https://blogs.gentoo.org/ago/2020/07/04/gentoo-tinderbox/ Issue: media-radio/unixcw-3.6.0 installs pkg-config files with mismatched Version. Discovered on: amd64 (internal ref: ci)
Created attachment 791573 [details] build.log build log and emerge --info
I think it is a false positive here. It is right that the package is version 3.6 and libcw.pc states version 7.0.0 But the .pc file is not for the package but for the library /usr/lib/libcw.so. And if you look a little bit closer thelibrary version is libcw.so.7 and that is what the .pc file states: > Name: libcw > Description: CW (Morse code) library > Version: 7.0.0 Maybe you should adapt your check logic in the tinderbox.
(In reply to Thomas Beierlein from comment #2) > Maybe you should adapt your check logic in the tinderbox. tinderbox's job is to copy paste the error, if there is anything wrong with the notice itself, the best choice is get in touch with portage devs or who added the qa notice.
(In reply to Agostino Sarubbo from comment #3) > (In reply to Thomas Beierlein from comment #2) > > Maybe you should adapt your check logic in the tinderbox. > > tinderbox's job is to copy paste the error, if there is anything wrong with > the notice itself, the best choice is get in touch with portage devs or who > added the qa notice. I had the impression that commenting on the bug is some kind of 'getting in touch' with the devs who run the tinderbox, but I will gladly retry on irc.
See https://bugs.gentoo.org/857654#c1. This check has been dropped in portage-3.0.35 because of false positives, usually relating to SONAME versioning not corresponding to ${PV}. It may be restored at a later time in an opt-in fashion (either for users/developers to set in e.g. make.conf, or for ebuilds to enable when upstream is known to make errors (possibly could be by setting QA_PKGCONFIG_VERSION?)). Apologies for the noise until now. It was an experiment and while some legitimate bugs were found, it was too noisy and that's unfair to developers.
