Closed LawnGnome closed 3 years ago
Dear all,
This is your release captain speaking. 🚂🚂🚂
Branch cut for the 3.19 release is scheduled for tomorrow.
Is this issue / PR going to make it in time? Please change the milestone accordingly. When in doubt, reach out!
Thank you
Closing this because it looks like we don't think it's important. Reopen if necessary.
@LawnGnome Something that came to my mind this morning: do we have any integration-y tests for GitLab? As far as I can tell in #11757 the GitLab client is mocked, so we don't record request/response in cassettes.
In our resolver tests we do have this one here,
TestCampaigns
that tracks a Bitbucket Server and a GitHub changeset and makes sure that the changeset resolvers work. But in #11675 I already ripped out that part, because thecreateChangesets
mutation won't exist anymore.I'm thinking that maybe we should extract that part of the test and create a
TestChangesetResolver
test that uses VCR-recorded and the rest of our stack to make sure that a fully-synced changeset from GH/GL/BBS works with theChangesetResolver
. What do you think?The other test that I thought about it is this one: https://github.com/sourcegraph/sourcegraph/blob/870221e07c86de5e67d3900a804ce6547b595fd3/enterprise/internal/campaigns/resolvers/resolver_test.go#L651
That test currently only tests GitHub changesets on an integration-y level, but since
changesetCountsOverTime
, which is used for the burndown chart, has been tricky in the past, I wonder whether we shouldn't extend that.And now to sum all of this up: I think we should have a test somewhere that makes sure that the flow of
actual-data-from-gitLab->Sourcegraph->Sourcegraph-GraphQL-API
works, but I leave it up to you to decide where/how/when to add this 😄Originally posted by @mrnugget in https://github.com/sourcegraph/sourcegraph/issues/11586#issuecomment-658582686