Hmm, I just wrote my email before fetching the emails that had been exchanged over the weekend. Not a good idea. On Mon, Dec 02, 2002 at 06:40:30PM +0100, Gour wrote:
Hans Hagen (pragma@wxs.nl) wrote:
a) c4 81 -> amacron b) 0101 -> amacron
so, c4 is the trigger, and 81 the character; this means that the function attached to c4 has to map the 81 onto \amacron
I'm not sure whether c4 is the trigger for the 81 character.
c4 81 is two-byte representation in memory (that's what you'll see in some hexadecimal editor) of Unicode amacron character with the code U+0101, or simply said: utf-8 code for amacron :-)
can you make me a file with a list like:
amacron : 01/01 : c4/c8 : <utfcode> ^^^^^^^^^^^^^^^^^^^^^^^^^ ^^^^^^^^^ normal ascii real utf
So, the line for amacron should look like:
amacron : 01/01 c4/c8
since c4/c8 is utfcode for amacron.
Is this OK?
I do not think the mapping files should touch utf-8. The input mechanism should map utf-8 to unicode, and then the mapping should map unicode to a macro. In that way the same mapping can be used by other encodings, provided they have an input mapping to unicode. Simon -- Simon Pepping email: spepping@scaprea.hobby.nl