[Pkg-rust-maintainers] Bug#958547: Bug#958547: closing 958547 ( rust-curl-sys autopkgtest failure, looks for crate in the wrong place. )

[Ximin Luo]

peter green:
> On 23/04/2020 17:07, Ximin Luo wrote:
>> close 958547
>> thanks
>>
>> See https://alioth-lists.debian.net/pipermail/pkg-rust-maintainers/2020-January/009512.html
>>
>> Just be patient and wait a few days, the issue usually resolves itself.
> 
> I have read that thread and I am aware of the issue that testing migration tests sometimes install the wrong version of the package and that when a version mismatch happens it results in errors that look at first glance similar to this one. In the case of such version mismatch errors I wait a bit and if waiting doesn't seem to be helping follow up with the release team and/or debci team.
> 
> But I am pretty sure that is not what is going on here for a couple of reasons.
> 
> 1. I checked the test log and there was no version mismatch to be seen.
> 2. The failure is also happening in plain unstable tests. My understanding is that the version mismatch issue is a quirk of the testing migration tests.
> 
> It seems that the issue is that the upstream version is "0.4.30+curl-7.69.1", the Debian packaging seems to strip this down to 0.4.30 for the package versions and for the name of the directory in /usr/share/ but the autpkgtest uses the full version when looking for the source to test.
> 

You are right, I've fixed this in debcargo 293fb88f2156d0db6262349aa4b1c4d3a3b1186a. Affected packages will be fixed after the next release & ./update.

> Looking at previous test logs I also have a nasty feeling that if/when this is fixed the tests will fail for other reasons.

Why "nasty"? This, and 95% of recent similar bugs you and other people have filed, affects literally 0 users[*] and I guarantee you nobody else in the world cares about them existing. There are far worse things like the coronavirus to be worried about. It's not a big deal, at some point they will all be fixed, play a video game, read a book, play the guitar in the meantime, or something.

X

[*] obviously one can make the argument that indirectly users care, because they slow the migration progress to Debian Testing, but *we've set up these processes ourselves* so that is a circular argument. The fact that these processes are catching issues that no *users* care about is a bug about the process.

-- 
GPG: ed25519/56034877E1F87C35
GPG: rsa4096/1318EFAC5FBBDBCE
https://github.com/infinity0/pubkeys.git