[Bioperl-l] Bio::Species bug, was Species.t tests failing with bioperl-live

[George Hartzell]

Chris Fields writes:
 > 
 > On Sep 26, 2008, at 8:45 AM, Sendu Bala wrote:
 > 
 > > Chris Fields wrote:
 > >> On Sep 26, 2008, at 3:26 AM, Sendu Bala wrote:
 > >>> Chris Fields wrote:
 > >>>> The last commit to Bio::Species, which was an effort to fix a  
 > >>>> memory leak, has unfortunately broken Species.t tests.  The  
 > >>>> relevant commit used weaken(); my guess is something is being  
 > >>>> garbage collected way too early.  Any ideas?
 > >>>
 > >>> Was that last commit motivated by a bug report, and/or is there an  
 > >>> associated test? If not, perhaps it can just be reverted?
 > >> Here's the relevant list thread:
 > >> http://thread.gmane.org/gmane.comp.lang.perl.bio.general/17541/focus=17550
 > >> The proposed patch was George's, which gets rid of the leak but  
 > >> causes the Species.t failures.  I say we revert the patch, but  
 > >> we'll still need to address the original problem addressed by  
 > >> Rutger Vos, i.e. that there is a significant memory leak in  
 > >> Bio::Species.
 > >
 > > Thanks. Rather than revert I've kludged it for now. Everything  
 > > should be fine, it's just that I fear my solution isn't 'correct'  
 > > and could mess things up we don't happen to have tests for.
 > >
 > > See http://bugzilla.open-bio.org/show_bug.cgi?id=2594
 > 
 > Probably something that will go away when we quit using Bio::Species  
 > altogether (in favor of Bio::Taxon) after 1.6.   I'm not sure whether  
 > it will be worth the trouble, but it's worth discussing.
 > 
 > Anyway, I'm still a bit unfamiliar with some aspects of Bio::Species  
 > (and I'm pretty busy ATM) so it make take a few days.

I can try to take a look at it again and see if there's a way to make
it work [more] correctly.

g.