[Bioperl-l] Bio::DB::CUTG behaving weird

[Stefan Kirov]

Nailed it! SImple as compiling LWP/UserAgent with Get/Head/POST (last I 
believe not necessary, but...) enabled (disabled by default). UserAgent 
does not complain loudly these are used, but not available (LWP bug?). 
This may need to be in the docs, what do you think? All tests pass (have 
not tried the new ones, but will).
Stefan

Richard Adams wrote:

> Stefan,
> I'm looking into it although I don't have access to either of those  OSs.
> Does the test script (t/DBCUTG.t) run OK on both systems?
>
> I imagine the error is coming from  the fact that the parser is trying 
> to parse the mitochondrial table
> and I will try to sort out why that isn't parsing properly. Basically 
> all the DBCUTG module does to cope
> with a non-unique species is just select the  first in the list based 
> on regexp matching, this is usually the most likely choice.
>
>
> I'll put in some hopefully more challenging tests in the test script 
> to help find the bug. If you can run your script with
> $db->verbose(1) set (using the CVS version) and send me the output I 
> can look into it more...
>
> Best wishes,
>
> Richard
>
>