[Developers] Re: patch: include amsmath package in cactus latex format

[Tom Goodale]

Befoe adding this, I'd like to ask if anyone has seen a latex distribution 
where this is not available ?

Tom

On Sat, 4 Jun 2005, Jonathan Thornburg wrote:

> Summary
> =======
> doc/latex/cactus.sty  already includes 9 widely-available latex packages.
> I propose that we add a 10th, the  amsmath  package.  This would provide
> math-formatting capabilities which are currently lacking, and which
> would be useful in writing Cactus thorn documentation.
>
>
> Details
> =======
> In current-CVS cactus, equation numbers are just integers.  For example,
> if (in a thorn guide documentation.tex file) I write
>
>  As implemented here, the shift condition is
>  \begin{eqnarray}
>  \partial_t \beta^i      & = &   B^i                                     \\
>  \partial_t B^i          & = &   F(\alpha) \partial_t \tilde{\Gamma}^i
>                                  - \eta B^i 
> %%%\\
>  \end{eqnarray}
>  or
>  \begin{eqnarray}
>  \partial_t \beta^i      & = &   F(\alpha) B^i                           \\
>  \partial_t B^i          & = &   \partial_t \tilde{\Gamma}^i - \eta B^i 
> %%%\\
>  \end{eqnarray}
>  where in all cases
>  \begin{equation}
>  F(\alpha) = \frac{3}{4} \alpha
>  \end{equation}
>
> there is no way to have the first pair of equations numbered (1a) and (1b)
> and the second pair numbered (2a) and (2b), and \label these so I can
> later refer to equations (1) and (2) in describing a parameter which
> selects one or the other.
>
> Currently, doc/latex/cactus.sty includes
>
>  \RequirePackage[latin9]{inputenc}
>  \RequirePackage{fancyhdr}
>  \RequirePackage{graphicx}
>  \RequirePackage{latexsym}
>  \RequirePackage{amssymb}
>  \RequirePackage{ifthen}
>  \RequirePackage{calc}
>  \RequirePackage{thumbpdf}
>  \RequirePackage{hyperref}
>
> I propose that we add the  amsmath  package to this list (patch attached).
>
>
> Among other useful features, this defines a  subequations  environment,
> which nicely solves my problem:
>
>  As implemented here, the shift condition is
>  \begin{subequations}
>  \label{EvolvedShift/eqn-Gamma-driver-weak-freezing}
>  \begin{eqnarray}
>  \partial_t \beta^i      & = &   B^i                                     \\
>  \partial_t B^i          & = &   F(\alpha) \partial_t \tilde{\Gamma}^i
>                                  - \eta B^i 
> %%%\\
>  \end{eqnarray}
>  \end{subequations}
>  or
>  \begin{subequations}
>  \label{EvolvedShift/eqn-Gamma-driver-strong-freezing}
>  \begin{eqnarray}
>  \partial_t \beta^i      & = &   F(\alpha) B^i                           \\
>  \partial_t B^i          & = &   \partial_t \tilde{\Gamma}^i - \eta B^i 
> %%%\\
>  \end{eqnarray}
>  \end{subequations}
>  where in all cases
>  \begin{equation}
>  F(\alpha) = \frac{3}{4} \alpha
>  \end{equation}
>
>  This is controlled by the Boolean parameter \verb|strong_freezing|:
>  If this is false, we have the first
>  choice~\eqref{EvolvedShift/eqn-Gamma-driver-weak-freezing},
>  since $\alpha \to 0$ implies only $\partial_{tt} \beta^i \to 0$.
>  If this is true, we have the second
>  choice~\eqref{EvolvedShift/eqn-Gamma-driver-strong-freezing},
>  since $\alpha \to 0$ then implies $\partial_t \beta^i \to 0$.
>
> Comments?
>
> ciao,
>
>