Front page | perl.perl5.porters |
Postings from June 2012
On Sat Jun 23 23:38:36 2012, sprout wrote:
> On Fri Jun 22 15:18:16 2012, dmcbride@cpan.org wrote:
> > Trying to test another patch, I copy perl from git blead to my AIX 6.1
> > box, where perl 5.16.0 is compiling and testing just fine and get:
> >
> > ../lib/File/stat.t
> > (Wstat: 256 Tests: 5008 Failed: 1)
> > Failed test: 4911
> > Non-zero exit status: 1
> >
> > Which is:
> >
> > not ok 4911 - correct overload for -l for ./perl
>
> I just tried compiling the smoke-me/slop (slab for ops) branch with g++
> instead of gcc, and I get the same failure on Mac OS X. I doubt the
> slab allocator has anything to do with it.
It seems that gcc produces this:
$ ./perl -Ilib -e 'warn $^X'
/Users/sprout/Perl/perl.git/perl at -e line 1.
while g++ produces this:
$ ./perl -Ilib -e 'warn $^X'
./perl at -e line 1.
I had no idea the compiler could affect $^X. gcc seems to be making
perl resolve symlinks, too, because I get the same result in the t
directory. (I may be misdiagnosing it.)
The script is using $^X for testing, which is why it could produce
different results.
In any case, this produces the same output for both compilers:
use File::stat;
my $stat = File::stat::stat('./perl');
warn eval '-l $stat';
warn eval '-l "./perl"';
__END__
Warning: something's wrong at - line 3.
1 at - line 4.
So it seems File::stat *is* buggy, and the test is just not good enough
half the time. Or the test is wrong.
I think the test is wrong, as -l _ will fail after a stat. You have to
do lstat for -l _ to work. Similarly, -l $stat_obj should only return
true if the $stat_obj was returned by File::stat::lstat. For
File::stat::stat, -l should either return false or die. Currently it
returns false. -l _ dies after stat.
--
Father Chrysostomos
---
via perlbug: queue: perl5 status: open
https://rt.perl.org:443/rt3/Ticket/Display.html?id=113796
Thread Previous
|
Thread Next