[Bioperl-l] failures on the stable branch

[Jason Stajich]

On Wed, 21 Apr 2004, Dave Howorth wrote:

> Jason Stajich wrote:
> > These are due to problems with Bio::DB::Query::GenBank
> >
> > setting the env variable BIOPERLDEBUG=1 will give more information.
> >
> >
> >>ok 62 # could not connect to Genbank
> >>ok 63 # could not connect to Genbank
> >>ok 64 # could not connect to Genbank
> >>ok 65 # could not connect to Genbank
> >>ok 66 # could not connect to Genbank
> >>ok 67 # could not connect to Genbank
> >>ok 68 # could not connect to Genbank
> >>ok 69 # could not connect to Genbank
> >>ok 70 # could not connect to Genbank
> >>ok 71 # could not connect to Genbank
> >>ok 72 # could not connect to Genbank
> >>ok 73 # could not connect to Genbank
> >>ok 74 # could not connect to Genbank
> >>ok 75 # could not connect to Genbank
> >>ok 76 # could not connect to Genbank
> >>ok 77 # could not connect to Genbank
> >>ok 78 # could not connect to Genbank
>
> When I set BIOPERLDEBUG=1 I see:
>
> t/DB.........................Warning: Couldn't connect to complete
> GenBank query tests!
>
> ------------- EXCEPTION  -------------
> MSG: Error from Genbank: Range operation produced no results
> STACK Bio::DB::Query::GenBank::_parse_response Bio/DB/Query/GenBank.pm:267
> STACK Bio::DB::Query::WebQuery::_run_query Bio/DB/Query/WebQuery.pm:268
> STACK Bio::DB::Query::GenBank::count Bio/DB/Query/GenBank.pm:225
> STACK (eval) t/DB.t:251
> STACK toplevel t/DB.t:245
>
> --------------------------------------
>
>
> FAILED tests 79-87
>          Failed 9/78 tests, 88.46% okay (less 1 skipped test: 68 okay,
> 87.18%)
>
>
> I'm not sure why I see different results to you.

This is what I see as well.

>
> I think what you're telling me is that I don't need to worry about this
> failure of db.t as long as I don't want to access GenBank. Is that right?
>
It works in general but for some reason the specific range query that is
in the tests is failing.  I will try and fix it at some point but I don't
really have a burning desire to do it right now.  Or maybe Lincoln will
have a chance to look - presumably the date range query syntax changes at
Entrez.


> Also, if I've understood you right, I'm not sure why the db.t test fails
> when BIOPERLDEBUG=0 when most (all?) other tests just skip the tests and
> output a warning like:
>
> t/DBCUTG.....................ok
>          20/22 skipped: tests which require remote servers - set env
> variable BIOPERLDEBUG to test
>

not important, but if you want to know why its because those tests didn't
implement a respect for this variable - no time to get into it - but
that aspect of tests has not been standardized and some modules don't
even have debugging code if you turn verbose on anyways.

> Cheers, Dave
>

--
Jason Stajich
Duke University
jason at cgt.mc.duke.edu