[Bioperl-l] [ANNOUCEMENT] BioPerl 1.6 RC3

[Chris Fields]

On Jan 20, 2009, at 10:38 PM, Nathan S. Watson-Haigh wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Scott Cain wrote:
>> Hi Chris,
>>
>> I had really hoped to have Bio::DB::SeqFeature::Store::DBI::Pg  
>> working
>> by the release of 1.6, but it will have to wait for a point release  
>> at
>> some time in the near future.  I can't get it to pass tests, so  
>> please
>> remove it from the 1.6 release.
>>
>> Scott
>>
>>
>
>
> Can I ask a question regarding inclusion of modules back into  
> BioPerl after the 1.6 release? Is the
> intent to allow the addition of extra features/functionality to the  
> 1.6 branch or is the branch just
> for bug fixes? I would have thought it better to keep branches for  
> point releases i.e. bug fixes
> (merged from trunk over to the relevant branch or branches) and for  
> additional
> features/functionality to be kept until the next minor release such  
> as 1.7
>
> e.g.
>
>
> - -- 1.6 ---------------------------- 1.7 -------------- trunk
>     \                                \
>      \                                \----- 1.7.1 -- 1.7 branch
>       \
>        \----- 1.6.1 --- 1.6.2 --------1.6.3 --------- 1.6 branch
>
>
> Just thought some clarification might be needed?
> Nath

In this case the only additional functionality is allowing one to use  
an alternative to mysql/BDB/memory (it's a 'plugin' for  
SeqFeature::Store).  The interfaces for SF::Store are pretty well- 
defined so I would expect this module to respect that API.  (Not to  
mention the same tests are run for all the plugins, so it sorta forces  
one to try to get them all passing).

It's akin to adding a new SeqIO parser; the only thing that would  
change is having an additional format to parse, but the interface  
(next_seq/write_seq) remains the same.  I don't see a specific problem  
with that unless said module introduces API changes, in which case it  
would probably have to wait until 1.7.

chris