Learn more about these different git repos.
Other Git URLs
For a given Package I have in the Fedora Copr named python-supervisor which is in a subdir named python-supervisor in a tito project where the subdir has a specfile in it named python-supervisor.spec and a tarball named supervisor-3.0b1.tar.gz, i get a failed build and an automatically created new Package created called supervisor when I try to [re]build python-supervisor.
If I delete the new python-supervisor package and do the above again, it is once again re-created by Copr.
And on top of all of that, the [re]build of python-supervisor fails, but it fails in the newly created supervisor package, not my python-supervisor package.
The python-supervisor package.
The auto-created supervisor package.
The failed build.
I can't make heads or tails of this.
For the 3.0b1-4.fc25 release (the failed build), the name of the generated srpm was supervisor-version-release.src.rpm as per Tito configuration. The name of the package is inferred from srpm name in the current codebase. That's why also the COPR package got the name "supervisor" in the end.
I would mark this as not a bug.
So, in the end, I had to (wanted to) change the Name: of the package in the specfile from the unadorned supervisor to python-supervisor.
But even before making that change it seems at least surprising that I can create a package, even with incorrect content/configuration, that results in the implicit, automatic creation of a new package and the build actually happening in that new package.
IMHO, a build and error of some sort in the package where the build actually started would be better than that surprising (and confusing) behaviour.
This might be surprising but also is convenient at times as packages can get created automatically.
Metadata Update from @clime: - Issue assigned to clime
Glad we have figured it out.
Metadata Update from @clime: - Issue status updated to: Closed (was: Open)
Log in to comment on this ticket.