[Tex/LaTex] Writing “Proof of . . .” cleanly in svjour3

spacingsv-classestheorems

Perhaps I am spoiled by amsthm, but I rather like being able to freely rename the proof environment, especially when there are intervening statements. For example:

Example 1: nice behavior

\documentclass[11pt]{article}
\usepackage{amsthm}
\newtheorem{prop}{Proposition}[section]
\newtheorem{lem}[prop]{Lemma}

\begin{document}
\begin{prop} \label{eq:prop1}
 Some proposition.
\end{prop}
We prove the above proposition with the help of the following lemma (see [citation]).
\begin{lem}
A useful lemma.
\end{lem}
\begin{proof}[Proof of Prop.~\ref{eq:prop1}]
This is desired behaviour.
\end{proof}
\end{document}

A standard environment: the original "proof" is overwritten

svjour3, however, does not have this property. The proof environment as predefined does not omit the original "Proof." text, and puts your addition in parentheses. My workaround is to have a proofrename proof with an empty name, and following their suggestion, to use the theopargself environment to omit the parentheses around the optional argument.

Example 2: bad behavior 

\RequirePackage{fix-cm}
\documentclass[smallextended,envcountsame,envcountsect]{svjour3}    
\smartqed  % flush right qed marks, e.g. at end of proof
\usepackage{graphicx}
\usepackage{mathptmx}      % use Times fonts if available on your TeX system
\usepackage{amsmath,amssymb}

\spnewtheorem*{proofrename}{}{\itshape}{\rm}

\begin{document}

\begin{proposition} \label{prop:prop1}
The \texttt{proof} environment is strange in \texttt{svjour3}.
\end{proposition}
We prove the above proposition with the help of the following lemma (see [citation]).
\begin{lemma}
A useful lemma.
\end{lemma}

\begin{proof}[Proof of \ref{prop:prop1}, Version 1]
Optional arguments do \textit{not} supersede the ``proof'' statement.  
\end{proof}

\begin{theopargself}
\begin{proofrename}[Proof of \ref{prop:prop1}, Version 2.]
Even suppressing the name, there is still some whitespace.  
\end{proofrename}
\end{theopargself}
\end{document}

svjour3 behavior: the proof is not omitting, or there is some whitespace

Are there any suggestions to get a better proof environment? I know it is possible, because of one of the free articles at the journal I'm considering: see the free article by Z. Ercan at Positivity.

Note: I see that most of the other svjour3 answers use \def ; however, the template LaTeX file that comes with svjour3 requests that you eschew \def in favor of \newcommand . Therefore, I will prefer answers without \def, though I recognize that this may not be possible.

Best Answer

Using just standard svjour commands and exploiting the fact that \spnewtheorem*{foo}{Foo}{}{} stores Foo as the expansion of \fooname:

\documentclass[smallextended,envcountsame,envcountsect]{svjour3}
\smartqed  % flush right qed marks, e.g. at end of proof

\spnewtheorem*{xproof}{}{\itshape}{\rmfamily}% the label is assigned later
\newcommand\xprooftitle{}
\renewenvironment{proof}[1][\proofname]
 {\renewcommand\xproofname{#1}\xproof}
 {\endxproof}

\begin{document}
\begin{proposition}\label{prop1}
Some proposition.
\end{proposition}
We prove the above proposition with the help of the following lemma (see [citation]).
\begin{lemma}
A useful lemma.
\end{lemma}
\begin{proof}[Proof of Prop.~\ref{prop1}]
This is desired behaviour.\qed
\end{proof}
\end{document}

enter image description here

The advantage of this approach is that if the people at Springer don't like your change to proof, they'll just need to remove the five lines above. Mark them clearly in the file with comments specifying what you did:

% I redefine the `proof` environment in order that
% \begin{proof}[Proof of the main theorem]
% produces ``Proof of the main theorem'' and not
% ``Proof (Proof of the main theorem)''
\spnewtheorem*{xproof}{}{\itshape}{\rmfamily}% the label is assigned later
\newcommand\xprooftitle{}
\renewenvironment{proof}[1][\proofname]
 {\renewcommand\xproofname{#1}\xproof}
 {\endxproof}
% Comment the five lines above to restore the default behavior of `proof'
Related Question