Re: XeTeX encoding problem

[Gavin Smith]

On 7 February 2016 at 16:23, Gavin Smith <address@hidden> wrote:
> In a patch that I'm still working on, I have a definition like
>
> +\def\gdefchar#1#2{%
> +\gdef#1{%
> +   \ifpassthroughchars
> +     \string#1%
> +   \else
> +     #2%
> +   \fi
> +}}
>
> and then e.g. in \latonechardefs
>
>  \def\latonechardefs{%
> -  \gdef^^a0{\tie}
> +  \gdefchar^^a0{\tie}
>

> I hope it's clear what I'm trying to do here: instead of redefining,
> change the value of a conditional that is used within the macro.

I've finally finished this. It does appear a bit faster as well as
being simpler. Where it could potentially break is chapter names with
non-ASCII characters, and index entries with non-ASCII characters, as
well as macros using non-ASCII characters (either in the macro
definition or in the arguments), as well as macros giving chapter
names and index entries. I'll do a bit more testing later.

Masamichi, does the native multibyte support still work well? Is there
any more that you wanted to do with it? If it looks good we should
look to releasing it as an official version for non-developers.