Re: mutt to show hi-bit chars once & for all?
On Saturday, September 24, 2005 at 22:00:24 -0400, Russell Hoover wrote:
> On Sat 09/24/05 at 03:35 PM +0200, Alain Bench <messtic@xxxxxxxxx> wrote:
>> A bunch of aeiouy AEIOUY umlaut and a beta: äëïöüÿ ÄËÏÖÜ?? ß
> Of the upper case letters, I see the "A" with an umlaut correctly,
> then two square boxes, then the "O" with an umlaut correctly, then the
> "U" with an umlaut correctly. Then two question-marks
Everything you see and quote is normal, but:
- EI umlaut seen as boxes ==> font problem
- Y umlaut seen and quoted as question marks ==> this character I sent
in UTF-8 doesn't exist in Latin-1 charset: Mutt has ?-masked it as
unconvertable. That's "normal".
> Same thing when I try to produce -- in Vim -- the same letters: äëïöüÿ
> -- lower case, no problems. Upper case: ÄËÏÖÜ: -- not sure what *you*
> see here, but I can't seem to produce, in Vim, the capitals "E", "I"
> or "Y" with their umlauts.
I see "aeiouy" OK, and "AEIOU:" OK but the colon in place of "Y".
That seems nominal, knowing the "Y" is impossible unless you change your
terminal's charset (to perhaps Latin-9 or UTF-8).
> my mutt was compiled with S-Lang version 1.4.6, and I know that the
> new version 2.0.4 has "native support for Unicode via UTF-8 throughout
> the library." Using the new S-Lang may help things.
No: It would perhaps influence if you were using an UTF-8 terminal,
but not at all in your current setup.
On Sunday, September 25, 2005 at 2:19:28 -0400, Russell Hoover wrote:
> I now have set (and exported) _only_ LANG, as per your advice. I set
> it to "en_US.ISO8859-1".
>| locale
>| LANG="en_US.ISO8859-1"
>| LC_CTYPE="en_US.ISO8859-1"
>| LC_COLLATE="C"
>| LC_TIME="C"
>| LC_NUMERIC="C"
>| LC_MONETARY="C"
>| LC_MESSAGES="en_US.ISO8859-1"
>| LC_ALL=""
Fine: That seems the most correct setting, for a Latin-1 terminal on
NetBSD 2.0.
On Sunday, September 25, 2005 at 5:54:39 -0400, Russell Hoover wrote:
> I cut-and-pasted into a vim text-file the following chart from
> <http://www.in-ulm.de/~mascheck/locale/> :
The copy/paste was munged, with most characters OK, some lacking (as
0xAD), some apparently transcoded to something wrong (as 0xD0), and some
transliterated to something good but suboptimal (as 0xBC to "1 4"). Note
this last BC is what you see as pi, but what is a 1/4 symbol in Latin-1.
> my term program is MacSSh version 2.1fc3 -- I'm still running MacOS
> 8.1, believe it or not, on a vintage-1998 Mac beige G3
Last time we heard about MacSSH here, in March 2003 thread "Mac ssh
to Linux and charset (was: Mutt and UTF-8)", it was using really a
CP-437 charset, and had "translation" setting allegedly not used because
suboptimal: It was more a character reordering than a charset change.
Now note an interesting fact: In CP-437, uppercase umlaut AOU do
exist, but EIY do not. And there is a Greek small letter pi.
> I changed the fontsize (something that I normally never, ever do) of
> the Monaco font that i use from 12-point to 14-point, and *presto* --
> suddenly every single character in the chart displayed correctly.
And now what do you see doing:
| $ print "\xBC \xA6"
| ¼ ¦
Summary so far: Mutt and NetBSD settings are perfect for Latin-1.
But MacSSH or Monaco 12 do not properly agree. You have 2 main options:
-1) Keep MacSSH Monaco 14, and validate it's clean Latin-1.
-2) Take another terminal, preferably doing UTF-8, and change only to
LANG=en_US.UTF-8 (nothing to change in muttrc).
Note the way to go is (2) if you want to "become as
Unicode-compliant as [you] can". Better yet do the 2 ways with a
conditional setup depending on the used terminal.
Bye! Alain.
--
Mutt muttrc tip to send mails in best adapted first necessary and sufficient
charset (version for Western Latin-1/Latin-9/CP-850/CP-1252 terminal users):
set send_charset="us-ascii:iso-8859-1:iso-8859-15:windows-1252:utf-8"