1. 30 Jul, 2015 1 commit
  2. 26 Jun, 2015 1 commit
  3. 17 Jun, 2015 1 commit
    • Dave Cheney's avatar
      check: fix data race on C.status · 758efeaa
      Dave Cheney authored
      Fixes #43
      Updates #35
      This PR removes a data race on C.status if c.Error or c.Assert fails in a goroutine spawned inside a test suite.
      Renaming the variable to `C._status` is regretable, but made the change to the code accessing the variable via helpers smaller.
      Tests were _not_ run as they do not pass on recent versions of Go. This fix was verified by asserting that LP 1463643 was resolved.
  4. 10 Jun, 2015 1 commit
  5. 24 Oct, 2014 1 commit
  6. 23 Oct, 2014 1 commit
  7. 20 Oct, 2014 2 commits
  8. 27 Aug, 2014 1 commit
  9. 24 Jul, 2014 1 commit
  10. 09 Jul, 2014 1 commit
  11. 03 Jul, 2014 6 commits
  12. 01 Apr, 2014 9 commits
  13. 25 Feb, 2014 3 commits
    • Roger Peppe's avatar
      Use readable time stamps on log messages. · eb6ee6f8
      Roger Peppe authored
      On long running tests, the time might wrap:
      [LOG] 72.15087 JUJU state: opening state; zookeeper addresses: ["ec2-23-20-157-185.compute-1.amazonaws.com:2181"]
      [LOG] 78.46405 JUJU state: waiting for state to be initialized
      destroy env
      [LOG] 13.73135 JUJU environs/ec2: destroying environment "sample-6326978ae3647455"
      This CL makes the time stamp format as a duration since
      the start of the test in the format agreed before:
      [LOG] 0:00.568 DEBUG juju.environs.tools reading v1.* tools
      [LOG] 0:00.569 INFO juju environs/testing: uploading FAKE tools 1.17.3-trusty-amd64
      [LOG] 0:00.570 INFO juju environs/testing: uploading FAKE tools 1.17.3-precise-amd64
      [LOG] 0:00.571 DEBUG juju.environs.tools no architecture specified when finding tools, looking for any
    • Gustavo Niemeyer's avatar
      Fix panic traceback cutting. · 2368793d
      Gustavo Niemeyer authored
    • Gustavo Niemeyer's avatar
  14. 18 Feb, 2014 4 commits
  15. 27 Jan, 2014 1 commit
  16. 02 Mar, 2013 1 commit
  17. 01 Mar, 2013 5 commits
    • Gustavo Niemeyer's avatar
    • Dave Cheney's avatar
      gocheck: serialize log writes · 495014f4
      Dave Cheney authored
      Fixes LP #1084878.
      The race detector cannot profile gocheck tested code due to this race.
      R=rog, niemeyer, fwereade
    • Gustavo Niemeyer's avatar
      Add -gocheck.list flag. · 9d38355f
      Gustavo Niemeyer authored
      This is an integration of Roger's change:
      I've just fixed the conflicts of merging with tip.
    • Gustavo Niemeyer's avatar
      Merged trunk. · bfab481a
      Gustavo Niemeyer authored
    • John A Meinel's avatar
      This fixes a bunch of failing tests on Windows. · 11552f46
      John A Meinel authored
      Specifically, a lot of tests were asserting something like:
      FAIL: filename.go ...
      While on Windows, the full paths are given:
      FAIL: C:/dev/go/gocheck/filename.go
      I dug into it, and it turns out that nicePath was comparing
      function.FileLine() to os.Getwd(). However, function.FileLine() returns
      'C:/path/to/foo' while 'os.Getwd()' returns 'C:\path\to\foo'.
      I also did one more quick change, which was to always append the trailing '/'.
      That way we don't have to do that on every call to nicePath() (and I checked
      that only nicePath uses initWD).
      At the moment, there are only 2 remaining tests failing. Both because the
      failure messages don't include the module name. This is with golang r58. So it
      is possible that golang changed something (so it needs to be fixed anyway), or
      it is possible it just works differently on Windows :(.