How can I loop thru a list of inputs that come after \maketitle, yet use that list as part of the \maketitle command?

Question

Multilingual Document

Question

I would like to follow the principle of "don't repeat yourself" and create a single list of inputs, which are actually various versions of the same document in different languages. This question is related to this.

---

This question is part of a trail of questions. The goal mine was to use this list inside of a tikz node, but I removed that from this question in the interest of keeping it simple. So, for those interested, here is a quick list for reference:

Related Question for making a macro that can handle post-parsing-inputted list items, see:

• How can I create a list as a variable that can be appended and eventually typeset?

For that same solution with tikz node support, see:

• How can I use a list stored in aux and insert its output into a tikz node?

---

UPDATE 2015-03-08: The following code based on egreg's answer

works 50% of the time

see How can I use a list stored in aux and insert its output into a tikz node?

Situation

I'd like to create a command that has:

• iso 639-1 code as #1
• language name in native tongue as #2
• input file as #3

example: \inputlanguagefile{En}{English}{./Languages/en.tex}

Then I would like to create a list of the language codes from these inputs that can be used in my custom \maketitle command (warning: \maketitle inputs are sent directly to tikz nodes)

Problem

The difficulty lies in the fact that the list is called before it is created. I would probably need to create some kind of command to loop through the inputs beforehand. The document would look something like this:

%\maketitle[subtitle]{title}
\maketitle[listoflanguagecodes]{Multilingual Document}
\inputlanguagefile{En}{English}{./Languages/en.tex}
\inputlanguagefile{Da}{Dansk}{./Languages/da.tex}
\inputlanguagefile{De}{Deutsch}{./Languages/de.tex}
\inputlanguagefile{It}{Italiano}{./Languages/it.tex}
\inputlanguagefile{Hu}{Magyarország}{./Languages/hu.tex}

Redefined \maketitle

\maketitle[subtitle]{title} \maketitle[\listoflanguagecodes]{Title}

\renewcommand{\maketitle}[2][]{%
\thispagestyle{empty}
 #1\\
 #2
\clearpage
\setcounter{page}{1}
}%

The second argument i.e. the actual language names are only used to create a language directory with etoc.

MWE

I am trying to put together a minimal working example of a complex setup. Work in progress.

autolanglist/
├── Languages
│   ├── da.tex
│   ├── de.tex
│   └── en.tex
└── usermanual.tex

1 directory, 4 files

Contents usermanual.tex

\documentclass{article}
\usepackage{fontspec}
\usepackage{tikz}
\usetikzlibrary{calc,positioning}
\usepackage[explicit]{titlesec}
\usepackage{tocloft}
\usepackage{etoc}
\usepackage{lipsum}
\usepackage{hyperref}

\newcounter{runningsectioncounter}
\setcounter{runningsectioncounter}{0}
\titleformat{\section}[hang]{\color{red}\Huge\bfseries}{}{0pt}{\thesection\quad#1}[\stepcounter{runningsectioncounter}]

% Use this \newlangfile{<code>}{<name>}{<file>} to add external language varieties
\newcommand{\inputlanguagefile}[3]{%
  \newpage\pdfbookmark{#2}{bkm#1}%
  \etoctoccontentsline{part}{#2}%
  \setcounter{runningsectioncounter}{\value{section}}
  \setcounter{section}{0}
  \renewcommand*{\theHsection}{chX.\the\value{runningsectioncounter}} % Keeps hyperref happy (provides unique section numbers instead of using the section counter)
  \input{#3}%
}

\renewcommand{\maketitle}[2][]{%
    #1\\
    #2
       
\clearpage
}%

\newcommand{\usermanualfrontmatter}{%
    \maketitle[En, Da, De]{User Manual} % NOTE: This is the list of language codes should be automated!
    \newpage
    \setcounter{tocdepth}{0} % make ToC contain only parts
    \renewcommand{\contentsname}{Language Directory} % change ToC Title
    \tableofcontents
    \newpage
    \setcounter{tocdepth}{4}
    \renewcommand{\contentsname}{Table of Contents} % reset ToC Title back to default
}%

\begin{document}
\usermanualfrontmatter{}
\inputlanguagefile{En}{English}{./Languages/en.tex}
\inputlanguagefile{Da}{Dansk}{./Languages/da.tex}
\inputlanguagefile{De}{Deutsch}{./Languages/de.tex}
\end{document}

Contents of ./Languages/da.tex

\maketitle[da]{Brugsanvisning}
\newpage
\localtableofcontents
\newpage
\section{Test}
\lipsum[50]

Contents of ./Languages/de.tex

\maketitle[de]{Gebrauchsanweisung}
\newpage
\localtableofcontents
\newpage
\section{Test}
\lipsum[50]

Contents of ./Languages/en.tex

\maketitle[en]{User Manual}
\localtableofcontents
\clearpage
\section{Test}
\lipsum[50]

Output

The first page is scaled down A4 paper, but the rest were resized too 100mm x 150mm to save screen real estate on this website :) I attempted to clearly show which list should be generated automatically. The contents of which are not added until later in the document (\inputlanguagefile).

Answer 1

You have to write the data in the .aux file for having the information available at the next run; we exploit the fact that the .aux file is read as part of \begin{document}, so \manuallanguages will have the requested value (at the next run) when \maketitle is issued.

\documentclass{article}

\makeatletter 
\newcommand{\manuallanguages}{\@gobble} % initialization
\newcommand{\inputlanguagefile}[1]{%
  \input{\manualfile{#1}}%
  \protected@write\@auxout{}{\string\used@language{#1}}%
}
\AtEndDocument{\let\used@language\@gobble}
\newcommand{\used@language}[1]{%
  \g@addto@macro\manuallanguages{\language@sep\format@language{#1}}%
}
\newcommand{\language@sep}{, }% separator
\newcommand{\format@language}[1]{%
  \expandafter\expandafter\expandafter\@secondoftwo % or \@firstoftwo
  \csname language@#1\endcsname
}
\newcommand\definemanuallanguage[2]{%
  \@namedef{language@#1}{{#1}{#2}}%
}
\makeatother

\newcommand\manualfile[1]{./UserManual_#1}% use your own

\definemanuallanguage{en}{English}
\definemanuallanguage{de}{Deutsch}
\definemanuallanguage{da}{Dansk}

\begin{document}

\title{User Manuals\\
  \manuallanguages}

\author{macmadness86}

\maketitle

\inputlanguagefile{en}

\inputlanguagefile{da}

\inputlanguagefile{de}

\end{document}

Each \inputlanguagefile adds a note in the .aux file in the form

\used@language{xy}

where xy is the argument to \inputlanguagefile.

Define \manualfile to work according to your setup, here I have the manual files in the same directory as the main file.

If you prefer to have just the ISO abbreviations (which are lowercase), change \@secondoftwo into \@firstoftwo.

Answer 2

last update: shows how to adapt the concrete code which was added to the Question in an edit. See bottom.

As you mentioned etoc, I will explain that with it you can achieve what you want without having to write things to the aux file.

I have also added hyperref just to show that hyperlinks will be there.

The idea is that with etoc you can print your comma separated list of used languages via the command \tableofcontents, and placing this command as a node in a TikZ picture is no problem.

Update adds a variant.

\documentclass{article}
\usepackage[T1]{fontenc}
\usepackage[utf8]{inputenc}

\usepackage{tikz}
\usetikzlibrary{calc, positioning}

\usepackage{etoc}

% in this example, user manuals are each printed in their own
% \part. If there were no other parts, we would not need to define 
% a special division called "languagefile", we would simply use "part".

% I use a \part below in \inputlanguagefile
% to provide an hyperref target.

% Else, without \part, one would only need define a counter 
% and do \refstepcounter in the \inputlanguagefile command.

% Without hyperref, there is no need to do anything extra.
\etocsetlevel{languagefile}{6}

\usepackage{hyperref}% just to check compatibility

\makeatletter

\newcommand\manualfile[1]% #1 = typically, two-letter abbrev for language
           {./UserManual_#1.tex }% use your own preferred location


\newcommand\declaremanuallanguage[2]% #1 = abbrev, #2 = full name
          {\@namedef{languagename@#1}{#2}}


\newcommand{\inputlanguagefile}[1]{%
     \part{\@nameuse{languagename@#1}}
     \etoctoccontentsline{languagefile}{\@nameuse{languagename@#1}}%
     \input{\manualfile{#1}}}

\makeatother

\declaremanuallanguage{en}{English}
\declaremanuallanguage{de}{Deutsch}
\declaremanuallanguage{da}{Dansk}

\newcommand{\manuallanguages}{%
    \begingroup
      \etocsetnexttocdepth {-1}% limit the TOC to what we want
      \etocsetlevel {part}{0}% move parts out of the way
      \etocsetlevel {languagefile}{-1}% only interested in them
      \etocsetstyle {languagefile}
                    % do the trick to get a separator, this is executed once
                    {\def\manual@sep{\def\manual@sep{, }}}
                    % then each entry will do ", " except the first
                    {\manual@sep\etocname}{}{}%
      \etocsettocstyle {}{}% suppress standard Contents heading
% we will be in vertical mode.
\begin{tikzpicture}
    \node [font=\Huge\bfseries] (title) at 
      ($ (current page.north)!.25!(current page.south) $) {User manuals};
    \node [font=\Large\bfseries] (subtitle) [below= 2cm of title] 
          {\tableofcontents};% etoc's magic ! Each entry is a link.
\end{tikzpicture}
    \endgroup
%\clearpage
\setcounter{page}{1}
}%


% only for sample code, remove from final version
% as we don't provide an extension, .tex  suffix is appended.
\usepackage{filecontents}
\begin{filecontents}{UserManual_en}
This is the English manual.
\end{filecontents}
\begin{filecontents}{UserManual_de}
Dies ist die deutsche Bedienungsanleitung.
\end{filecontents}
\begin{filecontents}{UserManual_da}
Dette er den danske manual.
\end{filecontents}
\begin{filecontents}{UserManual_fr}
Ceci est le manuel français.
\end{filecontents}

\title{\manuallanguages}

\author{macmadness86}

\begin{document}\pagestyle{empty}% only for cropping pages for png's

\maketitle\thispagestyle{empty}

\vskip1cm
\hrule
\clearpage

\inputlanguagefile{en}

\vskip1cm
\hrule
\clearpage

\inputlanguagefile{da}

\vskip1cm
\hrule
\clearpage

\inputlanguagefile{de}
\vskip1cm
\hrule

\end{document}

(the rules are only for the pages to be the same width even after cropping to produce the images)

Variant:

\documentclass{article}
\usepackage[T1]{fontenc}
\usepackage[utf8]{inputenc}

\usepackage{tikz}
\usetikzlibrary{calc, positioning}

\usepackage{etoc}

\usepackage{hyperref}% just to check compatibility

% location of user manuals
\newcommand\manualfile[1]{./UserManualv2_#1.tex }%

% innocent wrapper
\newcommand{\inputlanguagefile}[1]{\input{\manualfile{#1}}}

% a fictional sectioning unit,
\etocsetlevel{languagefile}{6}

\newcommand{\thisismanualfor}[1]{% 
    \part{#1}% convenient way to create an hyperref target
    \etoctoccontentsline{languagefile}{#1}%
}%

\newcommand{\manuallanguages}{%
    \begingroup
      \etocsetnexttocdepth {-1}% limit the TOC to what we want
      \etocsetlevel {part}{0}% move parts out of the way
      \etocsetlevel {languagefile}{-1}% only interested in them
      \etocsetstyle {languagefile}
                    % do the trick to get a separator, this is executed once
                    {\def\manual@sep{\def\manual@sep{, }}}
                    % then each entry will do ", " except the first
                    {\manual@sep\etocname}{}{}%
      \etocsettocstyle {}{}% suppress standard Contents heading
% for the record: etoc always does a \par when doing \tableofcontents,
% which can be turned off via \etocinline command; it turns out
% we don't need to worry about this in the TikZ's node.
\begin{tikzpicture}
    \node [font=\Huge\bfseries] (title) at 
      ($ (current page.north)!.25!(current page.south) $) {User manuals};
    \node [font=\Large\bfseries] (subtitle) [below= .5cm of title] 
          {\tableofcontents};% etoc's magic ! Each entry is a link.
\end{tikzpicture}
    \endgroup
}%


% ONLY for sample code, REMOVE from final version
% as we don't provide an extension, .tex  suffix is appended.
\usepackage{filecontents}
\begin{filecontents}{UserManualv2_en}
\thisismanualfor{English}%
This is the English manual.
\end{filecontents}
\begin{filecontents}{UserManualv2_de}
\thisismanualfor{Deutsch}%
Dies ist die deutsche Bedienungsanleitung.
\end{filecontents}
\begin{filecontents}{UserManualv2_da}
\thisismanualfor{Dansk}%
Dette er den danske manual.
\end{filecontents}
\begin{filecontents}{UserManualv2_fr}
\thisismanualfor{Français}%
Ceci est le manuel français.
\end{filecontents}

\title{\manuallanguages}

\author{macmadness86}

\begin{document}

\maketitle

\part{Introduction}

\inputlanguagefile{fr}

\inputlanguagefile{en}

\inputlanguagefile{da}

\part{Conclusion}

\hrule

\end{document}

---

Now that the question has complete code sample I can provide an answer addressing more directly the specific aims.

This is a copy of the code with additions tagged by % ADD THIS lines.

\documentclass{article}
%% \usepackage{fontspec} % commented out to compile with pdflatex
\usepackage{tikz}
\usetikzlibrary{calc,positioning}
\usepackage[explicit]{titlesec}
\usepackage{tocloft}
\usepackage{etoc}
% ADDTHIS
    \etocsetlevel{languageabbrev}{6}
%
\usepackage{lipsum}
\usepackage{hyperref}

\newcounter{runningsectioncounter}
\setcounter{runningsectioncounter}{0}
\titleformat{\section}[hang]{\color{red}\Huge\bfseries}{}{0pt}{\thesection\quad#1}[\stepcounter{runningsectioncounter}]

% Use this \newlangfile{<code>}{<name>}{<file>} to add external language varieties
\newcommand{\inputlanguagefile}[3]{%
  \newpage\pdfbookmark{#2}{bkm#1}%
  \etoctoccontentsline{part}{#2}%
% ADDTHIS
\etoctoccontentsline{languageabbrev}{#1}%
%
  \setcounter{runningsectioncounter}{\value{section}}
  \setcounter{section}{0}
  \renewcommand*{\theHsection}{chX.\the\value{runningsectioncounter}} % Keeps hyperref happy (provides unique section numbers instead of using the section counter)
  \input{#3}%
}

\renewcommand{\maketitle}[2][]{%
    \begin{tikzpicture}[remember picture, overlay]
        \node [font=\Huge] (title) at ($ (current page.north)!.25!(current page.south) $) {#2};
        \node [font=\Large,below=2cm of title] (subtitle) {#1};
        %\node [color=red,draw,below=1cm of subtitle] (note) {This list should be generated automatically.};
        %\draw [->,ultra thick,red] (note) -- (subtitle);
    \end{tikzpicture}
\clearpage
}%

\newcommand{\usermanualfrontmatter}{%
% ADDTHIS
    \GetUsedLanguages
%
    \maketitle[\UsedLanguages]{User Manual}
    \newpage
    \setcounter{tocdepth}{0} % make ToC contain only parts
    \renewcommand{\contentsname}{Language Directory} % change ToC Title
    \tableofcontents
    \newpage
    \setcounter{tocdepth}{4}
    \renewcommand{\contentsname}{Table of Contents} % reset ToC Title back to default
}%

% ADDTHIS
\makeatletter % to use \g@addto@macro from LaTeX's kernel
\newcommand{\GetUsedLanguages}{%
  \begingroup % work in a group to undo our changes automatically on exit
      \etocsetnexttocdepth {-1}% limit the TOC to what we want
      \etocsetlevel {part}{0}% move parts out of the way
      \etocsetlevel {languageabbrev}{-1}% only interested in them
      \etocsetstyle {languageabbrev}
                    {\gdef\UsedLanguages{}\etocskipfirstprefix}
                    {\g@addto@macro\UsedLanguages{, }}
                    {\expandafter\g@addto@macro\expandafter\UsedLanguages
                     \expandafter{\etocthename}}%
                    {}%
      \etocsettocstyle {}{}% suppress standard Contents heading
      \etocinline % suppress \par, thus TOC does not typeset anything at all
      \tableofcontents % fill up \UsedLanguages
  \endgroup
}
\makeatother



% ONLY FOR ALL IN ONE CODE SAMPLE
%
\usepackage{filecontents}

\begin{filecontents}{da.tex}
\maketitle[da]{Brugsanvisning}
\newpage
\localtableofcontents
\newpage
\section{Test}
\lipsum[50]
\end{filecontents}

\begin{filecontents}{de.tex}
\maketitle[de]{Gebrauchsanweisung}
\newpage
\localtableofcontents
\newpage
\section{Test}
\lipsum[50]
\end{filecontents}

\begin{filecontents}{en.tex}
\maketitle[en]{User Manual}
\localtableofcontents
\clearpage
\section{Test}
\lipsum[50]
\end{filecontents}

\begin{document}
\usermanualfrontmatter{}
\inputlanguagefile{En}{English}{./en.tex}
\inputlanguagefile{Da}{Dansk}{./da.tex}
\inputlanguagefile{De}{Deutsch}{./de.tex}
\end{document}