Hi all, incidentally, I found one issue in documents compiled to MathML, from two independent sources. Meaningless MWE:
\documentclass{article} \begin{document} [this is for test $a+b=c]$ \end{document} What is important is `]` at the end of inline math. It produces the following result with `mathml` option: <!--l. 3--><p class="noindent" >[this is for test <!--l. 3--><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline" ><mi >a</mi> <mo class="MathClass-bin">+</mo> <mi >b</mi> <mo class="MathClass-rel">=</mo> <mi >c</mi></mrow><mo class="MathClass-close">]</mo></mrow></math> </p> The issue is that there are two clossing `</mrow>` tags, but none is opened. I've found that it can be fixed using `mathml-` option: <!--l. 3--><p class="noindent" >[this is for test <!--l. 3--><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline" ><mi >a</mi> <mo class="MathClass-bin">+</mo> <mi >b</mi> <mo class="MathClass-rel">=</mo> <mi >c</mi><mo class="MathClass-close">]</mo></math> </p> It seems that one must use paired brackets (they don't have to be the same kind $(a+b=c]$ works as well) with default tex4ht configuration. Every opening bracket add two opening `<mrow>` tags and closing bracket closes that. As I am not math expert, my question is: Is the original example actually correct? Shouldn't be all brackets paired, or is the default tex4ht configuration wrong in assuming this? I know that one must use \left. if there is no opening bracket for \right). Is there similar mechanism for normal brackets? Could anybody test if `mathml-` option gives correct results for more complicated real world examples? Best regards, Michal