As http://bonsai.mozilla.org/cvsblame.cgi?file=mozilla/xpfe/global/resources/locale/en-US/brand.dtd&rev=&root=/cvsroot states, the version number in brand.dtd wasn't updated last night for 1.2b, the same is true for region.dtd and I guess that for some component even the localeVersion string in chrome.rdf wasn't updated correctly. If version strings changes are performed that incomplete and there's no time to check for it because they're flipped only very very very short times before the release, we won't ever be able to ship langauge packs again and we could drop locale support completely. I'll search for other version numbers that might be wrong as soon as I've pulled and build my tree.

BTW, this is the same thing that bug 168923 was about, and it makes bug 168066 a WONTFIX in my eyes :((

reassigning to self, making blocker as it is a blocker for L10n people. If bug 169074 would be solved, we could avoid such problems completely.

OK, this patch fixes all problem I could find. First, there are those two .dtd files, which still have 1.2a Then, inspector never gets updated to newer localeVersions (well, it was once on a branch - IIRC that was 1.1) and still has 0.9.4 Then, there is a built but empty mac donloadmanager directory which still has 1.0.0 And, CaScadeS looks a bit messed up with localeVersion strings entirely (having the string in locale/ but missing it in content/) and has 0.1.7 which looks imported from the previous mozdev project For L10n people, it's important that all components share the same version string, or it will get much harder to even provide correct and working L10n for those components.

Comment on attachment 103182 [details] [diff] [review] patch updating all known non-fitting locale version strings r=jbetak

leaf, tao: we really need to get the release scripts fixed. We miss brand.dtd and region.dtd in *every* release. You might have seen an email about it two weeks ago. I caught the four contents.rdf files while working on bug 137331 and wanted to file a follow-up bug after wrapping up bug 137331. Kudos to kairo for catching this!

oops should read bug 130331

leaf, are we using flipLocale.sh and strsubs.pl (see bug 154927)? If so, could we make them part of the Mozilla tree? :-)

asked bz for sr= (he sr'ed bug 168923 as well). Asa, can you a= this change? (If we get sr before, I'll mail you anyway...)

Comment on attachment 103182 [details] [diff] [review] patch updating all known non-fitting locale version strings sr=bzbarsky

*** Bug 175396 has been marked as a duplicate of this bug. ***

Comment on attachment 103182 [details] [diff] [review] patch updating all known non-fitting locale version strings a=asa for checkin to 1.2 (on behalf of drivers).

Checking in xpfe/global/resources/locale/en-US/brand.dtd; /cvsroot/mozilla/xpfe/global/resources/locale/en-US/brand.dtd,v <-- brand.dtd new revision: 1.10; previous revision: 1.9 done Checking in xpfe/global/resources/locale/en-US/region.dtd; /cvsroot/mozilla/xpfe/global/resources/locale/en-US/region.dtd,v <-- region.dtd new revision: 1.13; previous revision: 1.12 done Checking in extensions/inspector/resources/content/contents.rdf; /cvsroot/mozilla/extensions/inspector/resources/content/contents.rdf,v <-- contents.rdf new revision: 1.6; previous revision: 1.5 done Checking in extensions/inspector/resources/locale/en-US/contents.rdf; /cvsroot/mozilla/extensions/inspector/resources/locale/en-US/contents.rdf,v <-- contents.rdf new revision: 1.3; previous revision: 1.2 done Checking in xpfe/components/download-manager/resources/mac/contents.rdf; /cvsroot/mozilla/xpfe/components/download-manager/resources/mac/contents.rdf,v <-- contents.rdf new revision: 1.3; previous revision: 1.2 done Checking in extensions/editor/cascades/resources/locale/en-US/contents.rdf; /cvsroot/mozilla/extensions/editor/cascades/resources/locale/en-US/contents.rdf,v <-- contents.rdf new revision: 1.2; previous revision: 1.1 done Hopefully FIXED. Perhaps I'll file a bug with a patch to get _really all_ locale version strings updated for 1.2final correctly. I'm not sure we should trust leaf's script again...