RE: Understanding my package entry in update_excuses
> > > icu (1.8.1-2 to 2.0-1)
> > > Maintainer: Yves Arrouye
> > > 26 days old (needed 10 days)
> > > out of date on arm: icu, icu-locales, libicu-dev (from 1.8.1-2)
> > > out of date on sparc: icu, icu-locales, libicu-dev (from 1.8.1-2)
> > > icu (source, alpha, arm, hppa, i386, ia64, m68k, mips, mipsel,
> > > s390, sparc) is buggy! (1 > 0)
> > > Not considered
> > > - Will 2.0-2 be reconsidered for testing now that it has been
> > Yes. Note that the "days old" count will start over. If 2.0-2 fixes
> > #128484, AND no other bugs are reported for 10 days, AND the package is
> > built successfully for all appropriate architectures, then it will go
> > testing.
> AND there needs to be a xerces package with a libxercesicu1 binary package
> that depends on libicu20 instead of libicu18 in unstable AND this xerxes
> package must be ready to go into testing (no open RC bugs and waited it's
> 2/5/10 days and no other dependency problems) so that icu and xerces can
> go into testing at the same time (testing otherwise refuses to upgrade icu
> because it would make libxercesicu1 uninstallable in testing).
Argh! Guillaume, is [will] that [be] the case?
On the other hand, why couldn't libicu18 be kept on testing? Isn't that why
we have libfooN, libfooN+1 etc? So that one package that depends on libfooN
doesn't need to depend on libfooN+1 immediately?