[Top][All Lists]

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

Re: Bug#400453: Makefile misses to clean .gmo-files in the clean target,

From: Bruno Haible
Subject: Re: Bug#400453: Makefile misses to clean .gmo-files in the clean target, so they get never be updated after updating an .po file (fwd)
Date: Tue, 28 Nov 2006 19:30:47 +0100
User-agent: KMail/1.9.1

Daniel Leidert wrote:
> But there are also some practical problems resulting from this behaviour
> change (as far as I can see):
> - package distributors now always must patch the .po file _and_ update
> the.gmo file (old: they only had to update the .po file in many
> (probably most?) cases) to fix i18n bugs

Package distributors who include patches to an xx.po file also have to
include a "make xx.gmo" or "make update-gmo" command in their build sequence.

> - for at least Debian packages that means: package maintainers always
> _must_ apply the resulting changes (from a i18n bug-fix) to .gmo via
> patch system or dpkg-source will definitely fail with "unrepresentable
> changes to source", because source ships binary files (.gmo) that may
> not be altered directly - so adding update-gmo to any target in
> debian/rules will make dpkg-source fail at least at the second run (if
> run twice) - and because patching .po isn't longer enough, the above
> step is always necessary

This is Debian specific; I don't understand this.

> These issues will massively raise up as wider gettext >= 0.15 is used.

Huh? Did this change recently? AFAIK, this logic of po/Makefile rules is in
place since at least gettext 0.10.35, i.e. for more than 8 years.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]