Open angerhang opened 8 years ago
I am not even sure that the LaTeX is what it should be. I am currently looking at using a more standard "associative argument" markup:
\symdef[name=defined-piecewise,assocarg=1]{piecewise}[1]{\left\{\parray{rl}{\assoc{#1}{}}\right.}
\symtest{piecewise}{|x|:=\piecewise{\piece{x}{x>0},\piece{-x}{x<0},\otherwise{0}}}
and that generates
<omdoc:prototype>
<om:OMA>
<om:OMS cd="piecewise" cr="fun" name="defined-piecewise"/>
<omdoc:exprlist name="args">
<omdoc:expr name="arg"/>
</omdoc:exprlist>
</om:OMA>
</omdoc:prototype>
<omdoc:rendering>
<XMTok xmlns="http://dlmf.nist.gov/LaTeXML" role="OPEN">{</XMTok>
<XMArray xmlns="http://dlmf.nist.gov/LaTeXML">
<omdoc:iterate name="args">
<omdoc:separator>
<omdoc:render name="arg1"/>
</omdoc:separator>
<omdoc:render name="arg"/>
</omdoc:iterate>
</XMArray>
<XMHint xmlns="http://dlmf.nist.gov/LaTeXML"/>
</omdoc:rendering>
</omdoc:notation>
which looks a bit better. But it does not work at all in the PDFLaTeX case, and there are quite a few schema errors in the LaTeXML case as well.
I have the feeling that this will be the more fruitful avenue.
In
.../MathHub/smglom/mv/source/piecewise.tex
we havewhich generates the notation definition
This is wrong, instead of
<m:mtr/>
it should bethere is also an error that seems to describe what is happening: