1. 03 Nov, 2020 3 commits
  2. 17 Nov, 2020 1 commit
  3. 15 Oct, 2020 1 commit
  4. 12 Oct, 2020 1 commit
    • Georges Racinet's avatar
      New `--heptapod-hg-native` option · b2a5da28ceb5
      Georges Racinet authored
      This option makes the functional tests create Mercurial native
      projects instead of the default type (currently `hg_git`).
      
      Hence it will be the main tool to validate HGitaly related
      developments. Later on, we might want to introduce dedicated
      CI jobs. Switching to parametrization can also be considered,
      but it wouldn't change so much in CI, because we now have tests
      sharding.
      
      The new flag is stored on the `Heptapod` object, for lack of
      better choice in the short term (arguably, that's not a property
      of the Heptapod server).
      b2a5da28ceb5
  5. 06 Oct, 2020 1 commit
    • Georges Racinet's avatar
      Omnibus: stop hardcoding server-side path to hg · 55b1ee0b3b1b
      Georges Racinet authored
      For server-side operations, we should use the one provided by
      the server, obviously.
      
      The immediate need is that it varies amongst installation types
      With The new Omnibus packaging method (omnibus-heptapod!3)
      for Heptapod 0.17, it changes from /usr/local/bin (expected to be on PATH)
      to /opt/gitlab/embedded/bin.
      
      There are several TODOs:
      
      - read `hg_executable` from config, with remote instances being
        the only exceptions (but these wouldn't call server-side subprocesses
        from the tests).
      - provide a direct helper method on `Heptapod`
      - do the same for Git
      55b1ee0b3b1b
  6. 27 Sep, 2020 1 commit
    • Georges Racinet's avatar
      Project.get_merge_request_commit_links now returns the whole row · 911d63829706
      Georges Racinet authored
      We still wait on the link to be displayed, but returning the whole
      `li` element will be useful for finer assertions, e.g., about
      commit SHAs.
      
      The consequence is that the `text` attribute of the returned
      elements is broader, encompassing the commit title, authorship and
      short id.
      
      --HG--
      branch : stable
      911d63829706
  7. 25 Sep, 2020 1 commit
  8. 24 Sep, 2020 1 commit
  9. 28 Aug, 2020 1 commit
    • Georges Racinet's avatar
      Project: adapted webdriver_destroy() for GitLab 13.3 · 623db7fc5767
      Georges Racinet authored
      The confirmation modal has changed a lot (the principle
      stays the same), including the text to type, that we now
      simply copy for better independence (now has a verb and the
      full path).
      
      We've observed the `id` for the `modal` element to differ between
      source and Docker contexts. Using the `<div class="modal-dialog>`
      ancestor wouldn't be appropriate as the corresponding CSS selector
      returns another one.
      623db7fc5767
  10. 23 Jul, 2020 2 commits
  11. 17 Jul, 2020 1 commit
  12. 16 Jul, 2020 1 commit
    • Georges Racinet's avatar
      Project: high level methods to switch to legacy and back · 2335c75d6a79
      Georges Racinet authored
      Previously we were relying on the global application setting to
      infer if a project is on legacy or hashed storage. This was
      obviously approximative, but on GitLab 13, that setting doesn't
      exist any more and all new projects are on the hashed storage.
      
      We introduce here a new boolean property on the project and
      methods to switch from hashed to legacy and back.
      2335c75d6a79
  13. 03 Jun, 2020 1 commit
  14. 06 Jun, 2020 1 commit
    • Georges Racinet's avatar
      Better detection and cleanup in creation test by external user · 0ab9dbd37a6a
      Georges Racinet authored
      To recall, in this test, the user has no quota to create personal
      projects, but can still create a project in a group.
      
      In some cases (and especially on the future GitLab 12.10), the
      removal of the group wouldn't remove the just created project, or
      at least not right away, making subsequent runs fail because there's
      already a project with that name. We could have made the project
      name unique, but it's better to leak less and to tighten the detection.
      0ab9dbd37a6a
  15. 05 Jun, 2020 2 commits
  16. 03 Jun, 2020 3 commits
  17. 27 May, 2020 1 commit
    • Georges Racinet's avatar
      Project: API method for MR rebase · 6f7ee862e8c1
      Georges Racinet authored
      It will wait automatically for the rebase to be finished and
      assert its success.
      
      We could later on add options to test bad rebases if we need to.
      6f7ee862e8c1
  18. 09 May, 2020 1 commit
    • Georges Racinet's avatar
      Project: methods for deploy tokens handling · 4d5d5ae24372
      Georges Racinet authored
      The API method won't work for a while, wrote them before noticing
      the version information in online documentation.
      
      So we had to make a lengthy webdriver_create_deploy_token, hoping
      the page structure won't change too much before we can rely on
      the API methods.
      4d5d5ae24372
  19. 19 May, 2020 1 commit
  20. 14 May, 2020 1 commit
  21. 12 May, 2020 2 commits
  22. 07 May, 2020 1 commit
    • Georges Racinet's avatar
      Project: api method to retrieve a commit details · abc3edefc1ce
      Georges Racinet authored
      At this stage of Heptapod, the API returns Git SHAs, so we have to use it
      again to get the details. Usually our tests are made so that a commit
      can be identified by its message (`'title'` key of the returned dict)
      
      --HG--
      branch : heptapod-0-12
      abc3edefc1ce
  23. 28 Apr, 2020 2 commits
  24. 02 Apr, 2020 1 commit
    • Georges Racinet's avatar
      Project: invalidate cached FS path after transfer · d296f0a3bd71
      Georges Racinet authored
      By the time that cache had been introduced, the tests were
      running on the hashed storage only (GitLab 12 default). In that
      case, transferring a project does not change the repo path on
      the filesystem, but it does on the legacy storage.
      d296f0a3bd71
  25. 27 Mar, 2020 2 commits
  26. 25 Mar, 2020 1 commit
  27. 14 Mar, 2020 2 commits
    • Georges Racinet's avatar
      tests_hgrc_put: using new API GET calls · 0aa112f36cd7
      Georges Racinet authored
      This makes one less test to rely on FS access, and that one was
      creating doubt in case of failures in tests that use the PUT for
      their setup.
      0aa112f36cd7
    • Georges Racinet's avatar
      project: adapted FS helpers for GitLab 12 · 8036927dba8b
      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.
      8036927dba8b
  28. 10 Mar, 2020 1 commit
    • Georges Racinet's avatar
      MR detection: testing the commit lists panel instead of 'additional' · 1e61fd04ebf5
      Georges Racinet authored
      Our current solution for updating commits upon rebase/push
      does not produce "user x added n commits" in the discussion page,
      yet it fills the commit panel correctly. We consider that
      to be good enough.
      
      Reading the commit panel is done as a new method on the project
      so that we can easily test that it is as expected in other MR cases
      (has been problematic with GitLab 12)
      1e61fd04ebf5
  29. 29 Feb, 2020 1 commit
  30. 14 Mar, 2020 1 commit