Re: locales and coding systems

[brownh@hartford-hwp.com (Haines Brown)]

Vineet,

Well perhaps some progress, perhaps not.

> * Haines Brown (brownh@hartford-hwp.com) [031212 14:02]:
> > Logged and back in and went to save my test file. The test
> > failed. That is, I'm still being prompted for the coding system, and
> > utf-16-le is offered as the default.
> >=20
> > The command $ locales tells me that locales is installed (status is
> > ii).=20
> >=20
> > However, the $ locale command still only returns LANG=3DPOSTFIX for
> > LANG. It still lacks a line such as LANG=3Den-US.utf-8.  The coding
> > system I selected with locales configuration is not reported.
> 
> Try it after puttin something like this in your .bashrc:
> 
> LANG=3Den_US.UTF-8
> export LANG

Well, I put these lines into /etc/bash.bashrc. This did in fact put
LANG into bash's environment, and when I ran $ locales, it does
report LANG=en_US.UTF-8 rather than LANG=POSIX. However, that
information is not being used by emacs when it tries to find the
default coding system (when I try to save a file with accendted
characters, I'm told there's no default language and utf-16-le is
offered as the option by default).

Since I managed to install locales, I noticed that now I have a
/etc/locale.gen file, and when I look into it, I find that apparently
the coding system is supposed to be written there when I run
dpkg-reconfigure locales. For some reason nothing was written when I
configured locales initially, but now when I run dpkg-reconfigure, I
find added to it the line: en_US.UTF-8 UTF-8.

I relogged in again and now have:

$ locale
LANG=en_US.UTF-8
LC_CTYPE="en_US.UTF-8"
LC_NUMERIC="en_US.UTF-8"
LC_TIME="en_US.UTF-8"
LC_COLLATE="en_US.UTF-8"
LC_MONETARY="en_US.UTF-8"
LC_MESSAGES="en_US.UTF-8"
LC_PAPER="en_US.UTF-8"
LC_NAME="en_US.UTF-8"
LC_ADDRESS="en_US.UTF-8"
LC_TELEPHONE="en_US.UTF-8"
LC_MEASUREMENT="en_US.UTF-8"
LC_IDENTIFICATION="en_US.UTF-8"
LC_ALL=

Looks good, but then I tested to see if emacs would now see
UTF-8 as the coding system in use. It did not ;-(. I've slowly gotten
to the point where it seems that emacs should pick up the coding
system, but it fails to do so.

Do you happen to know which of these variable names is likely to be
the one used by emacs? At this point, I may need to approach the
problem from the emacs side, but my sense is that the problem is not
there.   

Haines Brown