Brian Denny <brian@briandenny.net> writes:
> > > so it seemed weird to me to add a test that would separately import
> > > an executable file. but it might be worth it in order to not affect
> > > the other tests (and not change the greek tree).
> >
> > I don't see a problem with testing the separate import. In fact, we
> > currently have no tests (that I know of) that test the import of a
> > single file -- perhaps this can be a "two-for" :-)
>
> testing the separate import seems to be going well. i should be able to
> submit a patch today or tonight that will take this approach, and revert
> the 'kappa' change.
>
> a "two-for" sounds good, but this test was going to be 'posix-only' since
> it tests the executable property. so maybe we should have a separate
> test (very similar, but simpler) that tests import on any platform.
The tests itself is posix-only? As in it only runs on posix
platforms? I don't recall your original test doing that, and don't
see the need for the new test to do so either. Further, I'm worried
about any situation that might result in a test seeming to exist on
one platform and not another (you know, it's nice to say "basic-test
9" instead of "basic-test 9, or basic-test 8 on windows"). :-)
It seems the 'posix-only' portions of the test are small enough
to just be conditional pieces of a generic test that should run
everywhere.
But whatever. You supply the patch, we'll supply the criticism. :-)
Thanks for looking into this.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Tue Oct 8 18:17:59 2002