Re: galois: action needed (or not?), please advise

Nilesh Patra <[email protected]>
Newsgroups gmane.linux.debian.devel.games
Message-ID <YHA0PRajZFRKO6hX@debian>
Hi Gerardo,

> Hello,
> I just checked the Package Tracker page for galois
> (https://tracker.debian.org/pkg/galois) and it shows a few entries in
> the "action needed" box. May I please ask for your advice:

Note: The things mentioned there are "good to have" fixes, and are not urgent in anyway

>1. "Fails to build during reproducibility testing": I followed the
> link but I am unable to see where it says that any build has failed.
> The light blue box on the left actually says "reproducible at
> 2021-03-25 22:48:00 UTC". I also looked at the buildinfo diff, that is
> indeed quite long, but as far as I understand, at least some
> differences are expected and result from intentionally changing some
> build parameters, e.g., the directory where files are extracted. Can
> you help me understand what the problem is (if there is indeed a
> problem) and what I should do to solve it?

Yeah, the unreproducible build report looks like a false positive.
However, to confirm it, you could run reprotest[1] locally to check that it
is indeed a false positive,
or even better you could simply enable salsa CI and check the reprotest
result.

> 2. "AppStream hints: 1 warning" (there is no Metainfo file): Tracker
> has been showing this for some time, I don't think it warrants
> immediate action, but while I'm at it, let me ask about it too. I've
> read the wiki page, I understand this is a "nice to have" feature but
> I'm not sure to which extent it is recommended practice in Debian or
> in the Games Team. Also, should it belong to the Debian packaging or
> is it better to add it upstream (I'm also the upstream maintainer so I
> can do both)?

I _think_ if possible, it'd be best if you can do it upstream, since the page here[2] states:
"write a MetaInfo file for this component and send it upstream."

> 3. "Build log checks report 1 warning": this is about CPPFLAGS and
> LDFLAGS not being set. Indeed, I don't think those are needed and I
> guess I can simply ignore this. Please let me know if you disagree.

Not really, it should _not_ be ignored. Sure, they are not essential, but they provide a hardening
flags which is good to have. Propagating hardening flags in the build is
mostly good, you might like checking out this tool blhc[3] to debug
these things further.

> 4. "Standards version of the package is outdated": I will of course
> update it in the next release.

ACK.

>In any case we are now in deep freeze so unless any of these would
> warrant an RC bug, they'll have to wait until after the Bullseye
> release.

Yes, as I mentioned none of these things is urgent. This package is good
enough and is in healthy shape for bullseye

[1]: https://wiki.debian.org/ReproducibleBuilds/Howto#Newer_method
[2]: https://appstream.debian.org/sid/main/issues/galois.html
[3]: https://tracker.debian.org/pkg/blhc

Nilesh
signature.asc (application/pgp-signature, 833 B)
-----BEGIN PGP SIGNATURE-----
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=rLp7
-----END PGP SIGNATURE-----
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.