Sort languages by display name, not locale code
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
language-selector (Ubuntu) |
Confirmed
|
Wishlist
|
Unassigned |
Bug Description
Binary package hint: language-selector
The lists of languages for both the Language and Text tabs seems to sort by locale code. This means that English (Nigeria) appears before English (New Zealand) and that Arabic text appears between English A and B characters.
The vast majority of users will not know the locale code and how it compares to the other visible languages. So there's theoretically no way to know how to find a desired language besides just scanning the entire list. Though I assume locale codes match many users' expectations in practice.
I recommend the use of python-pyicu to sort Unicode. It's in main & on the CD and you can see code examples by looking at how Ubiquity uses it to sort its language list.
ProblemType: Bug
DistroRelease: Ubuntu 10.04
Package: language-selector 0.5.7
ProcVersionSign
Uname: Linux 2.6.32-22-generic i686
Architecture: i386
Date: Tue May 11 16:46:42 2010
PackageArchitec
ProcEnviron:
PATH=(custom, user)
LANG=en_US.utf8
SHELL=/bin/bash
SourcePackage: language-selector
Changed in language-selector (Ubuntu): | |
importance: | Undecided → Wishlist |
status: | New → Confirmed |
*In addition*, I believe you can use python-pyicu to get translated names for the locales, based on the current locale.
You may not want that, if the primary use case is to switch from a language you don't understand to one you do (how are you navigating the dialog at all at that point though). But I thought the primary use case was to alter settings between/for a couple languages both of which you at least understand a bit. In that case, I would perhaps expect that the list of languages is translated into the current language.