[Bioperl-l] Bio::SeqIO::FTHelper

Hilmar Lapp lapp@gnf.org
Wed, 24 Jan 2001 13:01:54 -0800

Jason Stajich wrote:
> Looking at FTHelper _parse_loc method.
> Do we want our good buddy FTHelper to continue to create tag/value pairs
> for sub_Features to represent things like '_part_feature' and
> '_zero_width_feature'?  Or are we happy with having Split/Fuzzy Locations
> handle this representation?

Maybe I'm missing something, but I think migrating semantics from
undocumented tags to explicit types was one of the objectives. If it
was really undocumented, there shouldn't be client code relying on
those tags outside of the Bioperl core itself. In theory :)

People out there, if you have a client that relies on undocumented
tags in a SeqFeature::Generic, please shout. Otherwise you can reckon
that these tags will be gone. Please also shout if you have clients
relying on documented tags pertaining to location and length (I recall
having added some tags to the documentation in summer or fall last
year, but hopefully no-one noticed :o)


Hilmar Lapp                            email: lapp@gnf.org
GNF, San Diego, Ca. 92121              phone: +1-858-812-1757