[Biopython-dev] [Biopython] Google Summer of Code Project: SearchIO in Biopython

[Wibowo Arindrarto]

On Mon, Apr 30, 2012 at 12:57, Peter Cock <p.j.a.cock at googlemail.com> wrote:
> On Mon, Apr 30, 2012 at 11:08 AM, Wibowo Arindrarto
> <w.arindrarto at gmail.com> wrote:
>>
>> I'm thinking of using the Search object as the object returned by
>> SearchIO.parse or SearchIO.read. That way, we can store attributes
>> common to the different search queries in it. For example:
>>
>>>>> search  = SearchIO.parse('blast_result.xml', 'blast-xml')
>>>>> search.format
>> 'blast-xml'
>>>>> search.algorithm
>> 'blastx'
>>>>> search.version
>> '2.2.26+'
>>>>> search.database
>> 'refseq_protein'
>>>>> search.results
>> <generator object results at ....>
>>
>> And iteration over the results would be done like this (for example):
>>>>> for result in search.results:
>> ... print result.query, print len(result)
>>
>> Additionaly, we can also define __iter__ and next for Search so we can
>> just do the following:
>>>>> for result in search:
>> ... print result.query, print len(result)
>>
>> What do you think?
>
> I think you'll get in a mess with multiple iterators all sharing the
> same handle and competing over using it - but maybe I'm not
> grasping what you have in mind.
>
> Initially keep it simple: The primary public API would be
>
> for result in Bio.SearchIO.parse(...):
>     print result.query, print len(result)
>
> where each iteration gives a complete result set for one query.
>
> Peter
>
> P.S. With SearchIO subject to name space discussions ;)

Hmm..Ok. I'll stick to the simpler API for the initial implementation
~ see if later it's feasible to add more details :) (and perhaps
change the namespace too, as touched earlier).

Bow