hzluo> http://bugzilla.lyx.org/show_bug.cgi?id=3496 I can't compile
hzluo> under my system without this patch. My system is XP SP2, and
hzluo> local set to Simplified Chinese. So these non-ASCII chars are
hzluo> read as first bytes of GBK, but they are supposedly latin-1.
hzluo> The compiled result is supposedly identical. Could anyone
hzluo> comment it? Thanks!
Jose'? This looks reasonable to me, we should not have explicit 8bit
code anyway. The only weird thing is the tables for hebrew, since
there should not be any of this now that we use unicode.
I suggest to put these counters to one or several data file(s),
including hebrewCounter(), lowerromanCounter(), romanCounter(),
and maybe loweralphaCounter()/alphaCounter(). I think there
should be arabCounter() for the consistency?
And hebrewCounter() should actually be localCounter() or not?
I'm not sure about this. If yes, I strongly suggest to put it to a data
file, so that we can adapt to different language. It's much better
than to add more counters such as chineseCounter(). And we
do need similar thing as localCounter(), when I use Chinese :)
Hangzai
