Bug#824048: Any change in blastp? (Was: Bug#824048: python-biopython: FTBFS: AssertionError: 10 != 1)

2016-05-19 Thread Peter Cock 
Great - I'm glad that made sense, my email had more typos than usual
as I was in a hurry to leave the office earlier.

Thanks,

Peter

On Thu, May 19, 2016 at 4:20 PM, Andreas Tille  wrote:
> Hi Peter,
>
> I've just uploaded a package featuring the patch.
>
> Thanks a lot for the quick and helpful response
>
>   Andreas.
>
> On Thu, May 19, 2016 at 03:31:59PM +0100, Peter Cock wrote:
>> RE: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=824048
>>
>> I've committed this as "fix",
>>
>> https://github.com/biopython/biopython/commit/cb5b8f7cf16dfa9aada8d3c71ab8d588ebf0693f
>>
>> This was a sanity test which failed due to a change in the NCBI output.
>>
>> We don't currently try to parse the plain text output in this test
>> (the NCBI discourage parsing the plain text output, and we'd like
>> to drop our plain text BLAST parser anyway).
>>
>> If it is easier, you could just comment out or remote this for
>> the Biopython 1.66 Debian package?
>>
>> Thanks for letting us know,
>>
>> Peter
>>
>
> --
> http://fam-tille.de

Bug#824048: Any change in blastp? (Was: Bug#824048: python-biopython: FTBFS: AssertionError: 10 != 1)

2016-05-19 Thread Andreas Tille 
Hi Peter,

I've just uploaded a package featuring the patch.

Thanks a lot for the quick and helpful response

  Andreas.

On Thu, May 19, 2016 at 03:31:59PM +0100, Peter Cock wrote:
> RE: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=824048
> 
> I've committed this as "fix",
> 
> https://github.com/biopython/biopython/commit/cb5b8f7cf16dfa9aada8d3c71ab8d588ebf0693f
> 
> This was a sanity test which failed due to a change in the NCBI output.
> 
> We don't currently try to parse the plain text output in this test
> (the NCBI discourage parsing the plain text output, and we'd like
> to drop our plain text BLAST parser anyway).
> 
> If it is easier, you could just comment out or remote this for
> the Biopython 1.66 Debian package?
> 
> Thanks for letting us know,
> 
> Peter
> 

-- 
http://fam-tille.de

Bug#824048: Any change in blastp? (Was: Bug#824048: python-biopython: FTBFS: AssertionError: 10 != 1)

2016-05-19 Thread Peter Cock 
RE: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=824048

I've committed this as "fix",

https://github.com/biopython/biopython/commit/cb5b8f7cf16dfa9aada8d3c71ab8d588ebf0693f

This was a sanity test which failed due to a change in the NCBI output.

We don't currently try to parse the plain text output in this test
(the NCBI discourage parsing the plain text output, and we'd like
to drop our plain text BLAST parser anyway).

If it is easier, you could just comment out or remote this for
the Biopython 1.66 Debian package?

Thanks for letting us know,

Peter

Bug#824048: Any change in blastp? (Was: Bug#824048: python-biopython: FTBFS: AssertionError: 10 != 1)

2016-05-19 Thread Andreas Tille 
Hi Peter

On Thu, May 19, 2016 at 12:32:17PM +0100, Peter Cock wrote:
> Thanks Andreas,

Thanks for your very quick response.
 
> Ah, https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=824048
> 
> It looks like yet another minor change to the plain text human
> readable output with the latest NCBI BLAST+ v2.3.0 release is to
> blame,
> 
> $ ~/ncbi_blast/2.2.31+/blastp -query Fasta/rose.pro -evalue 1 -subject
> GenBank/NC_005816.faa | grep "Query= "
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> $ ~/ncbi_blast/2.3.0+/blastp -query Fasta/rose.pro -evalue 1 -subject
> GenBank/NC_005816.faa | grep "Query= "
> Query= gi|4959044|gb|AAD34209.1|AF069992_1 LIM domain interacting RING
> 
> The Biopython test was expecting 10 lines with this pattern, but there
> is now only 1. This may or may not break the parser - I've not yet
> checked:
> 
> https://github.com/biopython/biopython/issues/827

This sounds like a very probable explanation for the effect.  It would
be great if you could come up with a patch (or a new release) so I could
update the Debian package.
 
Thanks a lot for the perfect cooperation

  Andreas.

-- 
http://fam-tille.de

Bug#824048: Any change in blastp? (Was: Bug#824048: python-biopython: FTBFS: AssertionError: 10 != 1)

2016-05-19 Thread Andreas Tille 
Hi Peter,

since some time the blastp test fails when building the Debian BioPython
package.  I've got  no answer what might be wrong here.  Do you have any
idea which is better than deactivating this specific test for the monent
which would be the only poor idea I'd have?

Please let me know if you need more information.

Kind regards

   Andreas.

On Wed, May 11, 2016 at 10:08:49PM +0200, Andreas Tille wrote:
> On Wed, May 11, 2016 at 05:35:27PM +0100, Chris Lamb wrote:
> > ...
> >   Bio.PDB.Selection docstring test ... ok
> >   ==
> >   FAIL: test_blastp (test_NCBI_BLAST_tools.Pairwise)
> >   Pairwise BLASTP search
> >   --
> >   Traceback (most recent call last):
> > File 
> > "/home/lamby/temp/cdt.20160511172943.3m3eRDBdi6.python-biopython/python-biopython-1.66+dfsg/.pybuild/pythonX.Y_2.7/build/Tests/test_NCBI_BLAST_tools.py",
> >  line 103, in test_blastp
> >   self.assertEqual(10, stdoutdata.count("Query= "))
> >   AssertionError: 10 != 1
> >   
> >   --
> >   Ran 228 tests in 101.627 seconds
> >   
> >   FAILED (failures = 1)
> 
> Does any body know about a recent change in blastp?
> 
> Kind regards
> 
>Andreas.
> 
> -- 
> http://fam-tille.de
> 
> ___
> Debian-med-packaging mailing list
> debian-med-packag...@lists.alioth.debian.org
> http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-packaging
> 

-- 
http://fam-tille.de

Bug#824048: Any change in blastp? (Was: Bug#824048: python-biopython: FTBFS: AssertionError: 10 != 1)

2016-05-11 Thread Andreas Tille 
On Wed, May 11, 2016 at 05:35:27PM +0100, Chris Lamb wrote:
> ...
>   Bio.PDB.Selection docstring test ... ok
>   ==
>   FAIL: test_blastp (test_NCBI_BLAST_tools.Pairwise)
>   Pairwise BLASTP search
>   --
>   Traceback (most recent call last):
> File 
> "/home/lamby/temp/cdt.20160511172943.3m3eRDBdi6.python-biopython/python-biopython-1.66+dfsg/.pybuild/pythonX.Y_2.7/build/Tests/test_NCBI_BLAST_tools.py",
>  line 103, in test_blastp
>   self.assertEqual(10, stdoutdata.count("Query= "))
>   AssertionError: 10 != 1
>   
>   --
>   Ran 228 tests in 101.627 seconds
>   
>   FAILED (failures = 1)

Does any body know about a recent change in blastp?

Kind regards

   Andreas.

-- 
http://fam-tille.de