cjfields at uiuc.edu
Thu Jun 19 17:29:45 EDT 2008
On Jun 19, 2008, at 3:47 PM, Mark Johnson wrote:
> On Thu, Jun 19, 2008 at 3:19 PM, Chris Fields <cjfields at uiuc.edu>
>> I think the general idea of Bio::FeatureIO will remain (read/write
>> data) but it will definitely undergo significant reimplementation.
>> typed SeqFeatureI class (Bio::SeqFeature::Annotated) would be
>> deprecated in
>> favor of something more lightweight.
>> However, I don't see that happening until after 1.6 is released
>> someone wants to take it on.
> I suspect the 'something' you mention is going to require a fair bit
> of discussion and design. The best time for that would be after 1.6.
> Otherwise, we're likely to end up with something just as problematic
> as Bio::SeqFeature::Annotated.
> However, people are using Bio::FeatureIO as it stands. Unless the
> plan is to remove it from source control (as opposed to just not
> packaging it up as part of the 1.6 release), would there be any
> objection to patching up the existing implementation? If nothing
> else, it will be the starting point for the reimplementation. Might
> as well correct obvious defects, so they *don't* get reimplemented.
I think patching for the current implementation is fine. We we plan
on making changes we'll try posting plans on the wiki.
Speaking of, I need to look at the FeatureIO API to see what is
expected. Hopefully changes can occur with minimal to no API changes.
More information about the Bioperl-l