[Bioperl-l] SeqFeature::Generic DESTROY [was New: Bug in Bio::SeqFeature::Generic.pm DESTROY causes object corruption.. fix included!]

Jason Stajich jason.stajich at duke.edu
Thu Dec 16 13:42:37 EST 2004


So is this bug going to get resolved by someone?

I don't have a clear idea of how to reproduce it. Has the removal of 
the get_tag_values is implemented removed all of this code in the first 
place?

The reason for that code was to free memory cycles introduced when a
Feature has a reference to the Sequence and then the Sequence also has 
a reference to the Feature.

But it looks like the whole DESTROY function was removed from 
SeqFeature::Generic so I'm unsure if the memory cycle problem has now 
been re-introduced.
Looks like it might be since there is no explict unlinking of 
$obj->{'_gsf_seq'}.

-jason

On Dec 10, 2004, at 11:59 AM, Aaron J. Mackey wrote:

>
> Probably because the offending code in get_tag_values doesn't check 
> for defined($hash{$key}) but just exists($hash{$key}) (and then 
> merrily tries to dereference it as an array).  It's a bug in BioPerl, 
> to be sure, but not as mystical as it seems.
>
> -Aaron
>
> On Dec 10, 2004, at 11:55 AM, Alok Saldanha wrote:
>
>> I still don't understand why adding delete() to DESTROY causes the 
>> problem to go away
>
> --
> Aaron J. Mackey, Ph.D.
> Dept. of Biology, Goddard 212
> University of Pennsylvania       email:  amackey at pcbi.upenn.edu
> 415 S. University Avenue         office: 215-898-1205
> Philadelphia, PA  19104-6017     fax:    215-746-6697
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>
>
--
Jason Stajich
jason.stajich at duke.edu
http://www.duke.edu/~jes12/



More information about the Bioperl-l mailing list