[Debian-med-packaging] Bug#982139: q2-quality-control: test skipped on ppc64el

[Étienne Mollier]

Control: reassign -1 vsearch
Control: retitle -1 vsearch: breaks q2-quality-control test suite on ppc64el
Control: affects -1 + q2-quality-control

Greetings,

The issue below:
> 	>       self.assertEqual(sorted(obs.index), sorted(
> 	            _dnafastaformats_to_series(self.query_seqs_short).index))
> 	E       AssertionError: Lists differ: ['1111783', '1111867', '1111886'] != ['1111750', '1111768', '1111783', '1111867', '1111886']
> 	E       
> 	E       First differing element 0:
> 	E       '1111783'
> 	E       '1111750'
> 	E       
> 	E       Second list contains 2 additional elements.
> 	E       First extra element 3:
> 	E       '1111867'
> 	E       
> 	E       - ['1111783', '1111867', '1111886']
> 	E       + ['1111750', '1111768', '1111783', '1111867', '1111886']
> 	E       ?  ++++++++++++++++++++++
> 	
looks to be actually caused by vsearch:
> 	Command: vsearch --usearch_global /tmp/autopkgtest-lxc.p_pskvmx/downtmp/autopkgtest_tmp/q2_quality_control/tests/data/query-sequences-short.fasta --id 0.97 --strand both --maxaccepts 1 --maxrejects 0 --db /tmp/autopkgtest-lxc.p_pskvmx/downtmp/autopkgtest_tmp/q2_quality_control/tests/data/bacterial-ref-sequences.fasta --threads 1 --userfields query+target+ql+qlo+qhi --userout /tmp/tmpbvtklulf
> 	
> 	----------------------------- Captured stderr call -----------------------------
[…]
> 	Matching unique query sequences: 3 of 5 (60.00%)

Extracting q2-quality-control test data, to run them against
standalone vsearch raises discrepancies in vsearch results:

  - on amd64:

	$ vsearch \
	        --usearch_global query-sequences-short.fasta
	        --id 0.97 --strand both --maxaccepts 1 \
	        --maxrejects 0 \
	        --db bacterial-ref-sequences.fasta --threads 1 \
	        --userfields query+target+ql+qlo+qhi \
	        --userout output.amd64
	[…]
	Matching unique query sequences: 5 of 5 (100.00%)
	$ cat output.amd64
	1111886	1111886	15	1	15
	1111867	1111867	15	1	15
	1111783	1111783	15	1	15
	1111768	1111768	15	1	15
	1111750	1111750	15	1	15

  - on ppc64el:

	$ vsearch \
	        --usearch_global query-sequences-short.fasta
	        --id 0.97 --strand both --maxaccepts 1 \
	        --maxrejects 0 \
	        --db bacterial-ref-sequences.fasta --threads 1 \
	        --userfields query+target+ql+qlo+qhi \
	        --userout output.ppc64el
	[…]
	Matching unique query sequences: 3 of 5 (60.00%)
	$ cat output.ppc64el
	1111886	1111886	15	1	15
	1111867	1111867	15	1	15
	1111783	1111783	15	1	15

I reassign the bug accordingly.

Kind Regards,
-- 
Étienne Mollier <emollier at emlwks999.eu>
Fingerprint:  8f91 b227 c7d6 f2b1 948c  8236 793c f67e 8f0d 11da
Sent from /dev/pts/4, please excuse my verbosity.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://alioth-lists.debian.net/pipermail/debian-med-packaging/attachments/20210705/97394fbe/attachment.sig>