[Bioperl-l] bioperl-db tests
Shawn
shawnh@fugu-sg.org
21 Aug 2002 18:35:14 +0800
Hi Hilmar,
thanks for the reply.
What I want to do is to have an option of caching sequences when
fetching. The default behavior is still lazy. This is because for
certain pipelines, we fetch quite a number of sequences at one time, and
this is done before sending the inputs to the analysis. So what we want
to do is really to fetch the actual sequences when we do the fetch and
not til we call ->seq thereby allowing us to release the db connection
and avoiding those too many dbconnection errors.
so it just involves adding an option. Maybe I should wait til your
changes are done then integrate them ?
shawn
On Wed, 2002-08-21 at 02:44, Hilmar Lapp wrote:
> You need to get and install the bioperl main trunk latest revision in order to work with the live bioperl-db code. The errors you're seeing are all related to the map database except remove.t. I changed the config files and some other things; look into t/DBHarness.conf.example for a documented example, then copy that file to both t/DBHarness.conf and t/DBHarness.markerdb.conf and edit accordingly. You can also ignore the markerdb-related errors if you don't care about that.
>
> Getting the config files right should make the get_MapAdaptor() error go away.
>
> As for the exception in remove.t, the new location parser doesn't recognize the replace(...) operator, as it violates the official Genbank FT definition. The replace(...) operator was an early version of what is meanwhile a 'variation' feature tag with a standard location. Unless replace() operators still occur in current genbank downloads, I'd prefer to remove this test. Having to deal with replace() in location strings would be /really/ ugly because that operator folds location and annotation into one string.
>
> Iùb=
>
> Can you give me an idea what you'll be changing? I'm currently orking on a reorganization and partial rewrite of the entire Bio::DB::* code layout, but I didn't want to commit before all tests pass.
>
> -hilmar
>
> > -----Original Message-----
> > From: Shawn [mailto:shawnh@fugu-sg.org]
> > Sent: Tuesday, August 20, 2002 7:01 PM
> > To: bioperl
> > Subject: [Bioperl-l] bioperl-db tests
> >
> >
> >
> > Hi,
> > I'm trying to find out which version of bioperl works
> > best with the
> > live version of bioperl-db. I'm trying to get make test to
> > work. So far
> > I've tried bioperl-live, bioperl-1.0.2 and bioperl-0-7-2. So far the
> > main trunk seems to give me the least errors.
> >
> > Failed Test Status Wstat Total Fail Failed List of Failed
> > --------------------------------------------------------------
> > ------------------
> > t/ePCRdb.t 255 65280 3 1 33.33% 3
> > t/map.t 255 65280 6 4 66.67% 3-6
> > t/marker.t 255 65280 19 15 78.95% 5-19
> > t/remove.t 20 1 5.00% 9
> >
> > Two main errors
> > one involves get_MapAdaptor not found in Bio::DB::BioSQL::DBAdaptor
> > while the other is a little less obvious to me :
> >
> > ------------- EXCEPTION -------------
> > MSG: operator "replace" unrecognized by parser
> > STACK Bio::Factory::FTLocationFactory::from_string
> > /usr/users/shawnh//src/bioperl-live//Bio/Factory/FTLocationFac
> > tory.pm:159
> > STACK (eval)
> > /usr/users/shawnh//src/bioperl-live//Bio/SeqIO/FTHelper.pm:124
> > STACK Bio::SeqIO::FTHelper::_generic_seqfeature
> > /usr/users/shawnh//src/bioperl-live//Bio/SeqIO/FTHelper.pm:123
> > STACK Bio::SeqIO::genbank::next_seq
> > /usr/users/shawnh//src/bioperl-live//Bio/SeqIO/genbank.pm:298
> > STACK BioSQLBase::store_seq t/BioSQLBase.pm:62
> > STACK (eval) t/remove.t:31
> > STACK toplevel t/remove.t:30
> >
> > basically I want to make sure the tests pass before I make
> > any commits.
> >
> > thanks,
> >
> > shawn
> >
> >
> >
> >
> > _______________________________________________
> > Bioperl-l mailing list
> > Bioperl-l@bioperl.org
> > http://bioperl.org/mailman/listinfo/bioperl-l
> >
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l
>