On Wed, 2012-04-04 at 21:17 +0200, Joachim Breitner wrote: > This makes all perfect sense to me, and what I expected, if there were > not for this entry: > > $ cat unstable/Packages_mips|grep-dctrl -FVersion 10.0.3esr-2|grep-dctrl > -FPackage iceweasel-l10n-all > Package: iceweasel-l10n-all > Source: iceweasel > Version: 10.0.3esr-2 > Installed-Size: 137 > > How can there be a iceweasel-l10n-all of version 10.0.3esr-2 if, as you > say, “iceweasel 10.0.3esr-2 doesn't build a binary package named > iceweasel-l10n-all.”
Ah, I see what's going on (I think). iceweasel 10.0.3esr-2 /did/ build an iceweasel-l10n-all package, with the same version as the source package. iceweasel-l10n also built such a binary package, with version 1:9.0+debian-1 - i.e higher than that built from iceweasel. The version of iceweasel-l10n-all in testing was therefore that built from iceweasel-l10n so when that source package was removed iceweasel-l10n-all was also removed. For unstable my suspicion would be that both sets of arch:all packages were in the archive simultaneously whilst both source packages were also in unstable; the source removal thus only dropped those with the higher version. For testing, otoh, britney would only have kept the package with the higher version so once that was removed it was gone. snapshot.d.o should be able to confirm whether my suspicion is correct, at least in terms of what unstable and testing looked like immediately before and after the removal. Apologies for the confusion. Regards, Adam -- To UNSUBSCRIBE, email to [email protected] with a subject of "unsubscribe". Trouble? Contact [email protected] Archive: http://lists.debian.org/[email protected]
