how to save œ other than utf8

[David Kastrup]

Uwe Brauer <oub at mat.ucm.es> writes:

>>>>>> "Aidan" == Aidan Kehoe <kehoea at parhasard.net> writes:
>
>    > Œ is not in ISO 8859-9 (nor is œ). It is in ISO 8859-15, but
>    > unless you’re using recent 21.5, you’ll need to make sure that
>    > the character being input gives the following output in the
>    > scratch buffer:
>
> That works, but the 15 coding gives me 
>  ½

Sigh.  latin-9 (aka iso-8859-15 for whatever reason) has œ at code point
0xbd, and that is how you save the file.  XEmacs is completely correct
here.

Then you read the file in, interpreting it as latin-1 (since XEmacs has
no way to guess that this file is supposed to be latin-9).  And latin-1
has ½ at that code point.

If you, for example, place
\usepackage[inputenc]{latin9}
at the top of a LaTeX file, TeX will know to interpret this character as
œ in spite of your editor showing it as ½ unless told about the right
encoding (newer versions of Emacs steal the encoding info from the above
LaTeX line, but that is probably not fair).
 
>    > for it to be saved correctly using the iso-8859-15 coding
>    > system. Other encodings (other Mule character sets) won’t work,
>    > again, unless you’re using recent XEmacs 21.5. It may be that
>    > latin-unity will work too, even on 21.4.
>
> Ok, but how do I use latin-unity?  Well it seems that
> latin-unity-install did the thing the coding was then saved
> correctly. But alas it is saved as UTF-8!
>
>
> So how can I save it as iso-8859-15?

You already did, successfully so.  Your question should rather be how to
tell XEmacs to _read_ it as iso-8859-15.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum