[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: reporting live-helper version



Daniel Baumann escreveu isso aí:
> Antonio Terceiro wrote:
> > The problem is that the version is declared both in functions/common.sh
> > and debian/changelog, so it would be nice if they were synchronized.
> 
> i'm usually doing this, but last time i forgot :/
> 
> > This could be done at package build-time, for example, using
> > dpkg-parsechangelog and putting the right version in functions/common.sh
> > with sed. If it's ok, I'll write the patch.
> 
> i'm not so fond of that, because i'd like live-helper to be the same
> *independent* of if it being build as a debian package or just a tarball.

Maybe the debian package build could fail if the versions in these files
differ, so that you don't forget again. :)

http://git.debian.org/?p=users/terceiro-guest/live-helper.git;a=commitdiff;h=c3f940a8bfeb5930ba8bdeda66eb600b7271981b

> > Other problem: in the --dump option of lh_config, dpkg-query is used for
> > getting the version. Isn't that redundant since we already have the
> > version explictly declared in the source?
> 
> current behaviour is ok. because.. if we are on a debian system, we want
> to have the exact version number of the package, which is more accurate
> and can be different (backports.org, or if people build custom packages,
> or are using git snapshots) than what is in ${VERSION}.
> 
> on the other hand, ${VERSION} is required when we can't relay on dpkg
> (because someone is building on a non-debian system).

What I noted is that I had live-helper 1.0~a48-1 removed from my system,
and the live-helper from git was reporting 1.0~a48-1 instead of the
actual current version (the one in functions/common.sh)

Anyway, the commit above already depends on a i18n fix for the very line
which uses this version from dpkg-query.

-- 
Antonio Terceiro <terceiro@softwarelivre.org>
http://people.softwarelivre.org/~terceiro/
GnuPG ID: 0F9CB28F


Attachment: signature.asc
Description: Digital signature


Reply to: