| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Ensure we set SUITE_FLAKY_RSPEC_REPORT_PATH to nil in RspecFlaky::Listener spec | Rémy Coutable | 2017-10-06 | 1 | -0/+1 |
| | | | | | Signed-off-by: Rémy Coutable <remy@rymai.me> | ||||
| * | Add new RSpecFlaky::FlakyExamplesCollection and RSpecFlaky::Config classes | Rémy Coutable | 2017-10-06 | 3 | -1/+181 |
| | | | | | Signed-off-by: Rémy Coutable <remy@rymai.me> | ||||
| * | Ensure RSpecFlaky doesn't automatically update flaky examples | Rémy Coutable | 2017-10-04 | 2 | -82/+220 |
| | | | | | | | | | | | | | | | | | | | | | | Previously, instantiating a RspecFlaky::FlakyExample object would automatically update its first_flaky_at, last_flaky_at and last_flaky_job. That was wrong because we would overwrite every time the suite report with this false data. We now: - Get the suite report and only read from it - Write only the currently detected flaky examples in the report, so that the final report is only updated with flaky examples that were actually detected in each job. Before, job1 could overwrite the legit report from job2! - Write the newly detected flaky examples by rejecting the already tracked flaky specs instead of using another hash. Signed-off-by: Rémy Coutable <remy@rymai.me> | ||||
| * | Use a new RspecFlakyListener to detect flaky specs | Rémy Coutable | 2017-08-11 | 3 | -0/+371 |
| Signed-off-by: Rémy Coutable <remy@rymai.me> | |||||
 |
index : delta/gitlab/gitlab-ce.git | |
| gitlab.com: gitlab-org/gitlab-ce.git |
| summaryrefslogtreecommitdiff |