You shouldn't have a failing test, you just need to update dependencies.tsv to 've correct before running tests... Which you'd need to do anyway because you changed the deps.
I don't know why Ian's godeps is broken... On Aug 20, 2014 8:37 PM, "Tim Penhey" <tim.pen...@canonical.com> wrote: > A while back I added some useful make targets: > > if the env var `JUJU_MAKE_GODEPS` is "true", then > > make godeps > > runs the godeps -u, and also makes sure it is run before build and check. > > I tend to use this. > > Also, if you are working on integrating changes in other branches that > aren't merged, you'll always have a failing test now, no? > > Tim > > On 21/08/14 12:33, Ian Booth wrote: > > Ah yes godeps -t ./... fails the same way. > > But I never run that. > > > > I just run godeps -u dependencies.tsv > > to update the revs of the imported libs as I switch juju-core branches. > > > > > > On 21/08/14 10:30, Nate Finch wrote: > >> And running godeps -t ./... Didn't give the same error? > >> On Aug 20, 2014 8:26 PM, "Ian Booth" <ian.bo...@canonical.com> wrote: > >> > >>> Don't think so > >>> > >>> ian@wallyworld:~$ which godeps > >>> /home/ian/juju/go/bin/godeps > >>> > >>> ian@wallyworld:~$ ls -l `which godeps` > >>> -rwxrwxr-x 1 ian ian 4821160 Aug 15 19:12 /home/ian/juju/go/bin/godeps > >>> > >>> > >>> > >>> On 21/08/14 10:20, Nate Finch wrote: > >>>> All the test does is run godeps -t ./... It looks for godeps in the > path > >>>> first and then looks in gopath/bin. Do you maybe have a version in > your > >>>> path that is old? > >>>> On Aug 20, 2014 8:15 PM, "Ian Booth" <ian.bo...@canonical.com> wrote: > >>>> > >>>>> I run godeps all the time as I switch between 1.20 and master. It > Just > >>>>> Works. > >>>>> > >>>>> On 21/08/14 10:12, Nate Finch wrote: > >>>>>> What happens when you run godeps normally? > >>>>>> > >>>>>> It should ignore the std lib stuff, but I don't actually know how > >>> that's > >>>>>> implemented. > >>>>>> On Aug 20, 2014 8:02 PM, "Ian Booth" <ian.bo...@canonical.com> > wrote: > >>>>>> > >>>>>>> Hmmmm. The test fails for me. > >>>>>>> > >>>>>>> > >>>>>>> FAIL: dependencies_test.go:42: dependenciesTest.TestGodepsIsRight > >>>>>>> > >>>>>>> dependencies_test.go:77: > >>>>>>> ... > >>>>>>> dependencies_test.go:70: > >>>>>>> c.Fatal(string(out)) > >>>>>>> ... Error: godeps: no version control system found for > >>>>>>> "/usr/lib/go/src/pkg/bufio" > >>>>>>> godeps: no version control system found for > >>> "/usr/lib/go/src/pkg/bytes" > >>>>>>> godeps: no version control system found for > >>> "/usr/lib/go/src/pkg/errors" > >>>>>>> godeps: no version control system found for > "/usr/lib/go/src/pkg/io" > >>>>>>> godeps: no version control system found for > "/usr/lib/go/src/pkg/sync" > >>>>>>> godeps: no version control system found for > >>>>>>> "/usr/lib/go/src/pkg/sync/atomic" > >>>>>>> ... > >>>>>>> ... > >>>>>>> > >>>>>>> > >>>>>>> On 21/08/14 05:18, Nate Finch wrote: > >>>>>>>> In an effort to make sure that dependencies.tsv is actually > correct, > >>>>> the > >>>>>>>> tests in github.com/juju/juju now run godeps -t ./... and compare > >>> its > >>>>>>>> output to dependencies.tsv. The test looks for godeps first in > the > >>>>>>> $PATH, > >>>>>>>> and then in the first $GOPATH/bin. If it is not found, it > currently > >>>>>>> skips > >>>>>>>> the test (because CI doesn't have godeps when it runs the tests). > >>>>>>>> Eventually I'd like to make the test not skippable, since anyone > >>>>> running > >>>>>>>> the tests (except CI) needs godeps anyway... but to get this in > >>> sooner > >>>>>>>> rather than later, I think it's ok to let it be skipped. > >>>>>>>> > >>>>>>>> Note that in submitting the PR, I actually had to fix > >>> dependencies.tsv > >>>>> to > >>>>>>>> get the test to pass, because it was wrong.... which is exactly > why > >>>>> this > >>>>>>>> test exists. > >>>>>>>> > >>>>>>>> You can now run go test from github.com/juju/juju and it'll tell > you > >>>>> if > >>>>>>>> something's wonky with dependencies.tsv... it'll tell you if there > >>> are > >>>>>>>> entries in it that aren't needed, it'll tell you if you're missing > >>>>>>> entries > >>>>>>>> that should be in it, and it'll tell you if one of your branches > is > >>> on > >>>>>>> the > >>>>>>>> wrong commit. > >>>>>>>> > >>>>>>>> *Note:* this means you can't just willy-nilly append your latest > >>>>>>> dependency > >>>>>>>> to dependencies.tsv. They need to be in alphabetical order. The > >>>>> easiest > >>>>>>>> way to do that is just to run godeps -t ./... > dependencies.tsv > from > >>>>>>>> github.com/juju/juju. That'll produce the right output. > Obviously, > >>>>>>> check > >>>>>>>> the diff to make sure the changes it produces are the ones you > >>> expect. > >>>>>>>> > >>>>>>>> This also means that we need to ensure godeps produces the right > >>> output > >>>>>>> on > >>>>>>>> all OSes. If you have a dependency that exists only in an > >>> OS-specific > >>>>>>>> file, the godeps test will fail on OSes other than that one (e.g. > if > >>>>>>>> there's a dependency in an _windows.go file, the test will fail on > >>>>> linux, > >>>>>>>> because godeps doesn't see the windows dependencies). To fix > this, > >>> the > >>>>>>>> easiest thing to do is import the same library in an OS-agnostic > >>> file, > >>>>>>> and > >>>>>>>> give it a name of underscore, which will prevent the unused > package > >>>>>>> error, > >>>>>>>> thusly: _ "foo.com/my/win/lib" > >>>>>>>> > >>>>>>>> -Nate > >>>>>>>> > >>>>>>>> > >>>>>>>> > >>>>>>> > >>>>>>> -- > >>>>>>> Juju-dev mailing list > >>>>>>> Juju-dev@lists.ubuntu.com > >>>>>>> Modify settings or unsubscribe at: > >>>>>>> https://lists.ubuntu.com/mailman/listinfo/juju-dev > >>>>>>> > >>>>>> > >>>>> > >>>> > >>> > >> > > > >
-- Juju-dev mailing list Juju-dev@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/juju-dev
