tools-iuc: Failing tests on release 20.09
https://gist.github.com/mvdbeek/dfad7627d8dc52104fff24a838484cb1 has the failing tests for the IUC repository (i’ve removed the ones that fail on 20.05 as well, that’s why the numbers don’t add up). We do have 16 additional failing tests as compared to last weeks tests against 20.05.
Most of it is minor, since we now verify the order within which elements appear in a collection
Failed to find identifier 'FL-E13-5_chr1_chr1_19093103_19093103_Tfap2d.txt' of test collection ['FL-E13-5_chr1_chr1_14300280_14300280_Eya1.txt', 'FL-E13-5_chr1_chr1_4487435_4487435_Sox17.txt', 'FL-E13-5_chr1_chr1_19093103_19093103_Tfap2d.txt', 'MB-E10-5_chr1_chr1_19093103_19093103_Tfap2d.txt', 'MB-E10-5_chr1_chr1_14300280_14300280_Eya1.txt', 'MB-E10-5_chr1_chr1_4487435_4487435_Sox17.txt'] in the tool generated collection elements ['FL-E13-5_chr1_chr1_14300280_14300280_Eya1.txt', 'FL-E13-5_chr1_chr1_19093103_19093103_Tfap2d.txt', 'FL-E13-5_chr1_chr1_4487435_4487435_Sox17.txt', 'MB-E10-5_chr1_chr1_14300280_14300280_Eya1.txt', 'MB-E10-5_chr1_chr1_19093103_19093103_Tfap2d.txt', 'MB-E10-5_chr1_chr1_4487435_4487435_Sox17.txt'] (at the correct position)
for instance means that the order of the elements in the collection are not what is specified in the tests
The scanpy_inspect and anndata_import tests however might be real bugs. For scanpy it appears that one of the input datasets couldn’t be uploaded, so the test couldn’t run. For anndata_import the job is in error. These are things we should check out.
About this issue
- Original URL
- State: closed
- Created 4 years ago
- Comments: 34 (32 by maintainers)
/run-all-tool-tests
/run-all-tool-tests
/run-all-tool-tests
My guess would be that you got the right combination of history id and and dataset id (which are just encoded integers) so that the test differences are small enough for the test to pass
Still, how should this affect a tool run that runs in a container?
On a side note I’m wirking on a fix … carefully adapting
lines_diff