On May 4, 2009, at 09:35, Daniel Berger wrote:
Luis Lavena wrote:
On Mon, May 4, 2009 at 10:39 AM, Daniel Berger <djber...@gmail.com>
wrote:
Luis Lavena wrote:
Hey guys,
The following are the results of RubyGems revision 2193 running
against MinGW and VC6 versions of Ruby.
http://pastie.org/466833
Looks like the one failure either needs to be skipped or the test
itself
needs to be adjusted.
I see VC6 skips 6 tests while MinGW only 4, maybe there is a problem
with the platform evaluation?
Not entirely. The skipped messages that include "VC++" are only
skipped for versions of Ruby built with Visual Studio, not the
cygwin/mingw builds.
Or rather, it looks for a platform match of "mswin". I thought that
excluded cygwin and mingw. Is that not the case?
I thought cygwin supported symlinks. Is that not true?
Perhaps a better approach would be to inspect CONFIG['CC'], or one
of the other attributes.
For now, I added a skip for the failing test with a reason. We can
come back and evaluate it later.
_______________________________________________
Rubygems-developers mailing list
http://rubyforge.org/projects/rubygems
Rubygems-developers@rubyforge.org
http://rubyforge.org/mailman/listinfo/rubygems-developers
