[Biopython-dev] Releasing (a beta of) Biopython 1.62 this week?

[Wibowo Arindrarto]

Hi Peter, everyone,

>> In preparation for the 1.62 release, I've made the following changes
>> to Bio.NCBIStandalone and Bio.ParserSupport:
>>
>> * Migrated the two modules under Bio.SearchIO._legacy
>> * Upgraded their PendingDeprecationWarning to BiopythonDeprecationWarning
>
> So basically you're proposing formally deprecating parsing plain
> text BLAST output (via NCBIStandalone and Bio.ParserSupport)
> but continuing to support this format via SearchIO (using a copy
> of the current parser as a private module)?
>
> This then gives you the freedom to rewrite the old text parser
> more simply (e.g. assuming only recent versions of the BLAST
> suite), which might be nice.

Yes. This seems like a sensible thing to do now.

>> I've pushed the changes to this branch:
>> https://github.com/bow/biopython/tree/bio_blast_migrate
>>
>> Tests seem to be running fine still, but now there is the awkward
>> situation where if users import Bio.NCBIStandalone and/or
>> Bio.ParserSupport directly they will be greeted with two warnings: the
>> BiopythonWarning for the modules' deprecation and the
>> BiopythonExperimentalWarning for SearchIO.
>>
>> We could suppress the SearchIO warning in Bio.NCBIStandalone and
>> Bio.ParserSupport. But before this is done, I was wondering if we have
>> a defined timeline for removing a BiopythonExperimentalWarning? (i.e.
>> if it will be removed in this release, then we could do that instead).
>
> It doesn't make sense to have a defined timetime for removing a
> BiopythonExperimentalWarning - it will be on a case by case basis.
>
> Do you think SearchIO is ready for that now (or in Biopython 1.63)?

Hmm..what I have in mind is actually as soon as we lift SearchIO's
BiopythonExperimentalWarning, we give Bio.Blast a
PendingDeprecationWarning. I think this gives users a clearer / firmer
choice, since it could be confusing to have two different modules that
handle BLAST parsing in Biopython.

As for the readiness, I think the important features that we planned
have been implemented in SearchIO. I don't have any major feature
change that I would like to implement anytime soon, too. So yes, I
think it is ready.

Best,
Bow