Geoffrey Young wrote:
# testing : $r->finfo->group()
# expected: 4294967294
# received: -2
not ok 8


$ id ianh
uid=501(ianh) gid=20(staff) groups=20(staff), 80(admin)


ok, well if neither perl nor apr give the real answer then we have no choice
but to simply sweep it under the carpet - in both cases we merely pass along
the results we're given, and if perl proper doesn't give the real answer
either there must be a problem within the OS itself.

so, give this patch a whirl.  I'm not happy about doing it, but as with
win32 it seems we're kind of out of options if the two pieces of software we
rely on can't give true answers on darwin.

--Geoff


------------------------------------------------------------------------


Index: Apache-Test/lib/Apache/TestConfig.pm
===================================================================
RCS file: /home/cvs/httpd-test/perl-framework/Apache-Test/lib/Apache/TestConfig.pm,v
retrieving revision 1.218
diff -u -r1.218 TestConfig.pm
--- Apache-Test/lib/Apache/TestConfig.pm 5 Apr 2004 04:35:21 -0000 1.218
+++ Apache-Test/lib/Apache/TestConfig.pm 8 Apr 2004 13:49:18 -0000
@@ -18,6 +18,7 @@
use warnings FATAL => 'all';
use constant WIN32 => $^O eq 'MSWin32';
+use constant OSX => $^O eq 'darwin';
use constant CYGWIN => $^O eq 'cygwin';
use constant NETWARE => $^O eq 'NetWare';
use constant SOLARIS => $^O eq 'solaris';
Index: t/response/TestAPR/finfo.pm
===================================================================
RCS file: /home/cvs/modperl-2.0/t/response/TestAPR/finfo.pm,v
retrieving revision 1.8
diff -u -r1.8 finfo.pm
--- t/response/TestAPR/finfo.pm 14 Jan 2004 21:27:41 -0000 1.8
+++ t/response/TestAPR/finfo.pm 8 Apr 2004 13:49:19 -0000
@@ -8,6 +8,7 @@
use Apache::TestTrace;
use Apache::TestConfig;
use constant WIN32 => Apache::TestConfig::WIN32;
+use constant OSX => Apache::TestConfig::OSX;
use Apache::RequestRec ();
use APR::Finfo ();
@@ -55,10 +56,17 @@
our ($device, $inode, $protection, $nlink, $user, $group,
undef, $size, $atime, $mtime, $ctime) = stat $file;
- # skip certain tests on Win32 (and others?)
- # atime is wrong on NTFS, but OK on FAT32
- my %skip = WIN32 ?
- (map {$_ => 1} qw(device inode user group atime) ) : ();
+ # skip certain tests on Win32 and others
+ my %skip = ();
+
+ if (WIN32) {
+ # atime is wrong on NTFS, but OK on FAT32
+ %skip = map {$_ => 1} qw(device inode user group atime);
+ }
+ elsif (OSX) {
+ # both apr and perl report incorrect group values
+ %skip = (group => 1);
+ }
# compare stat fields between perl and apr_stat
{

In which case, this needs to be documented in the APR::Finfo manpage. i.e. that one better not rely on the above attributes, since they won't work on all platforms, listing the explicit problems. Thanks.


__________________________________________________________________
Stas Bekman            JAm_pH ------> Just Another mod_perl Hacker
http://stason.org/     mod_perl Guide ---> http://perl.apache.org
mailto:[EMAIL PROTECTED] http://use.perl.org http://apacheweek.com
http://modperlbook.org http://apache.org   http://ticketmaster.com

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Reply via email to