\documentclass{book}

\usepackage{datagidx}

\usepackage[colorlinks]{hyperref}

\usepackage{fmtcount}

\DTLgidxAddLocationType{numberstringnum}

 % Define some convenient commands

\newcommand*{\principleloc}[1]{\textbf{#1}}
\newcommand*{\glossaryloc}[1]{\textit{#1}}
\newcommand*{\appname}[1]{\texttt{#1}}

 % Don't let \appname interfere with the sorting and labelling of
 % terms:

\appto\newtermlabelhook{\let\appname\DTLgidxNoFormat}

 % (datagidx also does this for \MakeUppercase, \MakeTextUppercase,
 % \MakeLowercase, \MakeTextLowercase, \textsc and \acronymfont.)

 % Make an index database labelled 'index' with given title.
 % (Omit optional argument to affect all indexing databases
 % or use a comma-separated list of labels for each database
 % to affect.)

\newgidx[heading=\chapter]{index}{Index}

 % Add custom fields to the 'index' database

\newtermaddfield[index]{Ed}{ed}{\field{text}ed}

\newtermaddfield[index]{Ing}{ing}{\field{text}ing}

\newtermaddfield[index]{AltPlural}{altplural}{}% no default value

 % Define some convenience commands to access these fields

\newcommand*{\glsed}[1]{\useentry{#1}{Ed}}
\newcommand*{\glsing}[1]{\useentry{#1}{Ing}}
\newcommand*{\glsaltpl}[1]{\useentry{#1}{AltPlural}}

 % Make a glossary database labelled 'glossary' with given title.

\newgidx[heading=\section]{glossary}{Glossary}

 % Make an acronym database labelled 'acronyms' with given title.
 % A different heading for the table of contents is supplied.

\newgidx[heading={\section[Acronyms]}]{acronyms}{List of Acronyms}

 % Make a notation database labelled 'notation' with given title

\newgidx[heading=\section]{notation}{Notation}

 % Define some terms for the index

 % Either use "database" key in optional argument
 % or set the default database. It's less typing to set the default
 % database, so let's do that:

\DTLgidxSetDefaultDB{index}

 % These definitions will add terms to the "index" database:

\newterm{singular}
\newterm{plural}
\newterm
 [% options
   label=pluralalt,
   parent=plural,
   see={altplural}
 ]
 {alternative}
\newterm[label=altplural]{alternative plural}
\newterm{group}
\newterm[see=group]{scope}
\newterm[label={ind-glossary}]{glossary}
\newterm
 [%
   label={ind-index},
   seealso={ind-glossary},
 ]
 {index}
\newterm
 [
   seealso={ind-glossary}
 ]
 {acronym}
\newterm[parent=acronym]{reset}
\newterm
 [%
   parent=acronym,
   label={firstuse},
 ]
 {first use}
\newterm
 [%
   parent=acronym,
   label={acronymlist},
   text={list of acronyms},
   plural={lists of acronyms}
 ]
 {list}

 % Be careful if a value contains a comma or an equal sign. The
 % entire value must be grouped.
\newterm
 [%
   label={comma},
   sort={,},
   text={comma (,)},
   plural={commas (,)}
 ]%
 {, (comma)}

\newterm
 [%
   label={equals},
   sort={=},
   text={equal sign (=)},
   plural={equal signs (=)},
 ]%
 {= (equal sign)}

 % Don't need to worry about makeindex's special characters (since
 % we're not using makeindex!)
\newterm
 [%
   label={"},
   sort={"},
   text={double quote (")},
   plural={double quotes (")},
 ]%
 {" (double quote)}

\newterm
 [%
   label={!},
   sort={!},
   text={exclamation mark (!)},
   plural={exclamation marks (!)},
 ]%
 {! (exclamation mark)}

\newterm
 [%
   label={|},
   sort={|},
   text={vertical bar (\textbar)},
   plural={vertical bars (\textbar)},
 ]%
 {\textbar\ (vertical bar)}


 % Be careful of special characters

\newterm
 [%
   label={amp},
   sort={\DTLgidxStripBackslash{\&}},
   text={ampersand (\&)},
   plural={ampersands (\&)},
 ]
 {\& (ampersand)}

\newterm
 [%
   label={underscore},
   sort={\DTLgidxStripBackslash{\_}},
   text={underscore (\_)},
   plural={underscores (\_)},
 ]
 {\_ (underscore)}

\newterm
 [%
   label={dollar},
   sort={\DTLgidxStripBackslash{\$}},
   text={dollar (\$)},
   plural={dollars (\$)},
 ]
 {\$ (dollar)}

\newterm
 [%
   label={circum},
   sort={\DTLgidxStripBackslash{\^}},
   text={circumflex (\textasciicircum)},
   plural={circumflexes (\textasciicircum)},
 ]
 {\textasciicircum\ (circumflex)}

\newterm
 [%
   label={tilde},
   sort={\string~},
   text={tilde (\textasciitilde)},
   plural={tildes (\textasciitilde)},
 ]
 {\textasciitilde\ (tilde)}

\newterm
 [
   label={0},
   sort={0},
 ]
 {0 (zero)}

\newterm
 [
   label={1},
   sort={1},
 ]
 {1 (one)}

\newterm
 [
   label={2},
   sort={2},
 ]
 {2 (two)}

\newterm
 [
   label={3},
   sort={3},
 ]
 {3 (three)}

 % Let's index a person
\newterm
 [%
   label={knuth},
   text={Knuth},
 ]
 {Knuth, Donald E.}

 % and have a few more entries in the same letter group to test case
 % ordering:
\newterm{kite}
\newterm{koala}

 % Earlier I modified \newtermlabelhook so
 % that \appname won't interfere with the sorting
 % and labelling mechanism, so I don't need to specify separate sort
 % and label keys here:

\newterm{\appname{makeindex}}% label and sort both set to just 'makeindex'
\newterm{\appname{xindy}}% label and sort both set to just 'xindy'

 % Define some terms for the glossary

\DTLgidxSetDefaultDB{glossary}

\newterm[description={sea mammal with flippers that eats fish}]{seal}
\newterm[description={large seal},seealso={seal}]{sea lion}

\newterm{bravo}
\newterm
  [%
    label=bravocry,
    description={cry of approval (pl.\ bravos)},
    parent=bravo
  ]
  {bravo}
\newterm
  [% options
    label=bravokiller,
    description={hired ruffian or killer (pl.\ bravoes)},
    plural=bravoes,
    parent=bravo
  ]
 {bravo}

\newterm
 [%
   plural={indices\glsadd{ind-index}},
   altplural={indexes},
   text={index\glsadd{ind-index}},
   description={an alphabetical list of names or subjects with
   references to their location in the document (pl.\ indices or
   indexes)}
 ]
 {index\glsadd{[glossaryloc]ind-index}}

\newterm
 [% options
   text={glossary\glsadd{ind-glossary}},
   plural={glossaries\glsadd{ind-glossary}},
   seealso={index}%
 ]%
 {glossary\glsadd{ind-glossary}}

\newterm
 [% options
   label=glossarycol,
   text={glossary\glsadd{ind-glossary}},
   plural={glossaries\glsadd{ind-glossary}},
   description={collection of glosses},
   parent=glossary
 ]
 {glossary\glsadd{[glossaryloc]ind-glossary}}

\newterm
 [% options
   label=glossarylist,
   text={glossary\glsadd{ind-glossary}},
   plural={glossaries\glsadd{ind-glossary}},
   description={list of technical words},
   parent=glossary
 ]
 {glossary\glsadd{[glossaryloc]ind-glossary}}

\newterm
  [% options
   label=pglist,
   % description contains commas so it must be grouped
   description={a list of individual pages or page ranges
    (e.g.\ 1,2,4,7--9)}
  ]
  {page list}

 % define some acronyms

\DTLgidxSetDefaultDB{acronyms}

 % Let's redefine \newacro so that the short form is also indexed:

\renewcommand{\newacro}[3][]{%
  \newterm[database=index,label={ind-#2-long},see={ind-#2}]{#3}%
  \newterm[database=index,label={ind-#2}]{\acronymfont{#2}}%
  \newterm
    [%
      description={\capitalisewords{#3}},% long form capitalised
      text={\DTLgidxAcrStyle{#3}{\acronymfont{#2}}},% produced via \gls
      plural={\DTLgidxAcrStyle{#3}{\acronymfont{#2}}},% produced via \glspl
      short={\acronymfont{#2}\glsadd{ind-#2}},% short form
      long={#3},% long form
      sort={#2},% sort on short form
      #1%
    ]%
    {#2\glsadd{[glossaryloc]ind-#2}}
}

\newacro{html}{hyper-text markup language}
\newacro{css}{cascading style sheet}
\newacro{xml}{extensible markup language}

% define some notation

\DTLgidxSetDefaultDB{notation}

\newterm
 [%
   symbol={\ensuremath{\mathcal{S}}},
   description={a collection of distinct objects},
 ]%
 {set}

\newterm
 [%
   parent=set,
   symbol={\ensuremath{\mathcal{U}}},
   text={universal set},
   description={the set containing everything}
 ]%
 {universal}

\newterm
 [%
   parent=set,
   symbol={\ensuremath{\emptyset}},
   text={empty set},
   description={the set with no elements}
 ]%
 {empty}

\newterm
 [%
   symbol={\ensuremath{|\mathcal{S}|}},
   description={the number of elements in the set \ensuremath{\mathcal{S}}}
 ]%
 {cardinality}

\title{Sample Document Using the datagidx Package}
\author{Nicola L. C. Talbot}

\begin{document}
\pagenumbering{alph}% Stop hyperref complaining about duplicate page identifiers
\maketitle
\thispagestyle{empty}%

\frontmatter

\tableofcontents

\chapter{Summary}

This is a sample document illustrating the use of the
\texttt{datagidx} package to create document \glspl{index},
\glspl{glossarylist} and \glspl{acronymlist} without the use of
external \glsing{index} % custom command defined earlier
applications, such as \gls{makeindex} or
\gls{xindy}. Please read the user guide (datatool-user.pdf) for
further guidance.

\mainmatter

\chapter{Introduction}
%\renewcommand{\thepage}{\numberstring{page}}

Words can be \glsed{index}. % custom command defined earlier.

A \gls{glossarylist}\glsadd{[principleloc]ind-glossary} is a useful addition to any technical document,
although a \gls{glossarycol} can also simply be a collection of
glosses, which is another thing entirely. Some documents have
multiple \glspl{glossarylist}.

A \gls{bravocry} is a cry of approval (plural \glspl{bravocry}) but a 
\gls{bravokiller} can also be a hired ruffian or killer (plural
\glspl{bravokiller}).

\section{Characters}

When defining entries be careful of \glspl{comma} and \glspl{equals}
so they don't interfere with the key=value mechanism. The sort and
label keys get expanded so be careful of special characters, such as
\gls{amp}, \gls{underscore}, \gls{circum}, \gls{dollar} and \gls{tilde}.

Since we're not using \gls{makeindex}, we don't need to worry about
\gls{makeindex}'s special characters, such as \gls{"}, \gls{!} and
\gls{|}. (Unless they've been made active by packages such as
\texttt{ngerman} or \texttt{babel}.)

Non-alphabetical characters are usually grouped at the start of an
index, and are usually followed by the number group. That is, the
group of entries that are numerical, such as \gls{0}, \gls{1},
\gls{2} and \gls{3}.

\section{Custom Fields}

You can add custom fields. For example, this document has added
three custom fields to the `index' database.

\section{Plurals}

The \gls{plural} of \gls{glossarylist} is
\glspl{glossarylist}. The \gls{plural} of \gls{index} is
\glspl{index}. Some words have an \gls{altplural}. For example,
an alternative to \glspl{index} is 
\glsaltpl{index}.% custom command defined earlier

\section{Sorting}

The only type of sorting available is letter sorting. If you want
word sorting you'll need to use \gls{makeindex} or \gls{xindy}.
So ``\gls{sea lion}'' comes after ``\gls{seal}''.

The default sort is case-insensitive so \gls{kite} before
\gls{knuth} and \gls{knuth} before \gls{koala}.

\section{Using without indexing}

Here's a defined entry that won't get into the glossary.
Name: \glsdispentry{pglist}{Name}.
Description: \glsdispentry{pglist}{Description}.
(Unless I~later reference it using a command like \verb|\gls|.)

\section{Links to Entries}

You can reference and index entries using \verb|\gls|, \verb|\Gls|,
\verb|\glspl|, \verb|\Glspl|, \verb|\glssym| and \verb|\Glssym|.
(Note, if you're used to using the \texttt{glossaries} package
the syntax is different.)

Or you can reference a particular field using \verb|\useentry| or
\verb|\Useentry|. So here's the description for \gls{seal}:
\useentry{seal}{Description}.

If the \texttt{hyperref} package has been loaded, commands like
\verb|\gls| will link to the relevant entry in the glossary or
index. Referencing using \verb|\glsdispentry| and
\verb|\Glsdispentry| won't have hyperlinks.

\subsection{Enabling and Disabling Hyperlinks}

If the \texttt{hyperref} package has been loaded, hyperlinks can be
enabled and disabled. Either globally
\DTLgidxDisableHyper
(here's a reference to \gls{seal} without a hyperlink 
\DTLgidxEnableHyper
 and here's a reference to \gls{seal} with a hyperlink)
or locally
({%
  \DTLgidxDisableHyper
  here's a reference to \gls{seal} without a hyperlink
}%
and here's a reference to \gls{seal} with a hyperlink).

\section{Acronyms}

\glsadd{firstuse}Here's an \gls{acronym} referenced using \verb|\acr|: \acr{html}. And here
it is again: \acr{html}. If you're used to the \texttt{glossaries}
package, note the difference in using \verb|\gls|: \gls{html}.
And again (no difference): \gls{html}.

Now let's switch to displaying acronyms with a footnote.
 % syntax: \DTLgidxAcrStyle{long}{short}
\renewcommand*{\DTLgidxAcrStyle}[2]{%
  #2\footnote{#1.}%
}%
First use: \acr{xml}. Next use: \acr{xml}.

However it would look better if the footnote text started with a
capital letter, so let's tweak things a bit.
 % syntax: \DTLgidxFormatAcr{label}{long field}{short field}
\renewcommand*{\DTLgidxFormatAcr}[3]{%
  \DTLgidxAcrStyle{\Glsdispentry{#1}{#2}}{\useentry{#1}{#3}}%
}%
 % syntax: \DTLgidxFormatAcr{label}{long field}{short field}
\renewcommand*{\DTLgidxFormatAcrUC}[3]{%
  \DTLgidxAcrStyle{\Glsdispentry{#1}{#2}}{\Useentry{#1}{#3}}%
}%
Try with another acronym: \acr{css}. Next use: \acr{css}.

\glsadd{reset}Reset: \glsresetall{acronyms}%
\renewcommand*{\DTLgidxAcrStyle}[2]{#1 (#2)}%
\renewcommand*{\DTLgidxFormatAcr}[3]{%
  \DTLgidxAcrStyle{\glsdispentry{#1}{#2}}{\useentry{#1}{#3}}%
}%
\renewcommand*{\DTLgidxFormatAcrUC}[3]{%
  \DTLgidxAcrStyle{\Glsdispentry{#1}{#2}}{\useentry{#1}{#3}}%
}%
Here are the acronyms again:
\acr{html}, \acr{xml} and \acr{css}.
Next use:
\acr{html}, \acr{xml} and \acr{css}.
Full form:
\gls{html}, \gls{xml} and \gls{css}.

Reset again. \glsresetall{acronyms}%
Start with a capital. \Acr{html}.
Next: \Acr{html}. Full: \Gls{html}.

Prefer small-caps?
\renewcommand{\acronymfont}[1]{\textsc{#1}}%
\Acr{css}. Next: \acr{css}. Full: \gls{css}.

Prefer capitals?
\renewcommand{\acronymfont}[1]{\MakeTextUppercase{#1}}%
\Acr{xml}. Next: \acr{xml}. Full: \gls{xml}.

\section{Conditionals}

You can test if a term has been defined using \verb|\iftermexists|.
For example: \iftermexists{seal}{seal exists}{seal doesn't exist}.
Another example: \iftermexists{jabberwocky}{jabberwocky
exists}{jabberwocky doesn't exist}.

You can test if a term has been used via \verb|ifentryused|.
For example: \ifentryused{seal}{seal has been used}{seal hasn't been
used.}
Another example: \ifentryused{pglist}{pglist has been used}{pglist
hasn't been used}.


\section{Symbols}

Terms may have an associated symbol. The symbol can be accessed
using \verb|\glssym| or if you don't want to add information to the
location list you can use \verb|\glsdispentry|. Here's the symbol
associated with the \gls{cardinality} entry:
\glsdispentry{cardinality}{Symbol}.

A \gls{set} (denoted \glssym{set}) is a collection of objects.
The \gls{universal} is the set of everything.
The \gls{empty} contains no elements.
The \gls{cardinality} of a set (denoted \glssym{cardinality}) is the
number of elements in the set.

\section{Location Ranges}

A range is formed if a location sequence contains more than 2
locations. Here's \gls{seal} again.

\backmatter
 % suppress section numbering
\setcounter{secnumdepth}{-1}

\chapter{Glossaries}

\printterms
 [
   database=glossary,% 'glossary' database
   columns=1,% one column page layout
   postdesc=dot,% put a full stop after the description
   style=gloss,% use 'gloss' style (sub-entries numbered)
   namefont={\textbf},% put the name in bold
   namecase=firstuc,% make the first letter of the name upper case
   child=noname,% don't print sub item names
   childsort=false,% don't sort the child lists
%   location=hide% hide the location list
 ]

\printterms
 [
   database=acronyms,% 'acronyms' database
   postdesc=dot,% put a full stop after the description
   columns=1,% one column page layout
   namefont={\textbf},% put the name (i.e. the abbreviation) in bold
   namecase=uc,% make the name upper case
   style=align,% use the 'align' style
%   location=hide% hide the location list
 ]

\printterms
 [
   database=notation,% 'notation' database
   postdesc=dot,% put a full stop after the description
   columns=1,% one page column layout
   namefont={\textbf},% make the name bold
   style=indexalign,
   namecase=firstuc,% make first letter of name a capital
   symboldesc={desc (symbol)},% put symbol to the right of the description
   sort={}% don't sort
 ]

\printterms
 [
   database=index,% 'index' database
   prelocation=dotfill,% put a dotted line before the location list
   columns=2,% page layout 
   style=index,% use 'index' style
   %balance=false,% don't balance columns
   postheading={Locations in bold indicate primary reference.
    Locations in italic indicate definitions in the
    glossaries.}
 ]


\end{document}