[Bioperl-l] feedback on small patch for bioperl-run's t/Neighbor.t

Jason Stajich jason.stajich at duke.edu
Thu Sep 29 21:36:20 EDT 2005


sure - we're testing if we can run neighbor not rigorously testing  
whether or not it gives us the right answer , etc.  There should only  
be 3 leaves so another way to do it is, but testing in the grep on  
the 'id' is fine since that is what we're using later on.

my @nodes = sort { $a->id cmp $b->id }  grep { $_->is_Leaf } $tree- 
 >get_nodes;

-j
On Sep 29, 2005, at 6:10 PM, George Hartzell wrote:

>
> t/Neighbor.t fails for me, the patch below fixes it.
>
> I'm running w/ -live and -run from CVS this morning, and Phylip3.6
> (and the proper env. variable set).
>
> I'm getting four nodes, one of which has an id that's ''.  By touching
> up the test code so that the id field must be defined and non-empty,
> things work, and the existence of that filter suggests that it's ok
> for there to be more than the three expected nodes in the list.
>
> Does this patch make sense, or am I missing something subtle?
>
> g.
>
> Index: t/Neighbor.t
> ===================================================================
> RCS file: /home/repository/bioperl/bioperl-run/t/Neighbor.t,v
> retrieving revision 1.7
> diff -c -3 -r1.7 Neighbor.t
> *** t/Neighbor.t    18 Dec 2002 13:43:45 -0000    1.7
> --- t/Neighbor.t    29 Sep 2005 21:15:22 -0000
> ***************
> *** 77,83 ****
>
>   ($tree) = $tree_factory->create_tree($inputfilename);
>   my @nodes = sort { $a->id cmp $b->id }
> !             grep { defined $_->id} $tree->get_nodes();
>
>   ok($nodes[2]->id, 'SINFRUP002');
>   ok($nodes[1]->id, 'SINFRUP001');
> --- 77,83 ----
>
>   ($tree) = $tree_factory->create_tree($inputfilename);
>   my @nodes = sort { $a->id cmp $b->id }
> !             grep { defined $_->id and $_->id} $tree->get_nodes();
>
>   ok($nodes[2]->id, 'SINFRUP002');
>   ok($nodes[1]->id, 'SINFRUP001');
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>

--
Jason Stajich
Duke University
http://www.duke.edu/~jes12




More information about the Bioperl-l mailing list