[Bioperl-l] Bio::SeqFeature::Generic add_sub_SeqFeature

[Jason Stajich]

What is the feeling here, we have this old way of doing things which
included using the value 'EXPAND' to determine if we should expand the
start/end space for a feature when adding a sub_SeqFeature to a feature?

I think this should likely be better modeled through a SplitLocationI
which is just a container of LocationObjects.  So I propose to remove all
references to 'EXPAND' which means removing the method _expand_region and
updating add_sub_Feature to deal with adding the locations.  Similarly the
flush_sub_SeqFeature should flush the locations, but I'm not sure about
what the start/end should be reset to...

I also had to update FeaturePair to add the method location() which
delegates to feature1()->location() otherwise things won't work correctly. 
start/end are defined by feature1 object so location should also reside
in feature1.

Jason

Jason Stajich
jason@chg.mc.duke.edu
Center for Human Genetics
Duke University Medical Center 
http://www.chg.duke.edu/