jeroen <[EMAIL PROTECTED]> writes: > Replace r561 (surpress dpkg-source warnings in testsuite) by surpressing > the generation of such warnings in lintian itself, which requires latest > dpkg-dev. We're not sarge-safe anyway, so...
Hm... so, if we're depending on the latest dpkg-dev, that does open the possibility of applying the patch already in this bug (#337034) without making it work with sarge's dpkg as well. (Well, there are some things broken about that patch, but something close to it at least.) However, looking at this, now I'm not so sure that this is the right thing to do at all, particularly given the recent discussion about how lintian's output should be predictable and consistent. Pulling architectures from the dpkg files means that the output will depend on what version of dpkg you have installed. I think having these values coded in lintian is not actually the worst idea, and may be better than pulling them from dpkg files. This is particularly the case since this patch uses /usr/share/dpkg/archtable to figure out what the standard architectures are, but the dpkg changelog says that file is deprecated. Right now, I'm inclined to do something similar to this patch in terms of recognizing arbitrary combinations of os and cpu, but not pull the data from external files and instead still include it in lintian itself. Does anyone have any other ideas or input on that? -- Russ Allbery ([EMAIL PROTECTED]) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
