LaTeX code output by Calc wrong

Question

If I enable calc-embedded mode (C-x * e) in a LaTeX document (latex mode by default) when the point is in the math environment of the document below:

\documentclass{article}
\usepackage{amsmath}
\begin{document}
\newcommand*{\evalto}{}
\renewcommand*{\to}{=}
\begin{math}
[[ a, b ][ c, d ]] * [[x][y]]   =>
\end{math}
\end{document}

calc embedded produces the following code

\documentclass{article}
\usepackage{amsmath}
\begin{document}
\newcommand*{\evalto}{}
\renewcommand*{\to}{=}
\begin{math}
\evalto \begin{pmatrix} a & b \\ c & d \end{pmatrix} \begin{pmatrix} x \\ y \end{pmatrix} \to \begin{pmatrix} a x + b y \\ c x + d y \end{pmatrix}
\end{math}
\end{document}

which perfectly render the desired output.

But this is not the case if I want to activate the latex 2 display mode to have a prettier source code:

% [calc mode: language: (latex 2)]
\begin{math}
\evalto \begin{pmatrix} \times \begin{pmatrix} \to \begin{pmatrix}
        a & b \\               x \\                a x + b y \\ 
        c & d                  y                   c x + d y
        \end{pmatrix}          \end{pmatrix}       \end{pmatrix}
\end{math}

Which does not render the desired output. So I have to move rectangular data to get it.

Is there a way to get the correct output in an easy way?

Answer 1

This is documented behaviour. From section 7.8.3 of the Calc manual:

If you invoke ‘d T’ or ‘d L’ with an argument of 2 or −2, then matrices will be displayed in two-dimensional form, such as

 \begin{pmatrix}
 a & b \\
 c & d
 \end{pmatrix}

This may be convenient for isolated matrices, but could lead to expressions being displayed like

 \begin{pmatrix} \times x
 a & b \\
 c & d
 \end{pmatrix}

While this wouldn’t bother Calc, it is incorrect LaTeX. (Similarly for TeX.)