path: root/Master/texmf-dist/doc/latex/chemnum/chemnum_en.tex

% arara: pdflatex: { shell: on }
% arara: makeindex: { sort: true, style: chemnum_en.ist }
% arara: pdflatex: { shell: on }
% arara: pdflatex: { shell: on }
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% - chemnum - chemnum_en.tex
% - a comprehensive approach for the numbering of chemical compounds
% -----------------------------------------------------------------------------
% - Clemens NIEDERBERGER
% - 
% - contact@mychemistry.eu
% -----------------------------------------------------------------------------
% - If you have any ideas, questions, suggestions or bugs to report, please
% - feel free to contact me.
% -----------------------------------------------------------------------------
% - Copyright 2011-2013 Clemens Niederberger
% -
% - This work may be distributed and/or modified under the
% - conditions of the LaTeX Project Public License, either version 1.3
% - of this license or (at your option) any later version.
% - The latest version of this license is in
% -   http://www.latex-project.org/lppl.txt
% - and version 1.3 or later is part of all distributions of LaTeX
% - version 2005/12/01 or later.
% -
% - This work has the LPPL maintenance status `maintained'.
% -
% - The Current Maintainer of this work is Clemens Niederberger.
% -
% - This work consists of the files
% -   chemnum.sty, chemnum_de.tex, chemnum_en.tex, chemnum_it.tex
% -   scheme-tmp.ps, scheme-bla.ps,
% -   README
% -   chemnum_de.pdf, chemnum_en.pdf, chemnum_it.pdf
% -
% - in order to compile this document you need the class `cnpkgdoc' with is
% - available under the LPPL here: https://bitbucket.org/cgnieder/cnpkgdoc/
% -
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\RequirePackage[ngerman,german,naustrian,austrian,british,american,canadian,frenchb,
  canadien,spanish,italian,english]{babel}
\RequirePackage[textgreek=artemisia]{chemnum}
\documentclass[DIV11,toc=index]{cnpkgdoc}
\docsetup{
  pkg      = chemnum,
  code-box = {skipabove=1ex,skipbelow=1ex}
}
\usepackage{fnpct}
\addcmds{
  @m,arrow,bf,
  ch,chemfig,chemname,cmpd,cmpdinit,cmpdref,cmpdreset,cmpdsetup,
    CNlabel,CNlabelnoref,CNref,CNlabelsub,CNlabelsubnoref,CNrefsub,
  color,compound,compoundprefix,compoundsuffix,
  declarecompound,fcite,GetTranslation,
  includegraphics,midrule,penalty,psfrag,renewcommand,
  schemename,schemestart,
  schemestop,selectlanguage,setarrowoffset,setatomsep,setcompoundsep,
    setdoublesep,section,sf,subsection,subsubsection,
  tableofcontents,texorpdfstring,textcolor,theffbibliography,
  uline
}

\usepackage{textcomp,longtable,tabu,booktabs,chemscheme,hologo}
\usepackage[iupac=strict]{chemmacros}
\chemsetup[chemformula]{format=\libertineLF}
\usepackage{chemfig}
\usepackage[runs=2]{auto-pst-pdf}

\usepackage{filecontents}
\begin{filecontents*}{\jobname.ist}
 preamble "\\begin{theindex}\n Section headings are typeset \\textbf{bold}, packages \\textsf{sans serif}, commands \\textcolor{code}{brown}, and options \\textcolor{key}{green}.\\newline\n\n"
 heading_prefix "{\\bfseries "
 heading_suffix "\\hfil}\\nopagebreak\n"
 headings_flag  1
 delim_0 "\\dotfill "
 delim_1 "\\dotfill "
 delim_2 "\\dotfill "
 delim_r "--"
 suffix_2p "\\nohyperpage{\\,f.}"
 suffix_3p "\\nohyperpage{\\,ff.}"
\end{filecontents*}

\let\prog\textsc
\let\name\textsc
\let\CNversion\cnpkgdocpkgversion
\let\CNdate\cnpkgdocpkgdate

\cmpdinit*{a,aa,b,c,d,e,f,g,h,i,j,k,l,m,n,o,p,q,benzene,r,s,t,{1,3-dichlorocyclohexanole}}

\TitlePicture{%
 \selectlanguage{english}
 \cmpdinit*{glucose.alpha, glucose.beta, glucose.chain}
 \cmpd*[sub-counter = greek]{glucose.alpha, glucose.beta}\cmpd*{glucose.chain}
 \setatomsep{2em}
 \definesubmol{x}{(-[4]H)(-[0]OH)}
 \definesubmol{y}{(-[0]H)(-[4]HO)}
 \begin{scheme}[h]
  \centering\small
  \schemestart
   \chemfig{?(-[:-170]HO)-[:-50](-[:170]HO)-[:10](-[:-55,0.7]OH)-[:-10](-[6,0.8]OH)-[:130]O-[:-170]?(-[:150,0.7]-[2,0.7]OH)}
   \arrow(alpha--chain){<=>}
   \chemfig{[2]OH-[3]-!x-!x-!y-!x-=[1]O}
   \arrow(--beta){<=>}
   \chemfig{?(-[:-170]HO)-[:-50](-[:170]HO)-[:10](-[:-55,0.7]OH)-[:-10](-[:10]OH)(-[6,,,,draw=none]\vphantom{OH})-[:130]O-[:-170]?(-[:150,0.7]-[2,0.7]OH)}
   \arrow(@chain--chainlabel){0}[-90,.2] \normalsize\cmpd[cmpd-delim]{glucose.chain}
   \arrow(@chainlabel--){0}[,2.5]        \normalsize\cmpd[cmpd-delim]{glucose.beta}
   \arrow(@chainlabel--){0}[180,2.3]     \normalsize\cmpd[cmpd-delim]{glucose.alpha}
  \schemestop
  \caption{In a solution the \a{} and the \b{} anomer of \D-glucose
    \cmpd[list-delim]{glucose.alpha,glucose.beta} are in equilibrium via the
    open-chain form \cmpd[cmpd-delim]{glucose.chain}.}
 \end{scheme}}

\makeindex
\begin{document}
\selectlanguage{english}\cmpdreset

\section{A very short readme}
\chemnum \CNversion\ underlies the \LaTeX{} Project Public License Version~1.3 or
later. (\url{http://www.latex-project.org/lppl.txt})

\chemnum internally loads the packages \paket*{expl3} and \paket{xparse}.
\paket{xparse} is part of the \paket{l3packages} bundle, \paket*{expl3} is part
of the \paket{l3kernel} bundle. This means, \chemnum needs \LaTeX3 Support.

\chemnum also needs the packages \paket{etoolbox}, \paket{psfrag} and either
\paket{textgreek} or \paket{bm}, see section~\ref{sec:package:options}.

\section{News and a word of apology}
\subsection{Version 0.4}
In the update to v0.4 there have been some changes made with the options. There
are a few new ones and some of the old ones got new names. They now have consistent
names which in my opinion better reflect their function. This means that your
documents set with an older version of \chemnum might not compile any more without
errors. I am sorry for that. \chemnum still is actively developped. Once it reaches
v1.0, such changes won't be made any more\footnote{Or only in absolutely necessary
cases.}.

These options have new names:
\begin{center}
 \begin{tabular}{ll}\toprule
  \bfseries old name     & \bfseries new name  \\\midrule
  \key{cmpd-sep}         & \key{sub-output-sep} \\
  \key{sub-marker}       & \key{sub-input-sep} \\
  \key{sub-range-sep}    & \key{subrange-output-sep} \\
  \key{sub-range-marker} & \key{subrange-input-sep} \\
  \key{strict}           & \key{init-strict} \\
  \key{sub-init}         & \key{init-sub} \\\bottomrule
 \end{tabular}
\end{center}
These ones are new:
\begin{center}
 \begin{tabular}{ll}\toprule
  \bfseries key            & \bfseries short info \\\midrule
  \key{sublist-output-sep} & list separation in the output of labels like
    \cmpd{bsp.{one,two}}, see page \pageref{sssec:list-output-sep}. \\
  \key{sublist-input-sep}  & list separation in the input of labels like
    \cmpd{bsp.{one,two}}, see page \pageref{sssec:list-input-sep}. \\
  \key{init-input-sep}     & input in \cmd{cmpdinit}, see page \pageref{sssec:init-input-sep}. \\
  \key{hyper-use}          & usage of hyperlinks, see page \pageref{sec:hyperref}. \\
  \key{hyper-target}       & \\
  \key{hyper-link}         & \\
  \key{ref-style}          & Output of labels with \cmd{cmpdref}, see
    pages \pageref{ssec:cmpdref} and \pageref{ssec:schriftstil}. \\\bottomrule
 \end{tabular}
\end{center}

Also new is the way how lists and ranges of sublabels are input. The option
\key{sub-list} no longer exists. More on this in section~\ref{ssec:sub_ranges}.

Then there is the possibility of creating hyperlinked labels together with the
package \paket{hyperref}, see section~\ref{sec:hyperref}.

\subsection{Version 0.5}
With v0.5 there is a new option \key{ref-pos} which allows to change the positioning
options of the \cmd{psfrag} command that is used by the \cmd{cmpdref} command
internally, see p.\,\pageref{ssec:marker_in_bildern}.

\achtung{The optional arguments of \cmd{cmpdref} have changed their order, see
p.\,\pageref{ssec:cmpdref}!}

\section{Package options}\label{sec:package:options}\secidx{Package options}
\chemnum has one package option.
\begin{beschreibung}
 \Option{textgreek}{artemisia/\default{cbgreek}/euler/false} Default: \code{false}
\end{beschreibung}
This document was set with \verb+\usepackage[textgreek=artemisia]{chemnum}+.
\secidx*{Package options}

\section{Motivation}
As far as I know there are three packages that are meant to help with numbering
chemical compounds. All of them have their weaknesses.

The first one -- \paket{chemcono} -- redefined bibliography commands for that
purpose. Compounds have to be declared in what is called \verb=\theffbibliography=.
Then one can reference them with \verb=\fcite=. However, it produces a list of
compounds in the text. So the package author suggests:
\begin{zitat}[Stefan Schulz]
 After compilation and printout, discard the last page.
\end{zitat}
Obviously that's not a perfect solution.

The second one -- \paket{chemcompounds} -- was written, because the author didn't
want to work with the weaknesses of \paket{chemcono} any more. When he wrote the
package he basically used the same mechanism to create the labels as \paket{chemcono}
did.
\begin{zitat}[Stephan Schenk]
 When taking a closer look at the chemcono package, I realised that the only thing
 one has to do is to get rid of everything which produces text. Thus, as a basis I
 used the mechanism of \verb=\bibitem= and \verb=\cite= in pretty much the same way
 as \paket{chemcono} does by extracting the corresponding code from \verb=article.cls=
 and \verb=latex.ltx= but deleting any unnecessary commands producing output. I
 also introduced several lines of code to make the printing of the compound names
 more customisable.
\end{zitat}
Some points still left me unsatisfied, though:
\begin{enumerate}
 \item Compounds usually need to be declared with \verb=\declarecompound=. They
   need to be declared in any case if you need a label like \cmpd{bsp.one}. Then,
   one even needs to choose the label by hand, what somehow undermines the automatic
   numbering principle.
 \item The layout can't be changed for a single label but only for all.
 \item The numbers can't be reset. \emph{Although in most cases this is neither
   necessary nor can it be recommended}, there can be individual cases where this
   would be useful.
 \item A list of several compounds \verb=\compound{a,b,c}= can only be customized
   with more effort than what would be convenient.
\end{enumerate}

Then there is \paket{bpchem}, which provides commands similar to \verb=\label=
\verb=\ref=: \verb=\CNlabel{}=, \verb=\CNlabelnoref{}= and \verb=\CNref{}=. It
provides commands for sublabels, too: \verb=\CNlabelsub{}{}=, \verb=\CNlabelsubnoref{}{}=
and \verb=\CNrefsub{}{}=. This makes it more flexible than the others regarding sublabels.
However, it barely provides possibilities to customize the labels, lists are not
possible and the fact that there are different commands for labels and sublabels
isn't the best solution, either.

\chemnum is intended to fill these gaps. For this all commands have been written
from scratch. Some of the ideas of \paket{chemcompounds} \eg regarding delimiters
and layout have been picked up, though.

If you notice any feature missing, please let me know by sending me an email.

\section{Numbering compounds}\label{sec:nummerieren}
\subsection{Main command}\secidx{Main command}\cmpdreset
The main command of this package is this one:
\begin{beschreibung}
 \Befehl{cmpd}{<label name>}
\end{beschreibung}
When \code{<label name>} is used the first time, the label is created, saved (=
declared) and printed. Each further use just prints the label.

\begin{beispiel}
 Compounds \cmpd{a} and \cmpd{b} are declared and can be used any time: \cmpd{a}.
 No pre-declaring is necessary. Compounds like \cmpd{c} are numbered in the order
 they appear in the text.\par
 Once again: \cmpd{b}, \cmpd{a}, \cmpd{c}.
\end{beispiel}

If it is necessary to declare a compound without printing the label, this is
possible with
\begin{beschreibung}
 \Befehl[cmpda]{cmpd*}\ma{<label name>}\label{cmd:unsichtbares_label}
\end{beschreibung}
This declares the label but doesn't print anything.
\achtung{This command is not needed when \cmd{cmpdinit} is used, see
section~\ref{sec:cmpdinit} (pages~\pageref{sec:cmpdinit}ff).}

\begin{beispiel}
 The hidden version\cmpd*{d} declares the label but doesn't print anything. The
 next \cmpd{e} continues to count with the next number. With \cmpd{d} the label
 can be used, of course.
\end{beispiel}

You can pretty much use what you like for a label name except for the separator
symbols (see sections~\ref{ssec:unterlabel}, \ref{ssec:listen}, \ref{ssec:sub_ranges}
and \ref{ssec:trenner}). You shouldn't leave blanks in the name, though. This may
not cause an error at all, but \emph{could} declare different compounds with the
same name.

\lstset{showspaces=true}
\begin{beispiel}
 \cmpd{aa  }, \cmpd{a a }, \cmpd{a  a}, \cmpd{ a a}, \cmpd{  aa} and \cmpd{ aa }
 all have the same label. As well as \cmpd{aa }, \cmpd{a a}, \cmpd{ aa} and \cmpd{aa}.
\end{beispiel}
\lstset{showspaces=false}\secidx*{Main command}

\subsection{Sublabel}\label{ssec:unterlabel}\secidx{Sublabel}
If you want a label like \cmpd{a.one}, you need to use the following syntax:
\begin{beschreibung}
 \Befehl{cmpd}{<label name>.<subname>}
\end{beschreibung}
\code{<label name>} is the main name which stays the same, \code{<subname>} varies.
This syntax means that the point \code{.} \emph{cannot} be a part of \code{<label name>}
or \code{<subname>}. Instead of the point you also can use another symbol, see
section~\ref{sssec:sub-input-sep}.
\begin{beispiel}
 \cmpd{f.one} and \cmpd{f.two} are related, as are \cmpd{g.one} and \cmpd{g.two}.
 Of course, these labels can be used again: \cmpd{g.two} and \cmpd{f.one}.
\end{beispiel}

This also works, if the main name has already been used.
\begin{beispiel}
 \cmpd{a} and its variants \cmpd{a.one} and \cmpd{a.two}
\end{beispiel}

The same way the main name of combined labels can be used solely.
\begin{beispiel}
 \cmpd{f} and \cmpd{g}
\end{beispiel}

How you can create a label like \cmpd{f.{one,two}} is explained in
section~\ref{ssec:sub_ranges}.
\secidx*{Sublabel}

\subsection{Lists}\label{ssec:listen}\secidx{Lists}
There is actually more to the \cmd{cmpd} command. It also prints lists of labels.

The right description would be something like:
\begin{beschreibung}
 \Befehl{cmpd}{<(possibly comma separated list of) label name(s)>}
\end{beschreibung}
This means that with default settings the comma can't be part of the label name
unless hidden in braces. You can change that, see section~\ref{sssec:list-input-sep}
on page~\pageref{sssec:list-input-sep}. Section~\ref{sssec:input-sep-bsp} shows
an example how a changed input could look like.

\begin{beispiel}
 More than one label can be put inside \verb=\cmpd=, separated by commas. Then a
 list like \cmpd{a, b, c, e, g.two} is printed.
\end{beispiel}
The Harvard comma (see~\ref{ssec:harvard-comma}) in \code{, and} between \cmpd{e}
and \cmpd{g.two} suggests that there are options to customize the list, see
sections~\ref{ssec:trenner} and \ref{sec:language_support}.
\secidx*{Lists}

\subsection{Lists and ranges of sublabels}\label{ssec:sub_ranges}\index{Sublabel!lists}\index{Sublabel!ranges}
Sometimes it can be useful to display a label with a list or a range of sublabels.
Suppose you have compounds \cmpd{q.one,q.two,q.three,q.four,q.five} which for
example differ in their substituents. It can be useful to refer to them all at
once: \cmpd[cmpd-all]{q}.

The syntax is rather intuitive -- you just input a list of sublabels:
\begin{beispiel}
 list of labels: \cmpd{q.one, q.two, q.three, q.four, q.five}\\
 label with list of sublabels: \cmpd{q.{one,two,three,four,five}}
\end{beispiel}
Since the sublist is input with a comma in the default setting, you have to put
them into braces.

Using this syntax you also can create ranges of sublabels. As a marker to indicate
the range \chemnum uses two points \verb=..=.
\begin{beispiel}
 \cmpd{q.{two..four}}\\
 \cmpd{q.{one,three..five}}\\
 \cmpd{q.{one..three,five}}
\end{beispiel}

This means you can neither use the comma nor the two points as part of a sublabel
name. How you can use other markers for the input you can read in
sections~\ref{sssec:range-input-sep} and \ref{sssec:list-input-sep}.

\chemnum provides an option (see section~\ref{sec:cmpdsetup} for more details),
with which all sublabels of a main label can be printed as range.
\begin{beschreibung}
 \Option{cmpd-all}{\default{true}/false}
\end{beschreibung}
\begin{beispiel}
 \cmpd[cmpd-all]{q}
\end{beispiel}

In order to get the right sublabels the command needs to know, which sublabels
have been declared. This means that they're read from the \verb=\jobname.cmpd=
file (supposing your source file is called \verb=\jobname.tex=). So you may need
to compile twice (or more) until the labels are printed right.

If one uses label with sublabel ranges, one might want to use \emph{only} the
sublabels in a different place. This is possible with this option:
\begin{beschreibung}
 \Option{sub-only}{\default{true}/false}
\end{beschreibung}
\begin{beispiel}
 % uses packages `chemfig', `chemmacros' and `booktabs'
 \cmpd*{benzene.H,benzene.Me,benzene.OH,benzene.NH2}
 \chemname{\chemfig{*6(=-=-(-R)=-)}}{\cmpd[cmpd-all]{benzene}}
 \begin{tabular}{lll}
                               & \ch{-R}   & Name \\\midrule
  \cmpd[sub-only]{benzene.H}   & \ch{-H}   & Benzene \\
  \cmpd[sub-only]{benzene.Me}  & \ch{-CH3} & Toluene \\
  \cmpd[sub-only]{benzene.OH}  & \ch{-OH}  & Phenol \\
  \cmpd[sub-only]{benzene.NH2} & \ch{-NH2} & Phenylamine (Aniline)
 \end{tabular}
\end{beispiel}

Ranges of sublabels require some care. For one thing one has to know which
sublabel is the lower one and which is the higher one. Then one has to know which
labels are in between them and what do they represent. As \chemnum can't take care
for you, you need to do that for yourself if you want to avoid labels like these:
\begin{beispiel}
 \cmpd{q.{five..three}} or \cmpd{q.{three,one,four}}.
\end{beispiel}

\section{Initialize labels at once}\label{sec:cmpdinit}\secidx{Initialize labels}
The commands described in this section are not essential for the usage of \chemnum,
but provide additional functionality that some users might find useful.

\subsection{Usage}
\subsubsection{Main labels}
Perhaps you're missing \paket{chemcompounds}' command \verb=\declarecompound=
already. For one thing one has a summary of all used label names. And one may be
warned if one uses an uninitialized label name\footnote{for instance due to a
typing error}. This behaviour can be realized with these commands:
\begin{beschreibung}
 \Befehl{cmpdinit}{<comma separated list of label names>}
 \Befehl[cmpdinita]{cmpdinit*}{<comma separated list of label names>}
\end{beschreibung}
All compounds set inside \cmd{cmpdinit} are initialized and declared \emph{in the
given order}. If one uses a label name that's not in the list, \chemnum produces
a warning message.

\cmpdreset
\begin{beispiel}
 \cmpdinit{A, B, C} \cmpd{B} \cmpd{A.a} \cmpd{C} \cmpd{D}
\end{beispiel}

\begin{mdframed}
 \small\ttfamily\noindent
 {}*************************************************\\
 {}* chemnum warning: "cmpd-init"\\
 {}* \\
 {}* You used \verb=\cmpdinit= but didn't initiate compound "D"{} on line 1.\\
 {}*************************************************
\end{mdframed}

Please note that for labels with sublabels only the \emph{main name} can and
should be initialized. This initializes all compounds with the same main name.

\cmd[cmpdinita]{cmpdinit*} also produces a warning but doesn't declare the labels.
They are declared when you first use them in the order of appearance in the text.

\cmpdreset
\begin{beispiel}
 \cmpdinit*{X, Y, Z} \cmpd{Y} \cmpd{X.x} \cmpd{Z} \cmpd{W}
\end{beispiel}

\begin{mdframed}
 \small\ttfamily\noindent
 {}*************************************************\\
 {}* chemnum warning: "cmpd-init"\\
 {}* \\
 {}* You used \verb=\cmpdinit= but didn't initiate compound "W"{} on line 1.\\
 {}*************************************************
\end{mdframed}

If you need to use own labels (see page \pageref{ssec:own_labels}) but still want
to initialize all labels at once, you should use \cmd[cmpdinita]{cmpdinit*}{},
because then no labels are declared when they're initialized.

This command can be used several times, with or without \verb=*=. A useful way
would be a single usage, though, proposedly in the preamble of your document after
the \chemnum setup (see section~\ref{sec:cmpdsetup}).

\subsubsection{Sublabels}
If you'd like a warning for label-sublabel combinations as well, or want to preset
their order, you can use
\begin{beschreibung}
 \Befehl{cmpdinit}[\key{init-sub}{true}]{<comma separated list of label names>}
\end{beschreibung}
This will give you a warning in these cases, too.

\cmpdreset
\begin{beispiel}
 \cmpdinit[init-sub = true]{E.{e,f,g}}
 \cmpd{E.e, E.h}
\end{beispiel}

\begin{mdframed}
 \small\ttfamily\noindent
 {}*************************************************\\
 {}* chemnum warning: "cmpd-init-sub"\\
 {}* \\
 {}* You used \verb=\cmpdinit= and "{}init-sub = true"{} but didn't initiate sub-compound\\
 {}* "{}E.h"{} on line 1.\\
 {}*************************************************
\end{mdframed}

\subsection{Error instead of warning}
If you want \chemnum to produce an error instead of a warning when you use an
unititialized label name, you can use \cmd{cmpdinit} like this:
\begin{beschreibung}
 \Befehl{cmpdinit}[\key{init-strict}{true}]{<comma separated list ...>}
\end{beschreibung}
A usage of \key{init-strict}{false} would be the usage without optional argument.
Of course the *-version has the same option.

With \key{init-strict}{true} the \LaTeX\ run will be aborted with a corresponding
error message, if a label name is used that hasn't been initialized before.

Of course you can use \cmd{cmpdinit} with both options
\begin{beschreibung}
 \Befehl{cmpdinit}[\key{init-strict}{true}, \key{init-sub}{true}]{<comma separated list ...>}
\end{beschreibung}
or use \cmd{cmpdsetup}, see section~\ref{sec:cmpdsetup}.

\subsection{Customized input}
When you're using other tokens for \key{sub-input-sep} (page~\pageref{sssec:sub-input-sep})
or \key{sublist-input-sep} (page~\pageref{sssec:list-input-sep}), you have to
use them with \cmd{cmpdinit} as well.

A change of \key{list-input-sep} (page \pageref{sssec:list-input-sep}) doesn't
affect \cmd{cmpdinit}. If you want to change the input of \cmd{cmpdinit} you need
to use \key{init-input-sep}, see section~\ref{sssec:init-input-sep} on
page~\pageref{sssec:init-input-sep}.

\subsection{Why initiate labels?}
As mentioned in the beginning of this section the initiation of labels is not
required. There are two possible reasons, why one would like to use them, anyway.
\begin{enumerate}
 \item The use of \cmd[cmpdinita]{cmpdinit*} is a way to keep track of which labels
 you have used and gives you a warning/error message if you use a new one or just
 have mispelled an existing one. Depending on the number of labels you (have to)
 use, this can be lots of additional work, though.
 \item The use of \cmd{cmpdinit} does the same thing as \cmd[cmpdinita]{cmpdinit*}
 \emph{and} declares all the labels in the given order. So if you use it, you
 won't have to use \cmd[cmpda]{cmpd*} (section~\ref{cmd:unsichtbares_label}) and
 \cmd[cmpdb]{cmpd+} (section~\ref{cmd:cmpd_plus}). This gives you direct control,
 which number is given to which compound. 
\end{enumerate}
\secidx*{Initialize labels}

\section{Resetting the counter}\label{sec:cmpdreset}\secidx{Reset the counter}
With the command
\begin{beschreibung}
 \Befehl{cmpdreset}[<number>]
\end{beschreibung}
it is possible to reset the counter. Without argument it is reset to \verb=1=. So
\cmd{cmpdreset} and \cmd{cmpdreset}[1] are the same. Unlike most other \chemnum
commands its impact is not only inside the local \TeX\ group but resets the
counter globally.
\begin{beispiel}
 \cmpdreset The numbering starts with 1 again: \cmpd{h, i, j}
\end{beispiel}

\achtung{You should be careful with how you use this command! Usually it cannot
be recommended to reset the counter. Different compounds can get the same label!}

\begin{beispiel}
 Same label: \cmpd{a}, \cmpd{h}
\end{beispiel}
\secidx*{Reset the counter}

\section{Special usage}\secidx{Special usage}
\subsection{Moving arguments -- section titles}
Using \cmd{cmpd} inside a section title in the first instant doesn't seem to be
a problem. This doesn't give any errors and the output looks like expected:
\begin{beispiel}[code only]
 \section{Compound \cmpd[cmpd-delim]{b}}
 \ldots
\end{beispiel}
However, if you use \verb=\tableofcontents= -- which is rather probable, I
guess -- something unwanted happens: because the table of contents is at the
beginning of the document, the label is declared in the table of contents instead
of at the first appearance in the text. There are two ways to circumvent this:
\begin{enumerate}
 \item You declare your labels in the order you want with \cmd{cmpdinit}{}= (see
   page~\pageref{sec:cmpdinit}) in the preamble or the beginning of the document.
 \item You use \cmd[cmpdb]{cmpd+} to set the label. It will read the label from
   the file \verb=<jobname>.cmpd= (supposing your main file is called \verb=<jobname>.tex=).
   This means \emph{at least two}, maybe more \LaTeX\ runs are necessary until
   all labels are set.
\end{enumerate}
The same is true for any other list of moving arguments that is created \emph{before}
the actual appearance of the label in the text.

The command\label{cmd:cmpd_plus}
\begin{beschreibung}
 \Befehl[cmpdb]{cmpd+}\ma{<label name>}
\end{beschreibung}
reads the label name that is to be printed from \verb=\jobname.cmpd=. In a way
this is the opposite of \cmd[cmpda]{cmpd*} since it doesn't declare a label but
only references and prints it. So in order to use \cmd[cmpdb]{cmpd+}{<label>},
\verb=<label>= must have been declared, either with \cmd{cmpd}{<label>} or with
\cmd[cmpda]{cmpd*}{<label>}. It doesn't matter, if this declaration is done before
or after the usage of \cmd[cmpdb]{cmpd+}. This command is useful, if the label is
used inside a moving argument like in \verb=\section{}=.

\achtung{It is not needed when \cmd{cmpdinit} is used, see section~\ref{sec:cmpdinit}
(pages~\pageref{sec:cmpdinit}\,ff).}

\begin{beispiel}[code only]
 \subsubsection{Test section title with compound label \cmpd+[cmpd-delim]{f.one}}
 % with hyperref:
 \subsubsection{Test section title with compound label \texorpdfstring{\cmpd+[cmpd-delim]{f.two}}{(7b)}}
\end{beispiel}

\subsection{Floating environments}\cmpdinit*{float1,float2}
The usage of \cmd{cmpd} inside a float shouldn't be a problem: labels are declared
in the order the floating environment appears in the code. However, to be on the
safe side you can of course use \cmd[cmpdb]{cmpd+} here as well.
\begin{beispiel}[code and float]
 % preamble:
 % \usepackage{chemscheme}
 % document:
 \cmpdreset\cmpd{float1, float2}. Now inside of a scheme:
 \begin{scheme}[h]
  \centering
  \caption{This label should be a 2: \cmpd{float2}}
  If 2 = \cmpd{float2} is true, everything's fine.
 \end{scheme}
\end{beispiel}

This also works, of the floating environment is placed before the code, for
instance by using the option \verb=[t]=.

\cmpdinit*{Alc,EtherBr,cpd}
\subsection{Schemes created by external programs}\label{ssec:cmpdref}\secidx{Schemes by external programs}
If you create your schemes by external programs like for \eg \prog{ChemDraw} or
\prog{Marvin\-Sketch}, and want to use \chemnum with them as well, you can use
this command:
\begin{beschreibung}
 \Befehl{cmpdref}[<tag>]\oa{<options>}\ma{<label name>}
 \Befehl[cmpdrefa]{cmpdref+}[<tag>]\oa{<options>}\ma{<label name>}
\end{beschreibung}
This command is inspired by the \verb=\schemeref= command from the package
\paket{chemscheme}. It works in a very similar way. You create the scheme and save
it with temporary labels as an \verb=eps= file.

Like \cmd{cmpd+} the plussed variant only references a label but does not
declare it.

\begin{beispiel}[code and float]
 \begin{scheme}[ht]
  \centering
  \includegraphics{scheme-tmp.ps}
  \caption{Scheme with temporary markers.}
 \end{scheme}
 \begin{scheme}[ht]
  \centering
  \cmpdref{Alc} % replaces TMP1
  \cmpdref{EtherBr} % replaces TMP2
  % \cmpdref{third} would replace TMP3
  \includegraphics{scheme-tmp.ps}
  \caption{Scheme with automated labels.}
 \end{scheme}
\end{beispiel}
So you nummerate the compounds with \textsf{TMP1}, \textsf{TMP2} \etc These markers
will then be replaced with the corresponding labels. Like \paket{chemscheme} \chemnum
uses the \cmd{psfrag} command of the \paket{psfrag} package to do that. This means
you have to either compile via \prog{latex}, \prog{dvips}, \prog{ps2pdf}, or for
instance use the package \paket{auto-pst-pdf} to be able to use \hologo{pdfLaTeX}.
It doesn't work with \hologo{XeLaTeX}, neither probably with \hologo{LuaLaTeX},
the author didn't test that, though.

In each case it is important that the marker is saved as \emph{text} in the \verb=eps=
file.

You can customize the labels with the options presented in section~\ref{sec:cmpdsetup}.
\begin{beispiel}[code and float]
 \begin{scheme}[ht]
  \centering
  \cmpdsetup{ref-style=\color{green}}
  \cmpdref{Alc} % replaces TMP1
  \cmpdref{EtherBr} % replaces TMP2
  \includegraphics{scheme-tmp.ps}
  \caption{Scheme with automated labels.}
 \end{scheme}
\end{beispiel}

You also can replace arbitrary text by using the second optional argument.
\begin{beispiel}[code and float]
 \begin{scheme}[ht]
  \centering
  \includegraphics{scheme-bla.ps}
  \caption{Scheme with arbitrary markers.}
 \end{scheme}
 \begin{scheme}[ht]
  \centering
  \cmpdref[blah]{Alc}
  \cmpdref[blub]{EtherBr}
  % \cmpdref{third} would replace TMP1
  \includegraphics{scheme-bla.ps}
  \caption{Scheme with explicitly placed labels.}
 \end{scheme}
\end{beispiel}

Of course sublabels work in the known way:
\begin{beispiel}[code and float]
 \begin{scheme}[ht]
  \centering
  \cmpdref{cpd.red}
  \cmpdref{cpd.ox}
  \includegraphics{scheme-tmp.ps}
  \caption{Scheme with automated labels and sublabels.}
 \end{scheme}
\end{beispiel}
\cmpdreset[10]\secidx*{Schemes by external programs}
\secidx*{Special usage}

\section{Usage with \paket{hyperref}}\label{sec:hyperref}\secidx[Usage with hyperref]{Usage with \textsf{hyperref}}
\subsection{Basics}
If you are using the package \paket{hyperref} you can make hyperlinked labels in
your \prog{pdf}.
\begin{beschreibung}
 \Option{hyper-use}{\default{true}/false}
\end{beschreibung}
Setting this option (see section~\ref{sec:cmpdsetup} for more information) to
\code{true}, the next usage of a label will create a link target and the ones
after that will create links to this target. This only works if the label does
\emph{not} contain a list or range of sublabels. Also a label set with \cmd[cmpda]{cmpd*}
can't be marked as target.

\begin{beispiel}
 \cmpdsetup{hyper-use}
 The first time targets are created: \cmpd{a.two,b,c,d,e,f}. Then links are created:
 \cmpd{a.two,d,e}.
 
 This doesn't work with labels containing a list of sublabels: \cmpd{a,f.{one,two},c}.
 Since \cmpd{a.two} and \cmpd{a} are different labels they have different link targets.
\end{beispiel}

\subsection{Selected usage}
Maybe the automated use isn't what you want or need. Maybe is is enough to create
a hyperlink in one or two places. This can be achieved with two options:
\begin{beschreibung}
 \Option{hyper-target}{<link name>}
 \Option{hyper-link}{<link name>}
\end{beschreibung}
With \key{hyper-target} you can set the target. \textbf{Caution}: if you use this
option with a list \emph{every} label of the list will get the same link name and
\paket{hyperref} will warn you:
\begin{mdframed}\ttfamily\small
 destination with the same identifier (name\{<link name>\}) has been already
 used, duplicate ignored
\end{mdframed}

\key{hyper-link} creates a link for the denominated target. \textbf{Caution}: if
you use this option with a list \emph{every} label of the list will refer to the
same target.
\begin{beispiel}
 Create a link \cmpd[hyper-link=mylink,cmpd-all]{f} but don't forget the
 corresponding target: \cmpd[hyper-target=mylink,cmpd-all]{f}.
\end{beispiel}

If you use \key{hyper-link} without the corresponding \key{hyper-target}
\paket{hyperref} will issue a warning:
\begin{mdframed}\ttfamily\small
 name\{<link name>\} has been referenced but does not exist, replaced by a
 fixed one
\end{mdframed}
\secidx*[Usage with hyperref]{Usage with \textsf{hyperref}}

\section{Customization}\label{sec:cmpdsetup}
\subsection{The principle}
\chemnum is customized with a key-value system. They can either be used for all
following commands with
\begin{beschreibung}
 \Befehl{cmpdsetup}{key1 = value1, key2 = value2}
\end{beschreibung}
or only for a single command directly with
\begin{beschreibung}
 \Befehl{cmpd}[key1 = value1, key2 = value2]{<label name>}
\end{beschreibung}

If you use keys without values like \cmd{cmpdsetup}{key1,key2}, default values\label{opt:default}
are used. These are \emph{not} necessarily empty are \emph{not} necessarily the
default setting of \chemnum.

The options are divided in to different classes\label{key_classes}, which indicate
the scope they're used for. They all follow the form \verb+<class>-option = <value>+:
\begin{itemize}
 \item \code{cmpd}: These options change the whole label or the main label.
 \item \code{sub}, \code{sublist}, \code{subref}: These options have an impact on sublabels.
 \item \code{list}: Options which change the output of lists.
 \item \code{ref}: Options which have an impact on \cmd{cmpdref}.
 \item \code{hyper}: Options which have an impact on the usage with \paket{hyperref}.
 \item \code{init}: Options which have an impact in the initialization of the labels.
\end{itemize}

\subsection{Font style}\label{ssec:schriftstil}\secidx{Font style}
With the keys
\begin{beschreibung}
 \Option{cmpd-style}{<style>}
 \option{cmpd-weight}{bold/normal} Default = \code{bold}
\end{beschreibung}
you can change the style with which the labels are printed.

\begin{beispiel}
 % preamble:
 % \usepackage[normalem]{ulem}
 % document:
 \cmpd{a, b, f.two, k}\\                      % default
 \cmpd[cmpd-style = \uline]{a, b, f.two, k}\\ % underlined
 \cmpd{a, b, f.two, k}\\                      % default again
 \cmpdsetup{cmpd-style = \itshape\uline}
 \cmpd{a, b, f.two, k}\\                      % underlined and in italics
 \cmpd[cmpd-weight=normal]{a, b, f.two, k}
\end{beispiel}
The separate handling of the bold face is because of the greek labels (see
section~\ref{ssec:counter}). Depending on the package option \code{textgreek},
the bold face must realized differently.
\secidx*{Font style}

\subsection{Own labels}\label{ssec:own_labels}\secidx{Own labels}
With the key
\begin{beschreibung}
 \Option{cmpd-label}{<label>}
\end{beschreibung}
one can choose the label associated with the label name.

\begin{beispiel}
 Own label: \cmpd[cmpd-label = XYZ]{l}; in a list \emph{all new compounds} get
 this label! \cmpd[cmpd-label = XYZ]{l, a, m}
\end{beispiel}
\secidx*{Own labels}

\subsection{Markers in images}\secidx{Markers in images}\label{ssec:marker_in_bildern}
There are several options that affect labels in external images placed with \cmd{cmpdref}.
\begin{beschreibung}
 \option{ref-marker}{<marker>} This changes the default marker that is to be
   replaced. Default = \code{TMP}
 \option{ref-pos}{\{<tex-pos>\}\{<ps-pos>\}} Default = \code{bb}
 \option{ref-style}{<style>} change the format of the labels in external schemes.
   Default = \verb=\sffamily=
\end{beschreibung}
With \key{ref-pos} you can change the default way how \cmd{cmpdref} places the
label. The command \cmd{psfrag} that is used internally has two options for that.
The first one determines the position of the \TeX-replacement, the second one how
it is placed in the box from the marker that is replaced. For details please refer
to the documentation of \paket{psfrag}.
\secidx*{Markers in images}

\subsection{Counter}\label{ssec:counter}\secidx{Counter}
You can change the format of the counters, too, if you like.
\begin{beschreibung}
 \option{cmpd-counter}{arabic/alph/Alph/greek/Greek/roman/Roman/Symbol} Default = \code{arabic}
 \option{sub-counter}{arabic/alph/Alph/greek/Greek/roman/Roman/Symbol} Default = \code{alph}
\end{beschreibung}

Please note: a change of format only affects labels \emph{which are not yet declared}!
\begin{beispiel}
 \cmpdsetup{cmpd-counter = Alph, sub-counter = arabic}
 \cmpd{a, b, f.two, k}\\ % no affect with existing labels
 \cmpd{n.one,n.two,o.one}\\
 \cmpd{f.three,f.four}\\ % be careful: the main label already exists and isn't changed!
 \cmpd[cmpd-counter,sub-counter = greek]{p.one,p.two}
\end{beispiel}
\secidx*{Counter}

\subsection{Delimiters}\secidx{Delimiters}
Both single labels and whole lists can get delimiters. This is done with the options
\begin{beschreibung}
 \option{cmpd-delim}{<odelim><cdelim>} values need \emph{two} tokens!
 \option{list-delim}{<odelim><cdelim>} values need \emph{two} tokens!
\end{beschreibung}
\begin{beispiel}
 \cmpd{a, b, f.two, k}\\                  % default
 \cmpd[cmpd-delim = ()]{a, b, f.two, k}\\ % in braces
 \cmpd{a, b, f.two, k}\\                  % default again
 \cmpdsetup{cmpd-delim = ()}
 \cmpd{a, b, f.two, k}                    % in braces
\end{beispiel}
Please be aware that the default values (see pages~\pageref{opt:default} and
\pageref{ssec:keys_uebersicht}) of \key{cmpd-delim} and \key{list-delim} and the
default setting of \chemnum are \emph{not} the same. By default \chemnum doesn't
use delimiters. Also please note that the list delimiters only are used if \cmd{cmpd}
contains at least two label names.
\begin{beispiel}
 \cmpd{a, b, f.two, k}\\                  % default
 \cmpd[list-delim = {[}{]}]{a, b, f.two, k}\\ % in braces
 \cmpd{a, b, f.two, k}\\                  % default again
 \cmpdsetup{list-delim = []}
 \cmpd{a, b, f.two, k}\\                  % in braces
 \cmpd{a}                                 % NOT a list!
\end{beispiel}
If you generally use labels in braces but want to use a single one without braces,
you have several possibilities:
\begin{beispiel}
 \cmpdsetup{cmpd-delim = ()}
 normally: \cmpd{b}, \cmpd{c}, \cmpd{d.one}, but sometimes so \cmpd[cmpd-delim = ]{e}
 or so \cmpd-{e}.
\end{beispiel}
With
\begin{beschreibung}\label{cmd:ohne_begrenzer}
 \Befehl[cmpdc]{cmpd-}\ma{<label name>}
 \Befehl[cmpdrefc]{cmpdref-}\ma{<label name>}
\end{beschreibung}
the braces of single labels can easily be removed (but not the ones of a list).
\begin{beispiel}
 \cmpd{a, b, f.two, k}\\  % default
 \cmpdsetup{cmpd-delim = (), list-delim = []}
 \cmpd{a, b, f.two, k}\\  % braced twice
 \cmpd-{a, b, f.two, k}   % braced once
\end{beispiel}

Using \cmd{cmpdsetup}{\key{cmpd-delim}{ }} and \cmd{cmpdsetup}{\key{list-delim}{ }}
will restore the default behaviour.
\secidx*{Delimiters}

\subsection{Suffix and prefix}\secidx{Suffix and prefix}
If you want to, you can also add a prefix and/or a suffix to a compound and/or a
list.
\begin{beschreibung}
 \Option{cmpd-prefix}{<prefix>}
 \Option{cmpd-suffix}{<prefix>}
 \Option{list-prefix}{<prefix>}
 \Option{list-suffix}{<prefix>}
\end{beschreibung}
The same rule as for the delimiters applies: list attributes are only used with
lists, \ie at least with two items.

An example for the label attributes:
\begin{beispiel}
 \cmpd{a, b, f.two, k}\\ % default
 \cmpd[cmpd-prefix = Nr.]{a, b, f.two, k}\\ 
 \cmpdsetup{cmpd-prefix = \(\rightarrow\)}
 \cmpd{a, b, f.two, k}\\
 \cmpd{a}             % NOT a list!
\end{beispiel}
An example for the list attributes:
\begin{beispiel}
 \cmpd{a, b, f.two, k}\\ % default
 \cmpd[list-prefix = list:]{a, b, f.two, k}\\ 
 \cmpdsetup{list-prefix = collection:}
 \cmpd{a, b, f.two, k}\\
 \cmpd{a}             % NOT a list!
\end{beispiel}

Between prefix, label and suffix there is a space inserted as default behaviour:
\begin{beispiel}
 \cmpd[cmpd-suffix=sx,cmpd-prefix=px]{k}
\end{beispiel}
It can be changed with
\begin{beschreibung}
 \Option{cmpd-space}{<space>}
\end{beschreibung}
\begin{beispiel}
 \cmpd[cmpd-suffix=sx,cmpd-prefix=px,cmpd-space={}]{k}
\end{beispiel}
\secidx*{Suffix and prefix}

\subsection{Separators}\label{ssec:trenner}\secidx{Separators}
\subsubsection{Labels -- input}\label{sssec:sub-input-sep}\index{Separators!labels!input}
As a default setting \chemnum use the point \code{.} as a marker to separate main
labels from sublabels. You can change that as you like. Tokens you \emph{can't use}
are \verb=%=, \verb=,= and \verb=#= and \emph{you shouldn't} use \verb=@=. Furthermore
you should be careful to use different symbols for the other separators (see the
next sections).
\begin{beschreibung}
 \option{sub-input-sep}{<separator>} Default = \code{.}
\end{beschreibung}
\begin{beispiel}
 \cmpdsetup{sub-input-sep = !}
 \cmpd{f!one, g!two}\\
 \cmpd[sub-input-sep= +]{f+one, g+two}\\
 \cmpd[sub-input-sep= ~]{f~one, g~two}\\
 \cmpd[sub-input-sep= &]{f&one, g&two}\\
 \cmpd[sub-input-sep= *]{f*one, g*two}
\end{beispiel}
You should -- not only for the sake of consistency -- decide only \emph{once} in
the beginning of your document \emph{before} the \verb=\tableofcontents= or in
the preamble, which marker you use. Otherwise, the \verb=\tableofcontents= won't
know the marker anymore and your labels might be at least displayed wrong in the
table of contents. Also you need to setup the marker \emph{before} you use it with
\cmd{cmpd} or \cmd{cmpdinit} etc.

\subsubsection{Labels -- output}\index{Separators!labels!output}
You can separate the labels from the sublabels if you like to.
\begin{beschreibung}
 \Option{sub-output-sep}{<separator>}
\end{beschreibung}
\begin{beispiel}
 \cmpd{a, b, f.two, f.three, k}\\ % default
 \cmpd[sub-output-sep = -]{a, b, f.two, f.three, k}\\
 \cmpdsetup{sub-output-sep = $\cdot$}
 \cmpd{a, b, f.two, f.three, k}\\
 \cmpd[sub-output-sep = :]{a, b, f.two, f.three, k}
\end{beispiel}

\subsubsection{Sublabels -- input}\label{sssec:range-input-sep}\index{Separators!subabels!input}
As a default setting \chemnum use two points \code{..} as a marker to detect two
sublabels as end and beginning of a range. But you can choose any tokens you like
to do that. Tokens you \emph{can't use} are \verb=%=, \verb=,= and \verb=#= and
\emph{you shouldn't} use \verb=@=. Furthermore you should be careful to use different
symbols for the other separators.
\begin{beschreibung}
 \option{subrange-input-sep}{<separator>} Default = \code{..}
\end{beschreibung}
\begin{beispiel}
 \cmpd[subrange-input-sep=:]{q.{one:four}}\\
 \cmpdsetup{subrange-input-sep=--}
 \cmpd{q.{one--four}}
\end{beispiel}

\subsubsection{Sublabels -- output}\index{Separators!subabels!output}
Maybe you don't like the output of \cmpd[cmpd-all]{q} and you want it displayed
differently? Then you need this key:
\begin{beschreibung}
 \option{subrange-output-sep}{<separator>} Default = \verb|--|
\end{beschreibung}
\begin{beispiel}
 \cmpd[subrange-output-sep={ {\normalfont to} }]{q.{one..four}}\\
 \cmpdsetup{subrange-output-sep=-}
 \cmpd{q.{one..four}}
\end{beispiel}

\subsubsection{Lists -- input}\label{sssec:list-input-sep}\index{Separators!lists!input}
If you like you can change the list input separator for both main label lists and
lists of sublabels. In both cases \chemnum uses the comma \code{,} as default. You
can change that as you like. Tokens you \emph{can't use} are \verb=%=, \verb=,=
and \verb=#= and \emph{you shouldn't} use \verb=@=. Furthermore you should be
careful to use different symbols for the other separators.
\begin{beschreibung}
 \option{list-input-sep}{<separator>} Default = \code{,}
 \option{sub-input-sep}{<separator>} Default = \code{,}
\end{beschreibung}
\begin{beispiel}
 \cmpd{a, b, f.two, k}\\ % default
 \cmpd[list-input-sep = +]{a + b + f.two + k}\\
 \cmpd{a, b, f.two, k}\\
 \cmpdsetup{list-input-sep = ;}
 \cmpd{a; b; f.two; k}
\end{beispiel}
With this you can for example use IUPAC names as label names\footnote{Thanks to
Christina L\"udigk who in a way suggested this feature.}.
\begin{beispiel}
 \cmpdsetup{list-input-sep=;}
 \cmpd{1,3-dichlorocyclohexanole}
\end{beispiel}

\subsubsection{Lists -- output}\label{sssec:list-output-sep}\index{Separators!lists!output}
You can change the separator symbol with which lists are printed as well. \chemnum
uses the comma for both cases in the default setting.
\begin{beschreibung}
 \option{list-output-sep}{<separator>} Default = \code{,}
 \option{sub-output-sep}{<separator>} Default = \code{,}
\end{beschreibung}
\begin{beispiel}
 \cmpd{a, b, f.two, k}\\ % default
 \cmpd[list-output-sep = ]{a, b, f.two, k}\\
 \cmpd{a, b, f.two, k}\\
 \cmpdsetup{list-output-sep = ;}
 \cmpd{a, b, f.two, k}
\end{beispiel}

\subsubsection{Lists -- last separator}\label{sssec:last-sep}\index{Separators!list!last seperator}
The last separator in the main label list is a special case. \chemnum provides
two options with which it can be customized.
\begin{beschreibung}
 \Option{list-last-sep}{<separator>}
 \option{list-lang}{US/GB/DE/FR/ES/IT} Default = \code{US}
\end{beschreibung}

\begin{beispiel}
 \cmpd[list-lang = US]{a, b, f.two, k} \cmpd[list-lang = US]{a, b}\\
 \cmpd[list-lang = GB]{a, b, f.two, k} \cmpd[list-lang = GB]{a, b}\\
 \cmpd[list-lang = DE]{a, b, f.two, k} \cmpd[list-lang = DE]{a, b}\\
 \cmpd[list-lang = FR]{a, b, f.two, k} \cmpd[list-lang = FR]{a, b}\\
 \cmpd[list-lang = ES]{a, b, f.two, k} \cmpd[list-lang = ES]{a, b}\\
 \cmpd[list-lang = IT]{a, b, f.two, k} \cmpd[list-lang = IT]{a, b}
\end{beispiel}
As you can see the language \key{list-lang}{US} uses the Harvard comma, all other
languages don't. You'll find more on language support and the Harvard comma in
section~\ref{sec:language_support}.

With the key \key{list-last-sep} you also can set the last separator individually.
\begin{beispiel}
 \cmpdsetup{list-lang = GB}%
 \cmpd[list-last-sep = {and also}]{a, b, f.two, k}\\
 \cmpd[list-last-sep = ]{a, b, f.two, k}\\
 \cmpd[list-last-sep = {as well as}]{a, b, f.two, k}\\
 \cmpd[list-last-sep = empty]{a, b, f.two, k}
\end{beispiel}

\key{list-last-sep}{empty} is a special value. If it is used, the separator set
with \key{list-output-sep} is used for the whole list.
\begin{beispiel}
 \cmpdsetup{list-last-sep = empty}%
 \cmpd{a, b, f.two, k}\\
 \cmpd[list-output-sep = ;]{a, b, f.two, k}\\
 \cmpd[list-output-sep = {\ und}]{a, b, f.two, k}
\end{beispiel}

\subsubsection{\code{\textbackslash\textcolor{code}{cmpdinit}} -- input}\label{sssec:init-input-sep}\index{Separators!cmpdinit@\textcolor{code}{\code{\textbackslash cmpdinit}}}
When you've changed the separator for \cmd{cmpd} you might want to change it for
\cmd{cmpdinit} as well. \chemnum uses the comma in the default setting. You can
change that as you like. Sie können aber eine beliebige andere Eingabe wählen.
Tokens you \emph{can't use} are \verb=%=, \verb=,= and \verb=#= and \emph{you
shouldn't} use \verb=@=.
\begin{beschreibung}
 \option{init-input-sep}{<separator>} Default = \code{,}
\end{beschreibung}

\subsubsection{An example}\label{sssec:input-sep-bsp}
A small example how a customized input could look like:
\begin{beispiel}
 \cmpdsetup{
   sub-input-sep      = ! ,
   subrange-input-sep = : ,
   sublist-input-sep  = / ,
   list-input-sep     = +
 }
 \cmpd{a + f ! one + q ! one / three : five}
\end{beispiel}
\secidx*{Separators}

\section{Language support}\label{sec:language_support}\secidx{Language support}
\subsection{Cooperation with \textsf{babel}}
\chemnum supports the language choice made with the \paket{babel} package for a
number of languages. This is done with the \verb=\GetTranslation= command of the
\paket*{translations}\footnote{\paket*{translations} is part of the
\paket*{exsheets} bundle.} package.

\begin{beispiel}
 \selectlanguage{ngerman}DE: \cmpd{r,s,t}\\
 \selectlanguage{english}EN: \cmpd{r,s,t}\\
 \selectlanguage{british}GB: \cmpd{r,s,t}\\
 \selectlanguage{american}US: \cmpd{r,s,t}\\
 \selectlanguage{french}FR: \cmpd{r,s,t}\\
 \selectlanguage{spanish}ES: \cmpd{r,s,t}\\
 \selectlanguage{italian}IT: \cmpd{r,s,t}
\end{beispiel}
If you find a language missing, please send me an
email\footnote{\href{mailto:contact@mychemistry.eu}{contatc@mychemistry.eu}} with
the corresponding \textsf{babel} option and the right translation, and I will
gladly add it.

\subsection{Harvard comma}\label{ssec:harvard-comma}\secidx{Harvard comma}
In American English the so-called Harvard comma is quite commonly used. In British
English and other European languages it is rather uncommon.
\begin{zitat}[\href{http://en.wikipedia.org/wiki/Serial_comma}{Wikipedia}, 2011/11/07]
 The serial comma (also known as the Oxford comma or Harvard comma, and sometimes
 referred to as the series comma) is the comma used immediately before a coordinating
 conjunction (usually and or or, and sometimes nor) preceding the final item in a
 list of three or more items. For example, a list of three countries can be
 punctuated as either ``Portugal, Spain, and France'' (with the serial comma) or as
 ``Portugal, Spain and France'' (without the serial comma).

 Opinions vary among writers and editors on the usage or avoidance of the serial
 comma. In American English, the serial comma is standard usage in non-journalistic
 writing that follows the \name{Chicago Manual of Style}. Journalists, however,
 usually follow the AP Stylebook, which advises against it. It is used less often
 in British English, where it is standard usage to leave it out, with some notable
 exceptions such as \name{Fowler's Modern English Usage}. In many languages (e.g.,
 French, German, Italian, Polish, Spanish), the serial comma is not the norm and
 may even go against punctuation rules. It may be recommended in many cases,
 however, to avoid ambiguity or to aid prosody.
\end{zitat}
\chemnum uses the Harvard comma when American English is used and doesn't, when
other languages are used. However, you have the possibility to disable or enable
the Harvard comma as you wish:

\begin{beschreibung}
 \Option{list-serial-comma}{\default{true}/false}
\end{beschreibung}

\begin{beispiel}
 \selectlanguage{ngerman}DE: \cmpd[list-serial-comma]{r,s,t} vs. \cmpd{r,s,t}\\
 \selectlanguage{american}US: \cmpd[list-serial-comma=false]{r,s,t} vs. \cmpd{r,s,t}
\end{beispiel}
If you choose the language manually, please note that this resets the default
behaviour for that language. This means you have to use \key{list-serial-comma}
\emph{after} \key{list-lang}.
\begin{beispiel}
 wrong: \cmpdsetup{list-serial-comma=false, list-lang=US}\cmpd{r,s,t}\\
 right: \cmpdsetup{list-lang=US, list-serial-comma=false}\cmpd{r,s,t}
\end{beispiel}
\secidx*{Harvard comma}\secidx*{Language support}

\section{Overview}
\subsection{Options}\label{ssec:keys_uebersicht}
In this section all options are listed.
\begin{center}
 \label{tab:keys}
 \begin{longtabu*}{>{\ttfamily\color{key}}l>{\ttfamily}l>{\ttfamily}l>{\ttfamily}l}\toprule
  \normalfont\bfseries\normalcolor option &
  \normalfont\bfseries default &
  \normalfont\bfseries predefined &
  \normalfont\bfseries type \\\midrule
  cmpd-style               &                           &                           & macro   \\
  cmpd-weight              & bold                      & bold                      & choice  \\
  cmpd-label               &                           &                           & literal \\
  cmpd-delim\color{black}$^\mathrm{a}$  & ()           &                           & literal \\
  cmpd-odelim              &                           &                           & literal \\
  cmpd-cdelim              &                           &                           & literal \\
  cmpd-prefix              &                           &                           & literal \\
  cmpd-suffix              &                           &                           & literal \\
  cmpd-space               & \verb=\penalty\@m\ = & \verb=\penalty\@m\ = & literal \\
  cmpd-counter             & arabic                    & arabic                    & choice  \\
  cmpd-all                 & true                      & false                     & boolean \\\midrule
  sub-only                 & true                      & false                     & boolean \\
  sub-counter              & alph                      & alph                      & choice  \\
  sub-output-sep           &                           & ,                         & literal \\
  sub-input-sep            & .                         & .                         & literal \\
  sublist-output-sep       & ,                         & ,                         & literal \\
  sublist-input-sep        & ,                         & ,                         & literal \\
  subrange-output-sep      & --                        & --                        & literal \\
  subrange-input-sep       & ..                        & ..                        & literal \\\midrule
  list-delim\color{black}$^\mathrm{a}$  & ()           &                           & literal \\
  list-odelim              &                           &                           & literal \\
  list-cdelim              &                           &                           & literal \\
  list-prefix              &                           &                           & literal \\
  list-suffix              &                           &                           & literal \\
  list-input-sep           & ,                         & ,                         & literal \\
  list-output-sep          & ,                         & ,                         & literal \\
  list-lang                & US                        & US$^\mathrm{b}$           & choice  \\
  list-last-sep            & and$^\mathrm{c}$          &                           & literal \\
  list-serial-comma        & true                      & true$^\mathrm{c}$         & boolean \\\midrule
  ref-tag                  & TMP                       & TMP                       & literal \\
  ref-style                &                           & \verb=\sffamily=          & macro   \\
  ref-pos\color{black}$^\mathrm{a}$     & bb           & bb                        & literal \\\midrule
  init-strict              & true                      & false                     & boolean \\
  init-sub                 & true                      & false                     & boolean \\
  init-input-sep           & ,                         & ,                         & literal \\\midrule
  hyper-use                & true                      & false                     & boolean \\
  hyper-target             &                           &                           & literal \\
  hyper-link               &                           &                           & literal \\\bottomrule
 \end{longtabu*}
 \setlength\extrarowheight{-1pt}
 \begin{tabular}{>{\footnotesize}r>{\footnotesize\raggedright}p{.7\linewidth}}
  a & value needs \emph{two} tokens. \tabularnewline
  b & depends in the usage of \paket*{babel}. \tabularnewline
  c & depends on the value of \key{list-lang} and the usage of \paket*{babel}.
 \end{tabular}
\end{center}

\subsection{Commands}
In this section all commands provided by \chemnum as well as their variants are
listed.
\begin{center}
 \label{tab:commands}
 \begin{tabu*} to \linewidth {X[l]X[p]}\toprule
  \bfseries command                                 & \bfseries description \tabularnewline\midrule
  \cmd*{cmpd}[<keyval>]{<labels>}                   & basic command, declares
    and prints labels or lists of labels, see pages \pageref{sec:nummerieren}ff. \tabularnewline
  \cmd*[cmpda]{cmpd*}{<labels>}                     & invisible, declares label,
    see page \pageref{cmd:unsichtbares_label} \tabularnewline
  \cmd*[cmpdc]{cmpd-}[<keyval>]{<labels>}           & without delimiters, see
    page \pageref{cmd:ohne_begrenzer} \tabularnewline
  \cmd*[cmpdd]{cmpd+-}[<keyval>]{<labels>}          & label is read from auxiliary
    file, see page \pageref{cmd:cmpd_plus} \tabularnewline\midrule
  \cmd*{cmpdinit}[<keyval>]{<labels>}               & initialise and declare labels,
    see pages \pageref{sec:cmpdinit}f. \tabularnewline
  \cmd*[cmpdinita]{cmpdinit*}[<keyval>]{<labels>}   & initialise labels, see
    pages \pageref{sec:cmpdinit}f. \tabularnewline\midrule
  \cmd*{cmpdref}[<keyval>]\oa[<marker>]\ma{<label>} & replace temporary markers
    in \code{eps} files by labels, see page \pageref{ssec:cmpdref} \tabularnewline
  \cmd*[cmpdrefc]{cmpdref-}[<keyval>]\oa[<marker>]\ma{<label>} & replace temporary
    markers in \code{eps} files by labels without delimiters, see page \pageref{cmd:ohne_begrenzer} \tabularnewline\midrule
  \cmd*{cmpdreset}[<num>]                           & reset counter, see
    page \pageref{sec:cmpdreset} \tabularnewline\midrule
  \cmd*{cmpdsetup}{<keyval>}                        & setup \chemnum, see
    pages \pageref{sec:cmpdsetup}ff. \tabularnewline\bottomrule
  \end{tabu*}
\end{center}

\appendix
\addsec{Credits}
I would like to thank Joseph \name{Wright} and Russell \name{Hewitt} who provided
me with valuable feedback and suggestions to improve \chemnum.

Many thanks also to Jonas \name{Rivetti} who provided the Italian translation.

\printindex

\end{document}