Using \ul from OPmac tricks in OpTeX

[robertbachmann]

Using the following code works for me. However if I use \typosize[16/18] I get the error message

! Argument of \wordscanB has an extra }.

I suspect it is related to \let\hyphentt=\tentt but didn't have time to investigate further.

% !TeX program = optex
\useOpTeX \fontfam[lm]

\typosize[10/12]
%\typosize[16/18]



\newdimen\hyphdim 
\let\hyphentt=\tentt  \hyphdim=\fontdimen6\hyphentt \divide\hyphdim by2 
 
\def\hyphenprocess#1{\def\tmp{#1}\let\listwparts=\undefined 
   \setbox0=\vbox\bgroup\hyphenpenalty=-10000 \hsize=0pt \hfuzz=\maxdimen 
      \edef\hychar{\hyphenchar\hyphentt=\the\hyphenchar\hyphentt}\hyphenchar\hyphentt=45 
      \def~{\nobreak\hskip.5em\relax}\tt\noindent\hskip0pt\relax #1\par \hychar 
      \hyphenprocessA 
   \expandafter\let\expandafter\listwparts\expandafter\empty 
      \expandafter\hyphenprocessB\listwparts 
} 
\def\hyphenprocessA{\setbox2=\lastbox 
   \ifvoid2 \egroup \else \unskip \unpenalty 
      \setbox2=\hbox{\unhbox2}% 
      \tmpnum=\wd2 \advance\tmpnum by100 \divide\tmpnum by\hyphdim 
      \ifx\listwparts\undefined \xdef\listwparts{,}% 
      \else \advance\tmpnum by-1 \xdef\listwparts{\the\tmpnum,\listwparts}\fi 
              \expandafter\hyphenprocessA \fi 
} 
\def\hyphenprocessB#1,{\if^#1^\expandafter\hyphenprocessC 
   \else \tmpnum=#1 \expandafter\hyphenprocessD\tmp\end 
   \fi 
} 
\def\hyphenprocessC{\expandafter\addto\expandafter\listwparts\expandafter{\tmp}} 
\def\hyphenprocessD#1#2\end{\addto\listwparts{#1}% 
   \advance\tmpnum by-1 
   \ifnum\tmpnum>0 \def\next{\hyphenprocessD#2\end}% 
   \else 
   \def\tmp{#2}\def\next{\addto\listwparts{\-}\expandafter\hyphenprocessB}\fi 
   \next 
} 

%%%%%%

\def\ul#1{{\ulRedefine\leavevmode\wordscanA #1 {} }} 
\def\wordscanA#1 {\ifx^#1^\unskip\else \wordscanB#1\-\end \expandafter\wordscanA\fi} 
\def\wordscanB#1\-#2\end{\ifx^#2^\wordprintA{#1}\else 
   \wordprintB{#1}\def\next{\wordscanB#2\end}\expandafter\next\fi} 
\def\wordprintA#1{\setbox0=\hbox{#1}\hbox{\rlap{\copy0}\uline\wd0}\uline\uspace\relax} 
\def\wordprintB#1{\setbox0=\hbox{#1}\hbox{\rlap{\copy0}\uline\wd0}\-} 
\def\uline{\leaders \vrule height-1.9pt depth2.3pt\hskip} 
\def\uspace{\fontdimen2\font plus\fontdimen3\font minus\fontdimen4\font} 
\def\ulRedefine{\def~{\egroup\hbox{\rlap{\copy0}\uline\wd0}\nobreak\uline\uspace\relax 
         \setbox0=\hbox\bgroup}} 
         
\def\wordscanA#1 {\ifx^#1^\unskip\else 
   \hyphenprocess{#1}\expandafter\wordscanB\listwparts\-\end \expandafter\wordscanA\fi} 

%%%%%%

Lorem ipsum dolor sit amet. \ul{Lorem ipsum dolor sit amet, consectetuer adipiscing elit. Aenean commodo ligula eget dolor.} 
Aenean massa. Cum sociis natoque penatibus et magnis dis parturient montes, nascetur ridiculus mus. 

\bye

[olsak]

The problem is that the internal hyphenated text is typeset by \tt and this is a macro in OpTeX which selects the font depending on the current size. But the macro expects only 10 pt font here. Solution: replace \tt at line 15 in your code by \hyphentt.

I upgraded the Opmac-tricks web page.

But this will work only for English text in OpTeX, because \tentt preloded in the format is EC font which covers English alphabet. If we are using somewhat different language (for example Czech), then we need use \fontfam first and then use \fontdef\hyphentt{\tt} instead \let\hyphentt=\tentt . This feature cannot be mentioned in Opmac-Tricks, because \fontdef isn't Opmac macro and Opmac supposes that \tentt is always correctly selected.