Re: Fits files in astroquery
I also found (again in tests) a zipped FITS file. Can it still be
considered as in the preferred form for modification?
Le 04/07/2020 à 16:02, Sergio Gelato a écrit :
> What license are those FITS files under? If redistribution is
permitted I would keep them. The test suite is (usually) a valuable part
of the source package; I hate it when a test tries to download its data
from the network (frowned upon by pbuilder etc), and even more when the
data is no longer available at the network location hardcoded into the
test (e.g., bytesalad.stsci.edu). It's nicer for the source package to
be self-contained in this respect.
>
> I don't think the "binary" nature of the data is a problem. We don't
expect all image bitmaps (for logos, etc.) to be distributed in SVG
form, do we? Given the definition of "source code" as "the preferred
form for modification", I'd argue that FITS files often are to be
counted as source.
> ________________________________________
> From: Vincent Prat <vivi@debian.org>
> Sent: Saturday, July 4, 2020 15:13
> To: debian-astro@lists.debian.org
> Subject: Fits files in astroquery
>
> Hello,
>
> While packaging the latest release of astroquery, I realised that the
> tarball contains binary FITS files, for testing purpose.
> Is there any consensus on whether I should exclude them and patch the
> corresponding tests (my first guess) or just keep them?
>
> Cheers,
> Vincent
>
>
Reply to: