Closed jayvdb closed 7 years ago
c.f. https://docs.gitlab.com/ce/user/project/merge_requests/index.html for the fact that it should be merge-requests
errrrr… you're most likely right, but I did not invent those cassette records, so it has worked at some point. Let me double check the whys and hows,and test your fix.
But as the RTFM say so, I'm pretty sure it's alright and merge material 😉
I've asked if the branch name has changed at https://gitter.im/gitlabhq/gitlabhq?at=58c6bca109e7ba8510bcd274
@jayvdb you're welcome on irc.freenode.org #git-repo for a chat!
Can we please get some traction on this.
Please look at https://gitlab.com/gitlab-org/gitlab-ce/commit/86bccb71a1d4 , which is August 2015, ensuring that all MRs (even MRs from branches instead of forks) have a usable ref. The documentation there uses merge-requests
. So for as long as it has been possible to always use a ref, the documentation has always said the ref will be merge-requests/*/head
. (see moves like https://gitlab.com/gitlab-org/gitlab-ce/commit/7973a22f and https://gitlab.com/gitlab-org/gitlab-ce/commit/4f5bb980 to follow that original .md to its current location).
tests/integration/cassettes/test_gitlab_test_19_request_fetch.json
and tests/integration/cassettes/test_gitlab_test_19_request_fetch__bad_request.json
both have only one commit (e58373e21, 11 Oct 2016), and neither mentions merge-requests
or merge_requests
. ergo, the cassettes themselves have no opinion on whether it should be merge-requests
or merge_requests
.
That same commit had test_19_request_fetch
using remote_branch='merge_requests'
and the gitlab adapter changed:
- 'merge-requests/{}/head'.format(request),
+ 'merge_requests/{}/head'.format(request),
But somehow you created those cassettes. My guess is that you obtained the cassette before the change, and then you changed the code above, thus creating the bug, and committed the cassettes and code changes together, resulting in a buggy test method. Anyway, it really doesnt matter if merge_requests
ever worked somehow ... if it did, it was luck alone, as it isnt in accordance with the documentation.
Sorry for the loooong delay merging this small commit! But better late than never! 😉
I'm preparing a release for this week end, so stay tuned
Reverts one chunk of e58373e2