pykdtree_1.3.6+ds-1_source.changes ACCEPTED into unstable

[Antonio Valentino]

Dear Sebastiaan,

Il 01/11/22 08:45, Sebastiaan Couwenberg ha scritto:
> On 11/1/22 08:37, Antonio Valentino wrote:
>> Dear Sebastiaan,
>>
>> Il 01/11/22 08:23, Sebastiaan Couwenberg ha scritto:
>>  > Testing migration is blocked by autopkgtest failures:
>>  >
>>  >   https://qa.debian.org/excuses.php?package=pykdtree
>>  >
>>  > Kind Regards,
>>  >
>>  > Bas
>>
>> satpy currently fails autopkdtest mainly due to the recent update of 
>> xarray (2022.10.0).
>>
>> I have pushed to git a small patch backported form upstream but the 
>> main issue is still there.
>>
>> According to [1], updating both xarray and dask to v2022.10.0 would work.
>> An keeping bot to the previous version would work as well.
>> Updating only xarray seems to create problems to satpy.
>>
>> There are to bugreports: 1022255 [2] for both xarray and satpy, and 
>> 1022338 [3] for satpy only, but I thing that the source of the problem 
>> is the same.
>>
>> Not sure what is the best way to proceed.
>> Maybe I could submit a patch for updating dask to the latest version.
>> Do you have recommendations?
> 
> Getting dask updated seems like the best option.
> 
> Getting satpy removed from testing would resolve this issue for pykdtree.
> 
> Kind Regards,
> 
> Bas

yes, satpy is currently scheduled for auto-removal due to its dependency 
on g2clib and python-pint that seem to have release critical bugs.
By the way I do not think it will be actually removed form testing.
The issue with xarray is in unstable only.

cheers
-- 
Antonio Valentino