On Fri, Sep 13, 2013 at 12:37:22PM -0400, David Kalnischkies wrote: > That was me and at the time I wrote this it was a big gain because this > code was to be backported to squeeze and I am pretty sure our chessy > l10n god would have killed me otherwise (and I value my life). ;) > And as I didn't want to be killed by the release team either it was introduced > in unstable/wheezy unchanged first (a month before in that case).
Fair enough. > That the fixme was never fixed is just what happens if you are distracted by > other stuff for a month, so that you forget it – and apt-inst code is code not > read/changed very often, so you either rediscover it by accident or you don't… It certainly looked forgotten. I wonder if any of us ever remember to grep our own code for 'FIXME' tags. I better go check my own code now... > I know its easier said than done, but the take home lesson is: > If you think something else has a bug, think again and check that > your inputs are correct before proceeding. (I fell into that trap far to > often myself, so yes it frustrating, but anger doesn't help) I didn't even expect a bug. I thought I must have something installed that was old enough to not support xz compression somehow. I just didn't think that was supposed to be the case. > And for the record, I haven't changed the string yet, but I guess it will be > "This is not a valid DEB archive, it has no data.tar member." given that > there are plans for uncompressed members and its not unlikely we would > have the same "lets backport $compressor-support" in the future. I was thinking that would be a better string. Avoid translation issues in the future if additional types are added. The error could get pretty long too. Thanks. -- Len Sorensen -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
