name.
-
- Nov 09, 2019
-
-
Georges Racinet authored
-
- Sep 08, 2019
-
-
Georges Racinet authored
Because it is forbidden to remove the default GitLab branch, we have two cases.
-
- Oct 10, 2019
-
-
Georges Racinet authored
test_push was getting too big, and besides, all tests start with pushes. Actually test_push should focus on basic operations that can't be more precisely labeled.
-
- Oct 04, 2019
-
-
Georges Racinet authored
-