On 10-3-2010 10:54, Heiko Oberdiek wrote:
Using `o' it remains hidden whether the font contains `aT' kerning.
Therefore I think `a' should be used instead of `o'.
sure, it's just an example that assumes that there is no aT kern but
maybe a Ta which is then wrong since the shape is non symmetrical hence
the o.
\ifdim\wd0=\wd2
\setbox0\hbox dir TRT{To}%
I don't see the point in using TRT. What's the difference to \hbox{oT}?
lack of kerns of oT in most fonts (we often have upper-lower kerns but
not the reverse); the TRT is just a touch of luatex -)
A variant that tries the implicite kernings in the order of
`aT', `oT', `To'. Also the code is compatible to plain-TeX:
you mean traditional (non-etex) i guess
\def\LuaTeX{%
\begingroup
Lua%
\setbox0\hbox{aT}%
\setbox2\hbox{a\kern0ptT}%
\ifdim\wd0<\wd2 %
\else
\setbox0\hbox{oT}%
\setbox2\hbox{o\kern0ptT}%
\ifdim\wd0<\wd2 %
\else
\setbox0\hbox{To}%
\setbox2\hbox{T\kern0pto}%
\fi
\ifdim\wd0<\wd2 %
\dimen0=\wd0 %
\advance\dimen0-\wd2 %
\kern\dimen0 %
\fi
\fi
\TeX
\endgroup
}
sure testing for those often non present kerns makes sense and we need a
bit of stay in hmode code and so and we get our complicated looking logo -)
btw, regular times makes a nice test case
Hans
-----------------------------------------------------------------
Hans Hagen | PRAGMA ADE
Ridderstraat 27 | 8061 GH Hasselt | The Netherlands
tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com
| www.pragma-pod.nl
-----------------------------------------------------------------
