Philip Hands <phil@hands.com> (2015-08-10): > Holger Levsen <holger@layer-acht.org> writes: > ... > > And that in spring 2014 I pointed this mistake out before it was in sid and > > then it was still uploaded - as "Büsingen" is obviously the correct german > > spelling of it(!)... since then, I kinda gave up and just whined on irc. Then > > you asked me to file this bug, I did, nothing happened, so I thought I'd ping > > this bug… > > This appears to be a misunderstanding. Misunderstanding, I'm not sure; miscommunication, certainly. > Holger seems to be working on the assumption that the patch was not > applied, whereas when I look at the git log I see that Christian > committed that change on Oct 22 last year, and that the change was in > the changelog as version 1:0.60 Holger commented on the patch Christian pushed: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=765976#45 so I /think/ he saw the patch had been applied… > I've not checked when that got into an installer, but I guess it did > quite a long time ago. The version in the current installer is 1:0.66 Certainly. In any version past [2014-11-02] which is the date where 1:0.60 entered testing. Those can be checked here: https://www.debian.org/devel/debian-installer/News/2014/ https://www.debian.org/devel/debian-installer/News/2015/ > Presumably (judging from the fact that Holger's still talking about > it) the patch didn't work for some reason. Presumably, given the silent reopen both Christian and I mentioned already: https://bugs.debian.org/cgi-bin/bugreport.cgi?msg=55;bug=765976 but with no explanations whatsoever. Holger's dancing around my questions in today's mails doesn't make it easier to understand what his point actually was… Did the patch get tested at the beginning? Did it work at some point? Regressed later? I have no idea at all (other than best guessing based on when components were updated as you did further down in your mail). > OK, so I just tried it with the debian-8.1.0-amd64-netinst.iso image, > and Holger is certainly right (see screenshot attached). > > I note that there is only Berlin listed if I run this on the console: > > # grep DE /usr/share/tzsetup/tzmap > DE Europe/Berlin > > Whereas there is mention of both in /var/lib/cdebconf/templates.dat in > tzsetup/country/DE in the Choices: and Choices-C fields. I guess that's > the cause of this bug. > > In tzsetup there is a line: > > zone=$(grep "^$CC" /usr/share/tzsetup/tzmap | cut -d ' ' -f 2) > > which I suppose we were expecting to do the trick, but that's in an if's > else branch, so I guess we're actually using the then branch instead. > > tzsetup has not been touched since 2009, so it's not been introduced > recently there. > > Did anyone actually test that the patch did what we had hoped after > Christian's commit? Presumably the bug reporter who submitted the patch, and tagged it as such? But again, we would need slightly more thoughtful input from Holger. KiBi.
Attachment:
signature.asc
Description: Digital signature