- 17 Dec, 2020 1 commit
-
-
Georges Racinet authored
This will be, I hope, more readable than the pervasive "log with `:` separated template / split" pattern we've been using. Especially, I expect failures to be more readily understandable. Using namedtuples for the extracts so that it's admissible in dicts and sets. The drawback is that we have to return the class as well, for litteral comparison. Each call gives a new class but instances of two such namedtuple classes are comparable.
-
- 14 Dec, 2020 2 commits
-
-
Georges Racinet authored
Reproduce a blocking problem with Heptapod 0.18.0rc1
-
Georges Racinet authored
Using in in `test_push.py` for now, but this is a long waiting dedup.
-
- 02 Dec, 2020 4 commits
-
-
Raphaël Gomès authored
-
Raphaël Gomès authored
-
Raphaël Gomès authored
-
Georges Racinet authored
Preferred to call them ProjectMirror rather than the context-lacking RemoteMirror. We should do something a like for the more feature-full Merge Requests. Amazingly, there was not direct GET for a given mirror, so we're resorting to listing all of them. We could later on add more interesting attributes, but this isn't what we are doing for other objects (yet). Here's a poor man advantage: by calling `api_get()` explicitely, there's no ambiguity between local (maybe stale) state and the server state.
-
- 29 Nov, 2020 1 commit
-
-
Georges Racinet authored
In the current state of the main heptapod development branches, (to be released as 0.17.2), all functional tests pass, so let's now ensure it stays that way.
-
- 27 Nov, 2020 2 commits
-
-
Georges Racinet authored
On native projects, we won't have these old parameters to control the pruning of the closed branches, since the Mercurial repo will be the terminal one.
-
Georges Racinet authored
This introduces `@needs.hg_git` and `@needs.hg_native` markers to skip tests if `--heptapod-hg-native` is passed or isn't, respectively.
-
- 26 Nov, 2020 2 commits
-
-
Georges Racinet authored
With the previously forgotten enabling by the tests of the Mercurial native repositories, these have passed for the first time while testing the Heptapod 0.17.1 release. Let's now require them to pass, hoping nothing too flaky will be there. At this point of development of the native mode, the inside tests still fail, and that's not so surprising, probably a real problem to fix, not only test harness.
-
Georges Racinet authored
Now that it inherits from OmnibusHeptapod, it had the hardcoded expected `git@localhost`, which isn't appropriate because DockerHeptapod is meant to be used from outside the container (and OmnibusHeptapod can be used from inside the container)
-
- 24 Nov, 2020 5 commits
-
-
Georges Racinet authored
Now it's really basic again.
-
Georges Racinet authored
it will be reused
-
Georges Racinet authored
It's a bit cleaner: we can have the tear down, without the extra indention that the `with` statement would need.
-
Georges Racinet authored
This reproduces the problems encountered with the Mercurial native mode, see hgitaly#23.
-
Georges Racinet authored
This is just convenient and would also mean a fixture would be very easy to create.
-
- 25 Nov, 2020 1 commit
-
-
Georges Racinet authored
which I've been doing manually on my development workspaces. Of course the hg-native mode hadn't a chance to pass on CI before this.
-
- 24 Nov, 2020 2 commits
-
-
Georges Racinet authored
The stable branch is still for Heptapod 0.16, but there probably won't be any release on it, now.
-
Georges Racinet authored
-
- 19 Nov, 2020 5 commits
-
-
Georges Racinet authored
-
Georges Racinet authored
With the native mode, something pushed is pushed. In that case if we don't want a closing changeset of the GitLab branch, we have to prevent pushing it in the first place. People can always change the default branch setting if needed
-
Georges Racinet authored
For native repos, the shadowing happens at read time, so we can't make a difference between sliding (not a new bookmark) or not. Also, we cannot just let the bookmark at a previous position for the non-shadowed case of the default GitLab branch. This latter case is done identically for non-native repos because it was arguably a logical discrepancy.
-
Georges Racinet authored
The branchmap rules make it so that if it's useful for visualisation of instabilities, in stack topics case, the older one is still visible after publication (because it's base of the one that now has to be rebased). Hoping this can change at some point in the future, though.
-
Georges Racinet authored
For the very few tests that need to make a difference, it's best not to assume we will use a global flag available on the Heptapod project forever. It could very well depend on the given project.
-
- 17 Nov, 2020 1 commit
-
-
Georges Racinet authored
This one wasn't absorbed. Doesn't seem worth it to make it fit inside that series that predates the flake8 version bump.
-
- 14 Nov, 2020 1 commit
-
-
Georges Racinet authored
See heptapod#368 Now with two sets of scenarios with various refusals, and the check that accepting Merge Requests is recognized to go through another set of permissions (very important to avoid locking situations) --HG-- branch : stable
-
- 03 Nov, 2020 12 commits
-
-
Georges Racinet authored
This is again cleaner, and ultimately the reason why we've put a long timeout and `cleanup_teardown()` in `Project.__exit__()`, even though it's much less necessary now than it was before we made Project names unique.
-
Georges Racinet authored
Less pollution, even though try…finally wasn't so bad.
-
Georges Racinet authored
-
Georges Racinet authored
This works, but we still have uncertainties about MR creation prompt on push and the expected pipelines, so we don't test these.
-
Georges Racinet authored
This will allow for nicer cleanups, especially for those created in tests. We're adopting the strong cleanup capability and the larger timeout in `Project.__exit__()` to mimic what the Project fixtures do, in the hope to have them simply use the new Context Managers.
-
Georges Racinet authored
We chose to pass a whole Project instance, because it's seem very unlikely we'd have a use case for passing an id (caller would have created the target Project or would use a fixture), and therefore it makes things simpler on the average for the caller.
-
Georges Racinet authored
We already had a persistent normal user and the `external_user` fixture. This one is not external.
-
Georges Racinet authored
Proper testing of Git operation will need this.
-
Georges Racinet authored
A bit cleaner this way. One day, we may reduce the duplication by introducing a proper MergeRequest object.
-
Georges Racinet authored
Amazingly, the `owner_api_request()` shortcut was all we needed for a long time, but we'll now will need to be able to pass the user explicitely.
-
Georges Racinet authored
-
Georges Racinet authored
This will help reducing the boiler plate and over long statements. Leveraging in test_import for immediate testing.
-
- 17 Nov, 2020 1 commit
-
-
Georges Racinet authored
The good thing is that we finally get back to sanity with boolean operators. This is now legit: if something and othercondition:
-