Re: test-open failure on Hurd

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Eric Blake on 10/2/2009 6:32 PM: > I'll let others decide how Hurd will fix /dev/null, or if a fix is even > necessary. But meanwhile, the point of test-open was to test that opening > a non-directory with a trailing slash fails, not whet

Re: test-open failure on Hurd

Eric Blake, le Fri 02 Oct 2009 18:32:20 -0600, a écrit : > But meanwhile, the point of test-open was to test that opening > a non-directory with a trailing slash fails, Ok. > Therefore, we can fix the testsuite to be more portable by testing > something other than /dev/null/. I'm checking this i

Re: test-open failure on Hurd

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 According to Eric Blake on 10/2/2009 6:32 PM: > But meanwhile, the point of test-open was to test that opening > a non-directory with a trailing slash fails, not whether Hurd complies. > Therefore, we can fix the testsuite to be more portable by testin

test-open failure on Hurd

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 [adding bug-gnulib] According to Samuel Thibault on 10/2/2009 3:36 PM: > Eric Blake, le Fri 02 Oct 2009 15:32:01 -0600, a écrit : >> http://www.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap10.html > > Yep, that's what I was reading. > >> Alth