[Bug: 21.5-b27] Suboptimal interaction of latin-unity and VM.

[Aidan Kehoe]

 Ar an c?igi? l? is fiche de m? na Samhain, scr?obh Stephen J. Turnbull: 

 >  > it is very unclear to me why latin-unity-sanity-check passes the text
 >  > on to the Mule detection mechanism when writing, instead of accepting
 >  > 'binary as a reasonable way of encoding control-1, latin-iso8859-1 and
 >  > ascii.
 > 
 > "detection" is the wrong word.  What this means is that you're at the
 > mercy of whatever Mule decides to do with your content when b-f-c-s is
 > set to 'binary (eg, it typically corrupts Japanese).
 >
 > In fact, if you have your b-f-c-s set to binary, it does just accept
 > 'binary as a reasonable way of encoding those sets.  (At least it did
 > the last time I checked.) 

I?ve renamed my ~/.vm file to ~/.vm-. I start XEmacs:

$ xemacs-21.5-b27 -vanilla &

In *scratch*, I type: 

(latin-unity-install) C-j 
=> (latin-unity-sanity-check)

(setq font-running-xemacs t) ;; to work around an issue W3 is seeing
=> t

I type M-x vm-visit-folder RET /tmp/aidan/spam RET

I make a change, press S , and I get the coding system conflict message. It
says, again, that:

    ?All preapproved coding systems (buffer-default==binary
    preferred==utf-16-little-endian-bom) fail to appropriately encode some
    of the characters present in the buffer.?

This time it lists: 

    iso-8859-1 iso-8859-2 iso-8859-3 iso-8859-4 iso-8859-9 iso-8859-15 utf-8
iso-2022-7 ctext escape-quoted

as recommended because they can encode any character in the buffer. 
The folder in question is available at http://www.parhasard.net/spam.bz2 .

-- 
Santa Maradona, priez pour moi!