Re: [tlug] CJK Mixed in a Letter. Missing bdf font FOUND:hanglm24.bdf

[David Riggs <dariggs@example.com>]

Stephen wrote

>Your analysis is correct.  This is a long-standing design bug in Mule,
>and one of the primary applications of Han unification.  (The primary
>reason is the code space limitation, of course.)
>
>    David> But alas, it utter chokes when it cannot get a kanji. In my
>    David> case its the "setsu" kanji of "setsumei", which in the big5
>    David> form has the right hand "rabit ears" over the kuchi point
>    David> the opposite way from modern Japanese.
>
>    David> Emacs complains that "bdf file hanglm24.bdf does not exist"
>    David> and gives up the printing. This is a Korean font, which is
>    David> weird, since there is no Korean here.
>
>The only thing I can suggest offhand is to use M-x search-forward
>rather than isearch (which is unlikely to be terribly pleased with
>switching IMs on the fly, although it may work OK in GNU Emacs, pretty
>unlikely in XEmacs).  Then use C-u C-\ (or C-m C-x C-\) to change to a
>Chinese input method in the minibuffer.

---------
Thanks-- I will give some more thought about how to deal with this on my
own now that I realize its real and not just my latest screw up.


But my immediate problem is how to print the stuff out! Thanks for the
suggestion, Jim, about converting from pcf fonts, but I finally figured
out where the bdf fonts are kept by following a Emacs for the dark side
FAQ:

Found ftp://ftp.gnu.org/gnu/intlfonts/

downloaded the korean set, untared and copied hanglg16.bdf hanglm16.bdf
hanglm24.bdf to my /usr/share/emacs/fonts/bdf.

It works. I guess those fonts got left out somehow from the
international package. I also wonder if this is a Debian packaging problem.

The other, broader encoding mess is really a canonical problem: there
needs to be an agreed upon standard for a single code point for a kanji.
Its a constant hassel for folks like me who work in more than one
"kanji" system. Unicode alas does not solve this one, as there are
multiple forms of ji's that are really the same-- the two kinds of setsu
(as in setsumei) are really just font differences and should not be code
point differences, as far as I can see (which is not very far, of course).

Thanks for the input folks. I revel in my printout capabilities while I
work on the kanji encoding mess.

David Riggs