Skip Menu |

This queue is for tickets about the Want CPAN distribution.

Report information
The Basics
Id: 4628
Status: stalled
Priority: 0/
Queue: Want

People
Owner: Nobody in particular
Requestors: Marek.Rouchal [...] gmx.net
Cc:
AdminCc:

Bug Information
Severity: Important
Broken in: 0.07
Fixed in: (no value)



Subject: segfault in Set::Array
Hello, as you may probably know, there is a problem with Set::Array which uses Want: I get segmentation faults on Linux and Solaris with my perl-5.8.2, gcc-3.2.3, ithreads, shared libperl when e.g. running a difference on two Set::Array objects. Sorry, I had no time yet to put together a minimal test case, but this one triggers the issue: use Set::Array; my $set1 = Set::Array->new(qw(abc def ghi jkl mno)); my $set2 = Set::Array->new(qw(def jkl pqr)); print "BEFORE\n"; my @set3 = $set1 - $set2; print "AFTER\n"; (the problem occurs regardless of the type of set3) It seems to be triggered by the context of overloaded operators. The coredump happens in Want.pm at 107: my $n = parent_op_name(shift()); Let me know if you need any additional info to fix the problem. Unfortunately, I am not very familiar with the perl guts, so I can't help much with XS patches. Cheers, Marek
RT-Send-CC: djberg [...] cpan.org
This is a known limitation of Want, I'm afraid. I'm sorry it wasn't documented. In newer versions of Want it causes an error rather than a segfault, which doesn't necessarily help (though perhaps you can trap the error with an eval {} and fall back to some reasonable default?) I'm CCing this to the author of Set::Array. Robin