User Manual for glossaries.sty v4.46

Nicola L.C. Talbot
dickimaw-books.com/contact

2020-03-19

Abstract

The glossaries package provides a means to define terms or abbreviations or symbols that can be referenced within your document. Sorted lists with collated locations can be generated either using TeX or using a supplementary indexing application. Sample documents are provided with the glossaries package. These are listed in §19 Sample Documents.

Additional features not provided here may be available through the extension package glossaries-extra which, if required, needs to be installed separately. New features will be added to glossaries-extra. Versions of the glossaries package after v4.21 will mostly be just bug fixes or minor maintenance. Note that glossaries-extra provides an extra indexing option (bib2gls) which isn’t available with just the base glossaries package.

If you require multilingual support you must also separately install the relevant language module. Each language module is distributed under the name glossaries-language⟩, where ⟨language⟩ is the root language name. For example, glossaries-french or glossaries-german. If a language module is required, the glossaries package will automatically try to load it and will give a warning if the module isn’t found. See §1.3 Multi-Lingual Support for further details. If there isn’t any support available for your language, use the nolangwarn package option to suppress the warning and provide your own translations. (For example, use the title key in \printglossary.)

The glossaries package requires a number of other packages including, but not limited to, tracklang, mfirstuc, etoolbox, xkeyval (at least version dated 2006/11/18), textcase, xfor, datatool-base (part of the datatool bundle) and amsgen. These packages are all available in the latest TeX Live and MikTeX distributions. If any of them are missing, please update your TeX distribution using your update manager. For help on this see, for example, How do I update my TeX distribution? or (for Linux users) Updating TeX on Linux.

Note that occasionally you may find that certain packages need to be loaded after packages that are required by glossaries. For example, a package ⟨X⟩ might need to be loaded after amsgen. In which case, load the required package first (for example, amsgen), then ⟨X⟩, and finally load glossaries.

Documents have wide-ranging styles when it comes to presenting glossaries or lists of terms or notation. People have their own preferences and to a large extent this is determined by the kind of information that needs to go in the glossary. They may just have symbols with terse descriptions or they may have long technical words with complicated descriptions. The glossaries package is flexible enough to accommodate such varied requirements, but this flexibility comes at a price: a big manual.

😱 If you’re freaking out at the size of this manual, start with glossariesbegin.pdf (“The glossaries package: a guide for beginnners”). You should find it in the same directory as this document or try texdoc glossariesbegin.pdf. Once you’ve got to grips with the basics, then come back to this manual to find out how to adjust the settings.

The glossaries bundle comes with the following documentation:

glossariesbegin.pdf
If you are a complete beginner, start with “The glossaries package: a guide for beginners”.
glossary2glossaries.pdf
If you are moving over from the obsolete glossary package, read “Upgrading from the glossary package to the glossaries package”.
glossaries-user.pdf
This document is the main user guide for the glossaries package.
glossaries-code.pdf
Advanced users wishing to know more about the inner workings of all the packages provided in the glossaries bundle should read “Documented Code for glossaries v4.46”.
INSTALL
Installation instructions.
CHANGES
Change log.
README.md
Package summary.

Related resources:

If you use hyperref and glossaries, you must load hyperref first (although hyperref should be loaded after most other packages). Similarly the doc package must also be loaded before glossaries. (If doc is loaded, the file extensions for the default main glossary are changed to gls2, glo2 and glg2 to avoid conflict with doc’s changes glossary.)

If you are using hyperref, it’s best to use pdflatex rather than latex (DVI format) as pdflatex deals with hyperlinks much better. If you use the DVI format, you will encounter problems where you have long hyperlinks or hyperlinks in subscripts or superscripts. This is an issue with the DVI format not with glossaries. If you really need to use the DVI format and have a problem with hyperlinks in maths mode, I recommend you use glossaries-extra with the hyperoutside and textformat attributes set to appropriate values for problematic entries.

The glossaries package replaces the glossary package which is now obsolete. Please see the document “Upgrading from the glossary package to the glossaries package” for assistance in upgrading.

Contents

Glossary
1 Introduction
 1.1 Indexing Options
 1.2 Dummy Entries for Testing
 1.3 Multi-Lingual Support
  1.3.1 Changing the Fixed Names
 1.4 Generating the Associated Glossary Files
  1.4.1 Using the makeglossaries Perl Script
  1.4.2 Using the makeglossaries-lite Lua Script
  1.4.3 Using xindy explicitly (Option 3)
  1.4.4 Using makeindex explicitly (Option 2)
 1.5 Note to Front-End and Script Developers
  1.5.1 MakeIndex and Xindy
  1.5.2 Entry Labels
  1.5.3 Bib2Gls
2 Package Options
 2.1 General Options
 2.2 Sectioning, Headings and TOC Options
 2.3 Glossary Appearance Options
 2.4 Indexing Options
 2.5 Sorting Options
 2.6 Glossary Type Options
 2.7 Acronym and Abbreviation Options
  2.7.1 Deprecated Acronym Style Options
 2.8 Other Options
 2.9 Setting Options After the Package is Loaded
3 Setting Up
 3.1 Option 1
 3.2 Options 2 and 3
4 Defining Glossary Entries
 4.1 Plurals
 4.2 Other Grammatical Constructs
 4.3 Additional Keys
  4.3.1 Document Keys
  4.3.2 Storage Keys
 4.4 Expansion
 4.5 Sub-Entries
  4.5.1 Hierarchical Categories
  4.5.2 Homographs
 4.6 Loading Entries From a File
 4.7 Moving Entries to Another Glossary
 4.8 Drawbacks With Defining Entries in the Document Environment
  4.8.1 Technical Issues
  4.8.2 Good Practice Issues
5 Number lists
 5.1 Encap Values
 5.2 Locations
 5.3 Range Formations
 5.4 Style Hook
6 Links to Glossary Entries
 6.1 The \gls-Like Commands (First Use Flag Queried)
 6.2 The \glstext-Like Commands (First Use Flag Not Queried)
 6.3 Changing the format of the link text
 6.4 Enabling and disabling hyperlinks to glossary entries
7 Adding an Entry to the Glossary Without Generating Text
8 Cross-Referencing Entries
 8.1 Customising Cross-reference Text
9 Using Glossary Terms Without Links
10 Displaying a glossary
11 Xindy (Option 3)
 11.1 Language and Encodings
 11.2 Locations and Number lists
 11.3 Glossary Groups
12 Defining New Glossaries
13 Acronyms and Other Abbreviations
 13.1 Changing the Abbreviation Style
  13.1.1 Predefined Acronym Styles
  13.1.2 Defining A Custom Acronym Style
 13.2 Displaying the List of Acronyms
 13.3 Upgrading From the glossary Package
14 Unsetting and Resetting Entry Flags
 14.1 Counting the Number of Times an Entry has been Used (First Use Flag Unset)
15 Glossary Styles
 15.1 Predefined Styles
  15.1.1 List Styles
  15.1.2 Longtable Styles
  15.1.3 Longtable Styles (Ragged Right)
  15.1.4 Longtable Styles (booktabs)
  15.1.5 Supertabular Styles
  15.1.6 Supertabular Styles (Ragged Right)
  15.1.7 Tree-Like Styles
  15.1.8 Multicols Style
  15.1.9 In-Line Style
 15.2 Defining your own glossary style
16 Utilities
 16.1 Loops
 16.2 Conditionals
 16.3 Fetching and Updating the Value of a Field
17 Prefixes or Determiners
18 Accessibility Support
19 Sample Documents
 19.1 Basic
 19.2 Acronyms and First Use
 19.3 Non-Page Locations
 19.4 Multiple Glossaries
 19.5 Sorting
 19.6 Child Entries
 19.7 Cross-Referencing
 19.8 Custom Keys
 19.9 Xindy (Option 3)
 19.10 No Indexing Application (Option 1)
 19.11 Other
20 Troubleshooting

List of Tables

1.1 Glossary Options: Pros and Cons
1.2 Customised Text
1.3 Commands and package options that have no effect when using xindy or makeindex explicitly
4.1 Key to Field Mappings
6.1 Predefined Hyperlinked Location Formats
13.1 Synonyms provided by the package option shortcuts
13.2 The effect of using xspace
15.1 Glossary Styles
15.2 Multicolumn Styles

Glossary

This glossary style was setup using:

\usepackage[xindy,

nonumberlist,

toc,

nopostdot,

style=altlist,

nogroupskip]{glossaries}

bib2gls

An indexing application that combines two functions in one: (1) fetches entry definition from a bib file based on information provided in the aux file (similar to bibtex); (2) hierarchically sorts and collates location lists (similar to makeindex and xindy). This application is designed for use with glossaries-extra and can’t be used with just the base glossaries package. See Option 4.
Command Line Interface (CLI)

An application that doesn’t have a graphical user interface. That is, an application that doesn’t have any windows, buttons or menus and can be run in a command prompt or terminal.
convertgls2bib

An application provided with bib2gls that converts tex files containing entry definitions to bib files suitable for use with bib2gls. This application is designed for files that just contain entry definitions, but it can work on a complete document file. However, there will be a lot of “undefined command” warnings as convertgls2bib only has a limited set of known commands. You can limit it so that it only parses the preamble with the --preamble-only switch (requires at least bib2gls v2.0).
Entry location

The location of the entry in the document. This defaults to the page number on which the entry appears. An entry may have multiple locations.
Extended Latin Alphabet

An alphabet consisting of Latin characters and extended Latin characters.
Extended Latin Character

A character that’s created by combining Latin characters to form ligatures (e.g. æ) or by applying diacritical marks to a Latin character or characters (e.g. á). See also non-Latin character.
First use

The first time a glossary entry is used (from the start of the document or after a reset) with one of the following commands: \gls, \Gls, \GLS, \glspl, \Glspl, \GLSpl or \glsdisp. (See first use flag & first use text.)
First use flag

A conditional that determines whether or not the entry has been used according to the rules of first use. Commands to unset or reset this conditional are described in §14 Unsetting and Resetting Entry Flags.
First use text

The text that is displayed on first use, which is governed by the first and firstplural keys of \newglossaryentry. (May be overridden by \glsdisp or by \defglsentry.)
glossaries-extra

A separate package that extends the glossaries package, providing new features or improving existing features. If you want to use glossaries-extra, you must have both the glossaries package and the glossaries-extra package installed.
Indexing application

An application (piece of software) separate from TeX/LaTeX that collates and sorts information that has an associated page reference. Generally the information is an index entry but in this case the information is a glossary entry. There are two main indexing applications that are used with TeX: makeindex and xindy. These are both command line interface (CLI) applications.
Latin Alphabet

The alphabet consisting of Latin characters. See also extended Latin alphabet.
Latin Character

One of the letters a, …, z, A, …, Z. See also extended Latin character.
Link text

The text produced by commands such as \gls. It may or may not be a hyperlink to the glossary.
makeglossaries

A custom designed Perl script interface to xindy and makeindex provided with the glossaries package. TeX distributions on Windows convert the original makeglossaries script into an executable makeglossaries.exe for convenience (but Perl is still required).
makeglossariesgui

A Java GUI alternative to makeglossaries that also provides diagnostic tools. Available separately on CTAN.
makeglossaries-lite

A custom designed Lua script interface to xindy and makeindex provided with the glossaries package. This is a cut-down alternative to the Perl makeglossaries script. If you have Perl installed, use the Perl script instead. This script is actually distributed with the file name makeglossaries-lite.lua, but TeX Live (on Unix-like systems) creates a symbolic link called makeglossaries-lite (without the .lua extension) to the actual makeglossaries-lite.lua script.
makeindex

An indexing application. See Option 2.
Non-Latin Alphabet

An alphabet consisting of non-Latin characters.
Non-Latin Character

An extended Latin character or a character that isn’t a Latin character.
Number list

A list of entry locations (also called a location list). The number list can be suppressed using the nonumberlist package option.
Sanitize

Converts command names into character sequences. That is, a command called, say, \foo, is converted into the sequence of characters: \, f, o, o. Depending on the font, the backslash character may appear as a dash when used in the main document text, so \foo will appear as: —foo.

Earlier versions of glossaries used this technique to write information to the files used by the indexing applications to prevent problems caused by fragile commands. Now, this is only used for the sort key.

Small caps

Small capitals. The LaTeX kernel provides \textsc{text} to produce small capitals. This uses a font where lowercase letters have a small capital design. Uppercase letters have the standard height and there’s no noticeable difference with uppercase characters in corresponding non-small caps fonts. This means that for a small caps appearance, you need to use lowercase letters in the ⟨text⟩ argument. The package provides \textsmaller{text} which simulates small caps by reducing the size of the font, so in this case the contents of ⟨text⟩ should be uppercase (otherwise the effect is simply smaller lowercase letters). Some fonts don’t support small caps combined with bold or slanted properties. In this case, there will be a font substitution warning and one of the properties (such as small caps or slanted) will be dropped.
Standard LaTeX Extended Latin Character

An extended Latin character that can be created by a core LaTeX command, such as \o (ø) or \’e (é). That is, the character can be produced without the need to load a particular package.
xindy

A flexible indexing application with multilingual support written in Perl. See Option 3.

1. Introduction

The glossaries package is provided to assist generating lists of terms, symbols or abbreviations. (For convenience, these lists are all referred to as glossaries in this manual. The terms, symbols and abbreviations are collectively referred to as entries.) The package has a certain amount of flexibility, allowing the user to customize the format of the glossary and define multiple glossaries. It also supports glossary styles that include an associated symbol (in addition to a name and description) for each glossary entry.

There is provision for loading a database of glossary entries. Only those entries indexed1.1 in the document will be displayed in the glossary. (Unless you use Option 5, which doesn’t use any indexing but will instead list all defined entries in order of definition.)

It’s not necessary to actually have a glossary in the document. You may be interested in using this package just as means to consistently format certain types of terms, such as abbreviations, or you may prefer to have descriptions scattered about the document and be able to easily link to the relevant description (Option 6).

The simplest document is one without a glossary:

\documentclass{article}  
\usepackage[  
  sort=none % no sorting or indexing required  
]  
{glossaries}  
\newglossaryentry  
 {cafe}% label  
 {% definition:  
   name={caf\'e},  
   description={small restaurant selling refreshments}  
 }  
\setacronymstyle{long-short}  
\newacronym  
 {html}% label  
 {HTML}% short form  
 {hypertext markup language}% long form  
\newglossaryentry  
 {pi}% label  
 {% definition:  
   name={\ensuremath{\pi}},  
   description={Archimedes' Constant}  
 }  
\newglossaryentry  
 {distance}% label  
 {% definition:  
   name={distance},  
   description={the length between two points},  
   symbol={m}  
 }  
\begin{document}  
First use: \gls{cafe}, \gls{html}, \gls{pi}.  
Next use: \gls{cafe}, \gls{html}, \gls{pi}.  
\Gls{distance} (\glsentrydesc{distance}) is measured in  
\glssymbol{distance}.  
\end{document}

(This is a trivial example. For a real document I recommend you use siunitx for units.)


glossaries-extraThe glossaries-extra package, which is distributed as a separate bundle, extends the capabilities of the glossaries package. The simplest document with a glossary can be created with glossaries-extra (which internally loads the glossaries package):

\documentclass{article}

\usepackage[
 sort=none,% no sorting or indexing required
 abbreviations,% create list of abbreviations
 symbols,% create list of symbols
 postdot, % append a full stop after the descriptions
 stylemods,style=index % set the default glossary style
]{glossaries-extra}
\newglossaryentry % provided by glossaries.sty  
 {cafe}% label  
 {% definition:  
   name={caf\'e},  
   description={small restaurant selling refreshments}  
 }

% provided by glossaries-extra.sty:
\setabbreviationstyle{long-short}

\newabbreviation % provided by glossaries-extra.sty
 {html}% label
 {HTML}% short form
 {hypertext markup language}% long form

% provided by glossaries-extra.sty 'symbols' option:
\glsxtrnewsymbol
 [description={Archimedes' constant}]% options
 {pi}% label
 {\ensuremath{\pi}}% symbol
\newglossaryentry % provided by glossaries.sty  
 {distance}% label  
 {% definition:  
   name={distance},  
   description={the length between two points},  
   symbol={m}  
 }  
\begin{document}  
First use: \gls{cafe}, \gls{html}, \gls{pi}.  
Next use: \gls{cafe}, \gls{html}, \gls{pi}.  
\Gls{distance} is measured in \glssymbol{distance}.

\printunsrtglossaries % list all defined entries
\end{document}
Note the difference in the way the abbreviation (HTML) and symbol (π) are defined in the two above examples. The abbreviations, postdot and stylemods options are specific to glossaries-extra. Other options are passed to the base glossaries package.

In this document, commands and options displayed in teal, such as \newabbreviation and stylemods, are only available with the glossaries-extra package. There are also some commands and options (such as \makeglossaries and symbols) that are provided by the base glossaries package but are redefined by the glossaries-extra package. See the glossaries-extra user manual for further details of those commands.

One of the strengths of the glossaries package is its flexibility, however the drawback of this is the necessity of having a large manual that covers all the various settings. If you are daunted by the size of the manual, try starting off with the much shorter guide for beginners.

There’s a common misconception that you have to have Perl installed in order to use the glossaries package. Perl is not a requirement (as demonstrated by the above examples) but it does increase the available options, particularly if you use an extended Latin alphabet or a non-Latin alphabet.

This document uses the glossaries package. For example, when viewing the PDF version of this document in a hyperlinked-enabled PDF viewer (such as Adobe Reader or Okular) if you click on the word “xindy” you’ll be taken to the entry in the glossary where there’s a brief description of the term “xindy”. This is the way the glossaries mechanism works. An indexing application is used to generated the sorted list of terms. The indexing applications are command line interface (CLI) tools, which means they can be run directly from a command prompt or terminal, or can be integrated into some text editors, or you can use a build tool such as arara to run them.

Neither of the above two examples require an indexing application. The first is just using the glossaries package for consistent formatting, and there is no list. The second has lists but they are unsorted (see Option 5).

The remainder of this introductory section covers the following:

Top

1.1 Indexing Options

The basic idea behind the glossaries package is that you first define your entries (terms, symbols or abbreviations). Then you can reference these within your document (like \cite or \ref). You can also, optionally, display a list of the entries you have referenced in your document (the glossary). This last part, displaying the glossary, is the part that most new users find difficult. There are three options available with the base glossaries package (Options 13). The glossaries-extra extension package provides two extra options for lists (Options 4 and 5) as well as an option for standalone descriptions within the document body (Option 6).

An overview of Options 15 is given in table 1.1. Option 6 is omitted from the table as it doesn’t produce a list. For a more detailed comparison of the various methods, see the glossaries performance page.

If you are developing a class or package that loads glossaries, I recommend that you don’t force the user into a particular indexing method by adding an unconditional \makeglossaries into your class or package code. Aside from forcing the user into a particular indexing method, it means that they’re unable to use any commands that must come before \makeglossaries (such as \newglossary) and they can’t switch off the indexing whilst working on a draft document.

Strictly speaking, Options 5 and 6 aren’t actually indexing options as no indexing is performed. In the case of Option 5, all defined entries are listed in order of definition. In the case of Option 6, the entry hypertargets and descriptions are manually inserted at appropriate points in the document. These two options are included here for completeness and for comparison with the actual indexing options.


Table 1.1: Glossary Options: Pros and Cons

Option 1 Option 2 Option 3 Option 4 Option 5

Requires glossaries-extra?

Requires an external application?

Requires Perl?

Requires Java?

Can sort extended Latin alphabets or non-Latin alphabets?

* N/A

Efficient sort algorithm?

N/A

Can use a different sort method for each glossary?

N/A

Any problematic sort values?

Are entries with identical sort values treated as separate unique entries?

§

Can automatically form ranges in the location lists?

Can have non-standard locations in the location lists?

Maximum hierarchical depth (style-dependent)

# 3

\glsdisplaynumberlist reliable?

\newglossaryentry allowed in document environment? (Not recommended.)

Requires additional write registers?

Default value of sanitizesort package option

false true true true true

______________________________________________________________________________________

* Strips standard LaTeX accents (that is, accents generated by core LaTeX commands) so, for example, \AA is treated the same as A.

Only with the hybrid method provided with glossaries-extra.

Provided sort=none is used.

§ Entries with the same sort value are merged.

Requires some setting up.

The locations must be set explicitly through the custom location field provided by glossaries-extra.

# Unlimited but unreliable.

Entries are defined in bib format. \newglossaryentry should not be used explicitly.

Provided docdef=true or docdef=restricted but not recommended.

Provided docdef=false or docdef=restricted.

Irrelevant with sort=none. (The record=only option automatically switches this on.)


🔗Option 1 (TeX)

Example Document:

\documentclass{article}  
\usepackage{glossaries}

\makenoidxglossaries % use TeX to sort
\newglossaryentry{sample}{name={sample},  
  description={an example}}  
\begin{document}  
\gls{sample}.

\printnoidxglossary
\end{document}
You can place all your entry definitions in a separate file and load it in the preamble with \loadglsentries (after \makenoidxglossaries).

This option doesn’t require an external indexing application but, with the default alphabetic sorting, it’s very slow with severe limitations. If you want a sorted list, it doesn’t work well for extended Latin alphabets or non-Latin alphabets. However, if you use the sanitizesort=false package option (the default for Option 1) then the standard LaTeX accent commands will be ignored, so if an entry’s name is set to {\’e}lite then the sort value will default to elite if sanitizesort=false is used and will default to \’elite if sanitizesort=true is used. If you have any other kinds of commands that don’t expand to ASCII characters, such as \alpha or \si, then you must use sanitizesort=true or change the sort method (sort=use or sort=def) in the package options or explicitly set the sort key when you define the relevant entries. For example:

\newglossaryentry{alpha}{name={\ensuremath{\alpha}},
 sort={alpha},description={...}}

The glossaries-extra package has a modified symbols package option that provides \glsxtrnewsymbol, which automatically sets the sort key to the entry label (instead of the name).

This option works best with the sort=def or sort=use setting. For any other setting, be prepared for a long document build time, especially if you have a lot of entries defined. This option is intended as a last resort for alphabetical sorting. This option allows a mixture of sort methods. (For example, sorting by word order for one glossary and order of use for another.) This option is not suitable for hierarchical glossaries and does not form ranges in the number lists. If you really can’t use an indexing application consider using Option 5 instead.

  1. Add
    \makenoidxglossaries

    to your preamble (before you start defining your entries, as described in §4 Defining Glossary Entries).

  2. Put
    \printnoidxglossary

    where you want your list of entries to appear (described in §10 Displaying a glossary). Alternatively, to display all glossaries use the iterative command:

    \printnoidxglossaries

  3. Run LaTeX twice on your document. (As you would do to make a table of contents appear.) For example, click twice on the “typeset” or “build” or “PDFLaTeX” button in your editor.

🔗Option 2 (makeindex)

Example document:

\documentclass{article}  
\usepackage{glossaries}

\makeglossaries % open glossary files
\newglossaryentry{sample}{name={sample},  
 description={an example}}  
\begin{document}  
\gls{sample}.

\printglossary
\end{document}
You can place all your entry definitions in a separate file and load it in the preamble with \loadglsentries (after \makeglossaries).

This option uses a CLI application called makeindex to sort the entries. This application comes with all modern TeX distributions, but it’s hard-coded for the non-extended Latin alphabet. It can’t correctly sort accent commands (such as \’ or \c) and fails with UTF-8 characters, especially for any sort values that start with a UTF-8 character (as it separates the octets resulting in an invalid file encoding). This process involves making LaTeX write the glossary information to a temporary file which makeindex reads. Then makeindex writes a new file containing the code to typeset the glossary. Then \printglossary reads this file in on the next run.

This option works best if you want to sort entries according to the English alphabet and you don’t want to install Perl or Java. This method can also work with the restricted shell escape since makeindex is considered a trusted application. (So you should be able to use the automake package option provided the shell escape hasn’t been completely disabled.)

This method can form ranges in the number list but only accepts limited number formats: \arabic, \roman, \Roman, \alph and \Alph.

This option does not allow a mixture of sort methods. All glossaries must be sorted according to the same method: word/letter ordering or order of use or order of definition. If you need word ordering for one glossary and letter ordering for another you’ll have to explicitly call makeindex for each glossary type.

The glossaries-extra package allows a hybrid mix of Options 1 and 2 to provide word/letter ordering with Option 2 and order of use/definition with Option 1. See the glossaries-extra documentation for further details. See also the glossaries-extra alternative to sampleSort.tex in §19.5 Sorting.

  1. If you want to use makeindex’s -g option you must change the quote character using \GlsSetQuote. For example:
    \GlsSetQuote{+}

    This must be used before \makeglossaries. Note that if you are using babel, the shorthands aren’t enabled until the start of the document, so you won’t be able to use the shorthands in definitions made in the preamble.

  2. Add
    \makeglossaries

    to your preamble (before you start defining your entries, as described in §4 Defining Glossary Entries).

  3. Put
    \printglossary

    where you want your list of entries to appear (described in §10 Displaying a glossary). Alternatively, to display all glossaries use the iterative command:

    \printglossaries

  4. Run LaTeX on your document. This creates files with the extensions glo and ist (for example, if your LaTeX document is called myDoc.tex, then you’ll have two extra files called myDoc.glo and myDoc.ist). If you look at your document at this point, you won’t see the glossary as it hasn’t been created yet. (If you use glossaries-extra you’ll see the section heading and some boilerplate text.)

    If you have used package options such as symbols there will also be other sets of files corresponding to the extra glossaries that were created by those options.

  5. 🔗 Run makeindex with the .glo file as the input file and the .ist file as the style so that it creates an output file with the extension .gls. If you have access to a terminal or a command prompt (for example, the MSDOS command prompt for Windows users or the bash console for Unix-like users) then you need to run the command:

    
         
    makeindex -s myDoc.ist -o myDoc.gls myDoc.glo

    (Replace myDoc with the base name of your LaTeX document file. Avoid spaces in the file name if possible. The $ symbol indicates the command prompt and should be omitted.)

    The file extensions vary according to the glossary type. See §1.4.4 Using makeindex explicitly (Option 2) for further details. makeindex must be called for each set of files.

    If you don’t know how to use the command prompt, then you can probably access makeindex via your text editor, but each editor has a different method of doing this. See Incorporating makeglossaries or makeglossaries-lite or bib2gls into the document build for some examples.

    Alternatively, run makeindex indirectly via the makeglossaries script:

    
         
    makeglossaries myDoc

    Note that the file extension isn’t supplied in this case. (Replace makeglossaries with makeglossaries-lite if you don’t have Perl installed.) This will pick up all the file extensions from the aux file and run makeindex the appropriate number of times with all the necessary switches.

    The simplest approach is to use arara and add the following comment lines to the start of your document:

    % arara: pdflatex
    % arara: makeglossaries
    % arara: pdflatex
    (Replace makeglossaries with makeglossarieslite in the second line above if you don’t have Perl installed. Note that there’s no hyphen in this case.)

    The default sort is word order (“sea lion” comes before “seal”). If you want letter ordering you need to add the -l switch:

    
         
    makeindex -l -s myDoc.ist -o myDoc.gls myDoc.glo

    (See §1.4.4 Using makeindex explicitly (Option 2) for further details on using makeindex explicitly.) If you use makeglossaries or makeglossaries-lite then use the order=letter package option and the -l option will be added automatically.

  6. 🔗 Once you have successfully completed the previous step, you can now run LaTeX on your document again.

You’ll need to repeat the last step if you have used the toc option (unless you’re using glossaries-extra) to ensure the section heading is added to the table of contents. You’ll also need to repeat steps 5 and 6 if you have any cross-references which can’t be indexed until the glossary file has been created.

🔗Option 3 (xindy)

Example document:

\documentclass{article}
\usepackage[xindy]{glossaries}
\makeglossaries % open glossary files
\newglossaryentry{sample}{name={sample},  
 description={an example}}  
\begin{document}  
\gls{sample}.

\printglossary
\end{document}
You can place all your entry definitions in a separate file and load it in the preamble with \loadglsentries (after \makeglossaries).

This option uses a CLI application called xindy to sort the entries. This application is more flexible than makeindex and is able to sort extended Latin alphabets or non-Latin alphabets, however it does still have some limitations.

The xindy application comes with both TeX Live and MiKTeX, but since xindy is a Perl script, you will also need to install Perl, if you don’t already have it. In a similar way to Option 2, this option involves making LaTeX write the glossary information to a temporary file which xindy reads. Then xindy writes a new file containing the code to typeset the glossary. Then \printglossary reads this file in on the next run.

This is the best option with just the base glossaries package if you want to sort according to a language other than English or if you want non-standard location lists, but it can require some setting up (see §11 Xindy (Option 3)). There are some problems with certain sort values:

In these problematic cases, you must set the sort field explicitly. For example:

\newglossaryentry{alpha}{name={\ensuremath{\alpha}},
 sort={alpha},description={...}}

The glossaries-extra package has a modified symbols package option that provides \glsxtrnewsymbol, which automatically sets the sort key to the entry label (instead of the name).

All glossaries must be sorted according to the same method (word/letter ordering, order of use, or order of definition).

The glossaries-extra package allows a hybrid mix of Options 1 and 3 to provide word/letter ordering with Option 3 and order of use/definition with Option 1. See the glossaries-extra documentation for further details.

  1. Add the xindy option to the glossaries package option list:
    \usepackage[xindy]{glossaries}

    If you are using a non-Latin script you’ll also need to either switch off the creation of the number group:

    \usepackage[xindy={glsnumbers=false}]{glossaries}

    or use either \GlsSetXdyFirstLetterAfterDigits{letter} (to indicate the first letter group to follow the digits) or \GlsSetXdyNumberGroupOrder{spec} to indicate where the number group should be placed (see §11 Xindy (Option 3)).

  2. Add \makeglossaries to your preamble (before you start defining your entries, as described in §4 Defining Glossary Entries).
  3. Run LaTeX on your document. This creates files with the extensions glo and xdy (for example, if your LaTeX document is called myDoc.tex, then you’ll have two extra files called myDoc.glo and myDoc.xdy). If you look at your document at this point, you won’t see the glossary as it hasn’t been created yet. (If you’re using the glossaries-extra extension package, you’ll see the section header and some boilerplate text.)

    If you have used package options such as symbols there will also be other sets of files corresponding to the extra glossaries that were created by those options.

  4. Run xindy with the .glo file as the input file and the .xdy file as a module so that it creates an output file with the extension .gls. You also need to set the language name and input encoding. If you have access to a terminal or a command prompt (for example, the MSDOS command prompt for Windows users or the bash console for Unix-like users) then you need to run the command (all on one line):

    
         
    xindy -L english -C utf8 -I xindy -M myDoc -t myDoc.glg -o myDoc.gls myDoc.glo

    (Replace myDoc with the base name of your LaTeX document file. Avoid spaces in the file name. If necessary, also replace english with the name of your language and utf8 with your input encoding, for example, -L german -C din5007-utf8.)

    The file extensions vary according to the glossary type. See §1.4.3 Using xindy explicitly (Option 3) for further details. xindy must be called for each set of files.

    It’s much simpler to use makeglossaries instead:

    
         
    makeglossaries myDoc

    Note that the file extension isn’t supplied in this case. This will pick up all the file extensions from the aux file and run xindy the appropriate number of times with all the necessary switches.

    There’s no benefit in using makeglossaries-lite with xindy. (Remember that xindy is a Perl script so if you can use xindy then you can also use makeglossaries, and if you don’t want to use makeglossaries because you don’t want to install Perl, then you can’t use xindy either.)

    If you don’t know how to use the command prompt, then you can probably access xindy or makeglossaries via your text editor, but each editor has a different method of doing this. See Incorporating makeglossaries or makeglossaries-lite or bib2gls into the document build for some examples.

    Again, a convenient method is to use arara and add the follow comment lines to the start of your document:

    % arara: pdflatex
    % arara: makeglossaries
    % arara: pdflatex

    The default sort is word order (“sea lion” comes before “seal”). If you want letter ordering you need to add the order=letter package option:

    \usepackage[xindy,order=letter]{glossaries}

    (and return to the previous step). This option is picked up by makeglossaries. If you are explicitly using xindy then you’ll need to add -M ord/letorder to the options list. See §1.4.3 Using xindy explicitly (Option 3) for further details on using xindy explicitly.

  5. Once you have successfully completed the previous step, you can now run LaTeX on your document again. As with makeindex (Option 2), you may need to repeat the previous step and this step to ensure the table of contents and cross-references are resolved.

🔗Option 4 (bib2gls)


glossaries-extraThis option is only available with the glossaries-extra extension package. This method uses bib2gls to both fetch entry definitions from bib files and to hierarchically sort and collate.

Example document:

\documentclass{article}
\usepackage[record=nameref]{glossaries-extra}
\GlsXtrLoadResources[src={entries}]
\begin{document}
\gls{sample}, \gls{alpha}, \gls{html}.
\printunsrtglossary
\end{document}
where the file entries.bib contains:
@entry{sample,
  name={sample},
  description={an example}
}
@symbol{alpha,
  name={\ensuremath{\alpha}},
  description={...}
}
@abbreviation{html,
  short={HTML},
  long={hypertext markup language}
}

All entries must be provided in one or more bib files. See the bib2gls user manual for the required format.

Note that the sort key should not be used. Each entry type (@entry, @symbol, @abbreviation) has a particular field that’s used for the sort value by default (name, the label, short). You will break this mechanism if you explicitly use the sort key. See bib2gls gallery: sorting for examples.

The glossaries-extra package needs to be loaded with the record package option:

\usepackage[record]{glossaries-extra}
or (equivalently)
\usepackage[record=only]{glossaries-extra}
or (with at least glossaries-extra v1.37 and bib2gls v1.8):
\usepackage[record=nameref]{glossaries-extra}
The record=nameref option is the best method.

(It’s possible to use a hybrid of this method and Options 2 or 3 with record=alsoindex but in general there is little need for this and it complicates the build process.)

Each resource set is loaded with \GlsXtrLoadResources[options]. For example:

\GlsXtrLoadResources
[% definitions in entries1.bib and entries2.bib:
 src={entries1,entries2},
 sort={de-CH-1996}% sort according to this locale
]
The bib files are identified as a comma-separated list in the value of the src key. The sort option identifies the sorting method. This may be a locale identifier for alphabetic sorting, but there are other sort methods available, such as character code or numeric. One resource set may cover multiple glossaries or one glossary may be split across multiple resource sets, forming logical sub-blocks.

If you want to ensure that all entries are selected, even if they haven’t been referenced in the document, then add the option selection=all. (There are also ways of filtering the selection or you can even have a random selection by shuffling and truncating the list. See the bib2gls user manual for further details.)

The glossary is displayed using:

\printunsrtglossary
Alternatively all glossaries can be displayed using the iterative command:
\printunsrtglossaries
The document is built using:


   
pdflatex myDoc
bib2gls myDoc
pdflatex myDoc

If letter groups are required, you need the --group switch:


   
bib2gls --group myDoc

or with arara:

% arara: bib2gls: { group: on }
(You will also need an appropriate glossary style.)

Unlike Options 2 and 3, this method doesn’t create a file containing the typeset glossary but simply determines which entries are needed for the document, their associated locations and (if required) their associated letter group. This option allows a mixture of sort methods. For example, sorting by word order for one glossary and order of use for another or even sorting one block of the glossary differently to another block in the same glossary. See bib2gls gallery: sorting.

This method supports Unicode and uses the Common Locale Data Repository, which provides more extensive language support than xindy.1.2 The locations in the number list may be in any format. If bib2gls can deduce a numerical value it will attempt to form ranges otherwise it will simply list the locations.

See glossaries-extra and bib2gls: An Introductory Guide or the bib2gls user manual for further details of this method, and also Incorporating makeglossaries or makeglossaries-lite or bib2gls into the document build.

🔗Option 5 (no sorting)


glossaries-extraThis option is only available with the extension package glossaries-extra. No indexing application is required.

Example document:

\documentclass{article}  
\usepackage[sort=none]{glossaries-extra}  
\newglossaryentry{sample}{name={sample},  
 description={an example}}  
\newglossaryentry{alpha}{name={\ensuremath{\alpha}},  
 description={...}}  
\begin{document}  
\gls{sample}.

\printunsrtglossary
\end{document}

This method is best used with the package option sort=none (as shown above). There’s no “activation” command (such as \makeglossaries for Options 2 and 3).

All entries must be defined before the glossary is displayed (preferably in the preamble) in the required order, and child entries must be defined immediately after their parent entry if they must be kept together in the glossary. (Some glossary styles indent entries that have a parent but it’s the indexing application that ensures the child entries are listed immediately after the parent. If you’re opting to use this manual approach then it’s your responsibility to define the entries in the correct order.) You can place all your entry definitions in a separate file and load it in the preamble with \loadglsentries.

The glossary is displayed using:

\printunsrtglossary
Alternatively all glossaries can be displayed using the iterative command:
\printunsrtglossaries
This will display all defined entries, regardless of whether or not they have been used in the document. The number lists have to be set explicitly otherwise they won’t appear. Note that this uses the same command for displaying the glossary as Option 4. This is because bib2gls takes advantage of this method by defining the wanted entries in the required order and setting the locations (and letter group information, if required).

Therefore, the above example document has a glossary containing the entries: sample and α (in that order). Note that the alpha entry has been included even though it wasn’t referenced in the document.

This just requires a single LaTeX call:


   
pdflatex myDoc

unless the glossary needs to appear in the table of contents, in which case a second run is required:


   
pdflatex myDoc
pdflatex myDoc

(Naturally if the document also contains citations, and so on, then additional steps are required. Similarly for all the other options above.)

See the glossaries-extra documentation for further details of this method.

🔗Option 6 (standalone)


glossaries-extraThis option is only available with the glossaries-extra extension package.1.3 Instead of creating a list, this has standalone definitions throughout the document. The entry name may or may not be in a section heading.

You can either define entries in the document preamble (or in an external file loaded with \loadglsentries), as with Option 5, for example:

\documentclass{article}  
\usepackage[colorlinks]{hyperref}  
\usepackage[sort=none,  
   nostyles% <- no glossary styles are required  
 ]{glossaries-extra}  
\newglossaryentry{set}{name={set},  
  description={a collection of any kind of objects},  
  symbol={\ensuremath{\mathcal{S}}}  
}  
\newglossaryentry{function}{name={function},  
  description={a rule that assigns every element in the  
  domain \gls{set} to an element in the range \gls{set}},  
  symbol={\ensuremath{f(x)}}  
}

\newcommand*{\termdef}[1]{%
  \section{\glsxtrglossentry{#1} \glsentrysymbol{#1}}%
  \begin{quote}\em\Glsentrydesc{#1}.\end{quote}%
}
\begin{document}  
\tableofcontents  
\section{Introduction}  
Sample document about \glspl{function} and \glspl{set}.  
\termdef{set}  
More detailed information about \glspl{set} with examples.  
\termdef{function}  
More detailed information about \glspl{function} with examples.  
\end{document}

Or you can use bib2gls if you want to manage a large database of terms. For example (requires glossaries-extra v1.42, see below):

\documentclass{article}

\usepackage[colorlinks]{hyperref}
\usepackage[record,
   nostyles% <- no glossary styles are required
  ]{glossaries-extra}

\GlsXtrLoadResources[src=terms,sort=none,save-locations=false]

\newcommand*{\termdef}[1]{%
  \section{\glsxtrglossentry{#1} \glossentrysymbol{#1}}%
  \glsadd{#1}% <- index this entry
  \begin{quote}\em\Glsentrydesc{#1}.\end{quote}%
}
\begin{document}  
\tableofcontents  
\section{Introduction}  
Sample document about \glspl{function} and \glspl{set}.  
\termdef{set}  
More detailed information about \glspl{set} with examples.  
\termdef{function}  
More detailed information about \glspl{function} with examples.  
\end{document}

Where the file terms.bib contains:

@entry{set,
  name={set},
  description={a collection of any kind of objects},
  symbol={\ensuremath{\mathcal{S}}}
}
@entry{function,
  name={function},
  description={a rule that assigns every element in the domain
  \gls{set} to an element in the range \gls{set}},
  symbol={\ensuremath{f(x)}}
}
The advantage in this approach (with \loadglsentries or bib2gls) is that you can use an existing database of entries shared across multiple documents, ensuring consistent notation for all of them.

In both cases, there’s no need to load all the glossary styles packages, as they’re not required, so I’ve used the nostyles package option to prevent them from being loaded.

In the first case, you need the sort=none package option (as in Option 5) and then define the terms in the preamble. No external tool is required. Just run LaTeX as normal. (Twice to ensure that the table of contents is up to date.)


   
pdflatex myDoc
pdflatex myDoc

In the second case, you need the record package option (as in Option 4) since bib2gls is needed to select the required entries, but you don’t need a sorted list:

\GlsXtrLoadResources[src={terms},sort=none]
This will ensure that any entries indexed in the document (through commands like \gls or \glsadd) will be selected by bib2gls, but it will skip the sorting step. (The chances are you probably also won’t need location lists either. If so, you can add the option save-locations=false.)

Remember that for this second case you need to run bib2gls as per Option 4:


   
pdflatex myDoc
bib2gls myDoc
pdflatex myDoc
pdflatex myDoc

For both cases (with or without bib2gls), instead of listing all the entries using \printunsrtglossary, you use \glsxtrglossentry{label} where you want the name (and anchor with hyperref) to appear in the document. This will allow the link text created by commands like \gls to link to that point in the document. The description can simply be displayed with \glsentrydesc{label} or \Glsentrydesc{label}, as in the above examples. In both examples, I’ve defined a custom command \termdef to simplify the code and ensure consistency. Extra styling, such as placing the description in a coloured frame, can be added to this custom definition as required.

(Instead of using \glsentrydesc or \Glsentrydesc, you can use \glossentrydesc{label}, which will obey attributes such as glossdesc and glossdescfont. See the glossaries-extra manual for further details.)

The symbol (if required) can be displayed with either \glsentrysymbol{label} or \glossentrysymbol {label}. In the first example, I’ve used \glsentrysymbol. In the second I’ve used \glossentrysymbol. The latter is necessary with bib2gls if the symbol needs to go in a section title as the entries aren’t defined on the first LaTeX run.

In normal document text, \glsentrysymbol will silently do nothing if the entry hasn’t been defined, but when used in a section heading it will expand to an undefined internal command when written to the aux file, which triggers an error.

The \glossentrysymbol command performs an existence check, which triggers a warning if the entry is undefined. (All entries will be undefined before the first bib2gls call.) You need at least glossaries-extra v1.42 to use this command in a section title.1.4 If hyperref has been loaded, this will use \texorpdfstring to allow a simple expansion for the PDF bookmarks (see the glossaries-extra user manual for further details).

If you want to test if the symbol field has been set, you need to use \ifglshassymbol outside of the section title. For example:

\ifglshassymbol{#1}%
{\section{\glsxtrglossentry{#1} \glossentrysymbol{#1}}}
{\section{\glsxtrglossentry{#1}}}

In both of the above examples, the section titles start with a lowercase character (because the name value is all lowercase in entry definitions). You can apply automatic case-change with the glossname attribute. For example:

\glssetcategoryattribute{general}{glossname}{firstuc}
or (for title-case)
\glssetcategoryattribute{general}{glossname}{title}
However, this won’t apply the case-change in the table of contents or bookmarks.

In the second example, you can instead use bib2gls to apply a case-change:

\GlsXtrLoadResources[src=terms,
 sort=none,save-locations=false,
 replicate-fields={name=text},
 name-case-change=firstuc

]
(Or name-case-change=title for title-case.) This copies the name value to the text field and then applies a case-change to the name field (leaving the text field unchanged). The name in the section titles now starts with a capital but the link text produced by commands like \gls is still lowercase.

In the first example (without bib2gls) you need to do this manually. For example:

\newglossaryentry{set}{name={Set},text={set},
  description={a collection of any kind of objects},
  symbol={\ensuremath{\mathcal{S}}}
}

Note that if you use the default save-locations=true with bib2gls, it’s possible to combine Options 4 and 6 to have both standalone definitions and an index. Now I do need a glossary style. In this case I’m going to use bookindex, which is provided in the glossary-bookindex package (bundled with glossaries-extra). I don’t need any of the other style packages, so I can still keep the nostyles option and just load glossary-bookindex:

\usepackage[record=nameref,% <- using bib2gls
 nostyles,% <- don't load default style packages
 stylemods=bookindex,% <- load glossary-bookindex.sty
 style=bookindex% <- set the default style to 'bookindex'
]{glossaries-extra}
I also need to sort the entries, so the resource command is now:
\GlsXtrLoadResources[src=terms,% definitions in terms.bib
 sort=en-GB,% sort by this locale
 replicate-fields={name=text},
 name-case-change={firstuc}
]
At the end of the document, I can add the glossary:
\printunsrtglossary[title=Index,target=false]
Note that I’ve had to switch off the hypertargets with target=false (otherwise there would be duplicate targets). If you want letter group headings you need to use the --group switch:


   
bib2gls --group myDoc

or if you are using arara:

% arara: bib2gls: { group: on }

The bookindex style doesn’t show the description, so only the name and location is displayed. Remember that the name has had a case-conversion so it now starts with an initial capital. If you feel this is inappropriate for the index, you can adjust the bookindex style so that it uses the text field instead. For example:

\renewcommand*{\glsxtrbookindexname}[1]{%
  \glossentrynameother{#1}{text}}
See the glossaries-extra user manual for further details about this style.

Note that on the first LaTeX run none of the entries will be defined. Once they are defined, the page numbers may shift due to the increased amount of document text. You may therefore need to repeat the document build to ensure the page numbers are correct.

If there are extra terms that need to be included in the index that don’t have a description, you can define them with @index in the bib file. For example:

@index{element}
@index{member,alias={element}}
They can be used in the document as usual:
The objects that make up a set are the \glspl{element}  
or \glspl{member}.

See glossaries-extra and bib2gls: An Introductory Guide or the bib2gls user manual for further details.

The glossaries package comes with a number of sample documents that illustrate the various functions. These are listed in §19 Sample Documents.

Top

1.2 Dummy Entries for Testing

In addition to the sample files described above, glossaries also provides some files containing lorum ipsum dummy entries. These are provided for testing purposes and are on TeX’s path (in tex/latex/glossaries/test-entries) so they can be included via \input or \loadglsentries. The glossaries-extra package provides bib versions of all these files for use with bib2gls. The files are as follows:

example-glossaries-brief.tex
These entries all have brief descriptions. For example:
\newglossaryentry{lorem}{name={lorem},description={ipsum}}

example-glossaries-long.tex
These entries all have long descriptions. For example:
\newglossaryentry{loremipsum}{name={lorem ipsum},  
description={dolor sit amet, consectetuer adipiscing  
elit. Ut purus elit, vestibulum ut, placerat ac,  
adipiscing vitae, felis. Curabitur dictum gravida  
mauris.}}

example-glossaries-multipar.tex
These entries all have multi-paragraph descriptions. For example:
\longnewglossaryentry{loremi-ii}{name={lorem 1--2}}%  
{%  
Lorem ipsum ...  
Nam dui ligula...  
}

example-glossaries-symbols.tex
These entries all use the symbol key. For example:
\newglossaryentry{alpha}{name={alpha},  
symbol={\ensuremath{\alpha}},  
description={Quisque ullamcorper placerat ipsum.}}

example-glossaries-symbolnames.tex
Similar to the previous file but the symbol key isn’t used. Instead the symbol is stored in the name key. For example:
\newglossaryentry{sym.alpha}{sort={alpha},  
name={\ensuremath{\alpha}},  
description={Quisque ullamcorper placerat ipsum.}}

example-glossaries-images.tex
These entries use the user1 key to store the name of an image file. (The images are provided by the mwe package and should be on TeX’s path.) One entry doesn’t have an associated image to help test for a missing key. The descriptions are long to allow for tests with the text wrapping around the image. For example:
\longnewglossaryentry{sedfeugiat}{name={sed feugiat},  
user1={example-image}}%  
{%  
Cum sociis natoque...  
Etiam...  
}

example-glossaries-acronym.tex
These entries are all abbreviations. For example:
\newacronym[type=\glsdefaulttype]{lid}{LID}{lorem ipsum  
dolor}

If you use the glossaries-extra extension package, then \newacronym is redefined to use \newabbreviation with the category key set to acronym (rather than the default abbreviation). This means that you need to set the abbreviation style for the acronym category. For example:

\setabbreviationstyle[acronym]{long-short}

example-glossaries-acronym-desc.tex
This file contains entries that are all abbreviations that use the description key. For example:
\newacronym[type=\glsdefaulttype,  
  description={fringilla a, euismod sodales,  
  sollicitudin vel, wisi}]{ndl}{NDL}{nam dui ligula}

If you use the glossaries-extra extension package, then \newacronym is redefined to use \newabbreviation with the category key set to acronym (rather than the default abbreviation). This means that you need to set the abbreviation style for the acronym category. For example:

\setabbreviationstyle[acronym]{long-short-desc}

example-glossaries-acronyms-lang.tex
These entries are all abbreviations, where some of them have a translation supplied in the user1 key. For example:
\newacronym[type=\glsdefaulttype,user1={love itself}]  
 {li}{LI}{lorem ipsum}

If you use the glossaries-extra extension package, then \newacronym is redefined to use \newabbreviation with the category key set to acronym (rather than the default abbreviation). This means that you need to set the abbreviation style for the acronym category. For example:

\setabbreviationstyle[acronym]{long-short-user}

example-glossaries-parent.tex
These are hierarchical entries where the child entries use the name key. For example:
\newglossaryentry{sedmattis}{name={sed mattis},  
description={erat sit amet}  
\newglossaryentry{gravida}{parent={sedmattis},  
  name={gravida},description={malesuada}}

example-glossaries-childnoname.tex
These are hierarchical entries where the child entries don’t use the name key. For example:
\newglossaryentry{scelerisque}{name={scelerisque},  
  description={at}}  
\newglossaryentry{vestibulum}{parent={scelerisque},  
  description={eu, nulla}}

example-glossaries-cite.tex
These entries use the user1 key to store a citation key (or comma-separated list of citation keys). The citations are defined in xampl.bib, which should be available on all modern TeX distributions. One entry doesn’t have an associated citation to help test for a missing key. For example:
\newglossaryentry{fusce}{name={fusce},  
description={suscipit cursus sem},user1={article-minimal}}

example-glossaries-url.tex
These entries use the user1 key to store an URL associated with the entry. For example:
\newglossaryentry{aenean-url}{name={aenean},  
 description={adipiscing auctor est},  
 user1={http://uk.tug.org/}}

The sample file glossary-lipsum-examples.tex in the doc/latex/glossaries/samples directory uses all these files. See also https://www.dickimaw-books.com/gallery/#glossaries.


glossaries-extraThe glossaries-extra package provides the additional test file:

example-glossaries-xr.tex
These entries use the see key provided by the base glossaries package and also the alias and seealso keys that require glossaries-extra. For example:
\newglossaryentry{alias-lorem}{name={alias-lorem},  
 description={ipsum},alias={lorem}}  
\newglossaryentry{amet}{name={amet},description={consectetuer},  
 see={dolor}}  
\newglossaryentry{arcu}{name={arcu},description={libero},  
 seealso={placerat,vitae,curabitur}}

Top

1.3 Multi-Lingual Support

The glossaries package uses the tracklang package to determine the document languages. Unfortunately, because there isn’t a standard language identification framework provided with LaTeX, tracklang isn’t always able to detect the selected languages either as a result of using an unknown interface or where the interface doesn’t provide a way of detecting the language. See Localisation with tracklang.tex for further details.

As from version 1.17, the glossaries package can be used with xindy as well as makeindex. If you are writing in a language that uses an extended Latin alphabet or non-Latin alphabet it’s best to use Option 3 (xindy) or Option 4 (bib2gls) as makeindex (Option 2) is hard-coded for the non-extended Latin alphabet and Option 1 can only perform limited ASCII comparisons.

This means that with Options 3 or 4 you are not restricted to the A, …, Z letter groups. If you want to use xindy, remember to use the xindy package option. For example:

\documentclass[frenchb]{article}  
\usepackage[utf8]{inputenc}  
\usepackage[T1]{fontenc}  
\usepackage{babel}  
\usepackage[xindy]{glossaries}

If you want to use bib2gls, you need to use the record option with glossaries-extra and supply the definitions in bib files. (See the bib2gls user manual for further details.)

Note that although a non-Latin character, such as é, looks like a plain character in your tex file, with standard LaTeX it’s actually a macro and can therefore cause expansion problems. You may need to switch off the field expansions with \glsnoexpandfields. This issue doesn’t occur with XeLaTeX or LuaLaTeX.

With inputenc, if you use a non-Latin character (or other expandable) character at the start of an entry name, you must place it in a group, or it will cause a problem for commands that convert the first letter to upper case (e.g. \Gls). For example:

\newglossaryentry{elite}{name={{é}lite},
description={select group or class}}
For further details, see the “UTF-8” section in the mfirstuc user manual.

If you are using xindy or bib2gls, the application needs to know the encoding of the tex file. This information is added to the aux file and can be picked up by makeglossaries and bib2gls. If you use xindy explicitly instead of via makeglossaries, you may need to specify the encoding using the -C option. Read the xindy manual for further details of this option.

As from v4.24, if you are writing in German (for example, using the ngerman package1.5 or babel with the ngerman package option), and you want to use makeindex’s -g option, you’ll need to change makeindex’s quote character using:


\GlsSetQuote{character}

Note that ⟨character⟩ may not be one of ? (question mark), | (pipe) or ! (exclamation mark). For example:

\GlsSetQuote{+}

This must be done before \makeglossaries and any entry definitions. It’s only applicable for makeindex. This option in conjunction with ngerman will also cause makeglossaries to use the -g switch when invoking makeindex.

Be careful of babel’s shorthands. These aren’t switched on until the start of the document, so any entries defined in the preamble won’t be able to use those shorthands. However, if you define the entries in the document and any of those shorthands happen to be special characters for makeindex or xindy (such as the double-quote) then this will interfere with code that tries to escape any of those characters that occur in the sort key.

In general, it’s best not to use babel’s shorthands in entry definitions. For example:

\documentclass{article}  
\usepackage[ngerman]{babel}  
\usepackage{glossaries}  
\GlsSetQuote{+}  
\makeglossaries  
\newglossaryentry{rna}{name={ribonukleins\"aure},  
  sort={ribonukleins"aure},  
  description={eine Nukleins\"aure}}  
\begin{document}  
\gls{rna}  
\printglossaries  
\end{document}

The ngerman package has the shorthands on in the preamble, so they can be used in definitions if \GlsSetQuote has been used to change the makeindex quote character. Example:

\documentclass{article}  
\usepackage[ngerman]{babel}  
\usepackage{glossaries}  
\GlsSetQuote{+}  
\makeglossaries  
\newglossaryentry{rna}{name={ribonukleins"aure},  
  description={eine Nukleins"aure}}  
\begin{document}  
\gls{rna}  
\printglossaries  
\end{document}

Top

1.3.1 Changing the Fixed Names

The fixed names are produced using the commands listed in table 1.2. If you aren’t using a language package such as babel or polyglossia that uses caption hooks, you can just redefine these commands as appropriate. If you are using babel or polyglossia, you need to use their caption hooks to change the defaults. See changing the words babel uses or read the babel or polyglossia documentation. If you have loaded babel, then glossaries will attempt to load translator, unless you have used the notranslate, translate=false or translate=babel package options. If the translator package is loaded, the translations are provided by dictionary files (for example, glossaries-dictionary-English.dict). See the translator package for advice on changing translations provided by translator dictionaries. If you can’t work out how to modify these dictionary definitions, try switching to babel’s interface using translate=babel:

\documentclass[english,french]{article}  
\usepackage{babel}  
\usepackage[translate=babel]{glossaries}

and then use babel’s caption hook mechanism. Note that if you pass the language options directly to babel rather that using the document class options or otherwise passing the same options to translator, then translator won’t pick up the language and no dictionaries will be loaded and babel’s caption hooks will be used instead.


Table 1.2: Customised Text
Command Name

Translator Key Word

Purpose

\glossaryname

Glossary

Title of the main glossary.

\acronymname

Acronyms

Title of the list of acronyms (when used with package option acronym).

\entryname

Notation (glossaries)

Header for first column in the glossary (for 2, 3 or 4 column glossaries that support headers).

\descriptionname

Description (glossaries)

Header for second column in the glossary (for 2, 3 or 4 column glossaries that support headers).

\symbolname

Symbol (glossaries)

Header for symbol column in the glossary for glossary styles that support this option.

\pagelistname

Page List (glossaries)

Header for page list column in the glossary for glossaries that support this option.

\glssymbolsgroupname

Symbols (glossaries)

Header for symbols section of the glossary for glossary styles that support this option.

\glsnumbersgroupname

Numbers (glossaries)

Header for numbers section of the glossary for glossary styles that support this option.


As from version 4.12, multilingual support is provided by separate language modules that need to be installed in addition to installing the glossaries package. You only need to install the modules for the languages that you require. If the language module has an unmaintained status, you can volunteer to take over the maintenance by contacting me at http://www.dickimaw-books.com/contact.html. The translator dictionary files for glossaries are now provided by the appropriate language module. For further details about information specific to a given language, please see the documentation for that language module.

Examples of use:

  • Using babel and translator:
    \documentclass[english,french]{article}  
    \usepackage{babel}  
    \usepackage{glossaries}

    (translator is automatically loaded).

  • Using babel:
    \documentclass[english,french]{article}  
    \usepackage{babel}  
    \usepackage[translate=babel]{glossaries}

    (translator isn’t loaded). The glossaries-extra package has translate=babel as the default if babel has been loaded.

  • Using polyglossia:
    \documentclass{article}  
    \usepackage{polyglossia}  
    \setmainlanguage{english}  
    \usepackage{glossaries}

Due to the varied nature of glossaries, it’s likely that the predefined translations may not be appropriate. If you are using the babel package and the glossaries package option translate=babel, you need to be familiar with the advice given in changing the words babel uses. If you are using the translator package, then you can provide your own dictionary with the necessary modifications (using \deftranslation) and load it using \usedictionary. If you simply want to change the title of a glossary, you can use the title key in commands like \printglossary (but not the iterative commands like \printglossaries).

Note that the translator dictionaries are loaded at the beginning of the document, so it won’t have any effect if you put \deftranslation in the preamble. It should be put in your personal dictionary instead (as in the example below). See the translator documentation for further details. (Now with beamer documentation.)

Your custom dictionary doesn’t have to be just a translation from English to another language. You may prefer to have a dictionary for a particular type of document. For example, suppose your institution’s in-house reports have to have the glossary labelled as “Nomenclature” and the page list should be labelled “Location”, then you can create a file called, say,

myinstitute-glossaries-dictionary-English.dict

that contains the following:

\ProvidesDictionary{myinstitute-glossaries-dictionary}{English}  
\deftranslation{Glossary}{Nomenclature}  
\deftranslation{Page List (glossaries)}{Location}

You can now load it using:

\usedictionary{myinstitute-glossaries-dictionary}

(Make sure that myinstitute-glossaries-dictionary-English.dict can be found by TeX.) If you want to share your custom dictionary, you can upload it to CTAN.

If you are using babel and don’t want to use the translator interface, you can use the package option translate=babel. For example:

\documentclass[british]{article}  
\usepackage{babel}  
\usepackage[translate=babel]{glossaries}  
\addto\captionsbritish{%  
    \renewcommand*{\glossaryname}{List of Terms}%  
    \renewcommand*{\acronymname}{List of Acronyms}%  
}

Note that xindy and bib2gls provide much better multi-lingual support than makeindex, so I recommend that you use Options 3 or 4 if you have glossary entries that contain non-Latin characters. See §11 Xindy (Option 3) for further details on xindy, and see the bib2gls user manual for further details of that application.

Creating a New Language Module

The glossaries package now uses the tracklang package to determine which language modules need to be loaded. If you want to create a new language module, you should first read the tracklang documentation.

To create a new language module, you need to at least create two files called: glossaries-lang.ldf and glossaries-dictionary-Lang.dict where ⟨lang⟩ is the root language name (for example, english) and ⟨Lang⟩ is the language name used by translator (for example, English).

Here’s an example of glossaries-dictionary-English.dict:

\ProvidesDictionary{glossaries-dictionary}{English}  
\providetranslation{Glossary}{Glossary}  
\providetranslation{Acronyms}{Acronyms}  
\providetranslation{Notation (glossaries)}{Notation}  
\providetranslation{Description (glossaries)}{Description}  
\providetranslation{Symbol (glossaries)}{Symbol}  
\providetranslation{Page List (glossaries)}{Page List}  
\providetranslation{Symbols (glossaries)}{Symbols}  
\providetranslation{Numbers (glossaries)}{Numbers}

You can use this as a template for your dictionary file. Change English to the translator name for your language (so that it matches the file name glossaries-dictionary-Lang.dict) and, for each \providetranslation, change the second argument to the appropriate translation.

Here’s an example of glossaries-english.ldf:

\ProvidesGlossariesLang{english}  
\glsifusedtranslatordict{English}  
{%  
  \addglossarytocaptions{\CurrentTrackedLanguage}%  
  \addglossarytocaptions{\CurrentTrackedDialect}%  
}  
{%  
  \@ifpackageloaded{polyglossia}%  
  {%  
    \newcommand*{\glossariescaptionsenglish}{%  
      \renewcommand*{\glossaryname}{\textenglish{Glossary}}%  
      \renewcommand*{\acronymname}{\textenglish{Acronyms}}%  
      \renewcommand*{\entryname}{\textenglish{Notation}}%  
      \renewcommand*{\descriptionname}{\textenglish{Description}}%  
      \renewcommand*{\symbolname}{\textenglish{Symbol}}%  
      \renewcommand*{\pagelistname}{\textenglish{Page List}}%  
      \renewcommand*{\glssymbolsgroupname}{\textenglish{Symbols}}%  
      \renewcommand*{\glsnumbersgroupname}{\textenglish{Numbers}}%  
    }%  
  }%  
  {%  
    \newcommand*{\glossariescaptionsenglish}{%  
      \renewcommand*{\glossaryname}{Glossary}%  
      \renewcommand*{\acronymname}{Acronyms}%  
      \renewcommand*{\entryname}{Notation}%  
      \renewcommand*{\descriptionname}{Description}%  
      \renewcommand*{\symbolname}{Symbol}%  
      \renewcommand*{\pagelistname}{Page List}%  
      \renewcommand*{\glssymbolsgroupname}{Symbols}%  
      \renewcommand*{\glsnumbersgroupname}{Numbers}%  
    }%  
  }%  
  \ifcsdef{captions\CurrentTrackedDialect}  
  {%  
    \csappto{captions\CurrentTrackedDialect}%  
    {%  
      \glossariescaptionsenglish  
    }%  
  }%  
  {%  
    \ifcsdef{captions\CurrentTrackedLanguage}  
    {  
      \csappto{captions\CurrentTrackedLanguage}%  
      {%  
        \glossariescaptionsenglish  
      }%  
    }%  
    {%  
    }%  
  }%  
  \glossariescaptionsenglish  
}  
\renewcommand*{\glspluralsuffix}{s}  
\renewcommand*{\glsacrpluralsuffix}{\glspluralsuffix}  
\renewcommand*{\glsupacrpluralsuffix}{\glstextup{\glspluralsuffix}}

This is a somewhat longer file, but again you can use it as a template. Replace English with the translator language label ⟨Lang⟩ used for the dictionary file and replace english with the root language name ⟨lang⟩. Within the definition of \glossariescaptionslang⟩, replace the English text (such as “Glossaries”) with the appropriate translation.

The suffixes used to generate the plural forms when the plural hasn’t been specified are given by \glspluralsuffix (for general entries). For abbreviations defined with \newacronym, \glsupacrpluralsuffix is used for acronyms where the suffix needs to be set using \glstextup to counteract the effects of \textsc and \glsacrpluralsuffix for other acronym styles. There’s no guarantee when these commands will be expanded. They may be expanded on definition or they may be expanded on use, depending on the glossaries configuration.

Therefore these plural suffix command definitions aren’t included in the caption mechanism as that’s typically not switched on until the start of the document. This means that the suffix in effect will be for the last loaded language that redefined these commands. It’s best to initialise these commands to the most common suffix required in your document and use the plural, longplural, shortplural etc keys to override exceptions.

If you want to add a regional variation, create a file called glossaries-iso lang-iso country.ldf, where ⟨iso lang⟩ is the ISO language code and ⟨iso country⟩ is the ISO country code. For example, glossaries-en-GB.ldf. This file can load the root language file and make the appropriate changes, for example:

 \ProvidesGlossariesLang{en-GB}  
 \RequireGlossariesLang{english}  
 \glsifusedtranslatordict{British}  
 {%  
   \addglossarytocaptions{\CurrentTrackedLanguage}%  
   \addglossarytocaptions{\CurrentTrackedDialect}%  
 }  
 {%  
   \@ifpackageloaded{polyglossia}%  
   {%  
     % Modify \glossariescaptionsenglish as appropriate for  
     % polyglossia  
   }%  
   {%  
     % Modify \glossariescaptionsenglish as appropriate for  
     % non-polyglossia  
   }%  
 }

If the translations includes non-Latin characters, it’s necessary to provide code that’s independent of the input encoding. Remember that while some users may use UTF-8, others may use Latin-1 or any other supported encoding, but while users won’t appreciate you enforcing your preference on them, it’s useful to provide a UTF-8 version for XeLaTeX users.

The glossaries-irish.ldf file provides this as follows:

\ProvidesGlossariesLang{irish}  
\glsifusedtranslatordict{Irish}  
{%  
  \addglossarytocaptions{\CurrentTrackedLanguage}%  
  \addglossarytocaptions{\CurrentTrackedDialect}%  
}  
{%  
  \ifdefstring{\inputencodingname}{utf8}  
  {\input{glossaries-irish-utf8.ldf}}%  
  {%  
    \ifdef{\XeTeXinputencoding}% XeTeX defaults to UTF-8  
    {\input{glossaries-irish-utf8.ldf}}%  
    {\input{glossaries-irish-noenc.ldf}}  
  }  
  \ifcsdef{captions\CurrentTrackedDialect}  
  {%  
    \csappto{captions\CurrentTrackedDialect}%  
    {%  
      \glossariescaptionsirish  
    }%  
  }%  
  {%  
    \ifcsdef{captions\CurrentTrackedLanguage}  
    {  
      \csappto{captions\CurrentTrackedLanguage}%  
      {%  
        \glossariescaptionsirish  
      }%  
    }%  
    {%  
    }%  
  }%  
  \glossariescaptionsirish  
}

(Again you can use this as a template. Replace irish with your root language label and Irish with the translator dictionary label.)

There are now two extra files: glossaries-irish-noenc.ldf (no encoding information) and glossaries-irish-utf8.ldf (UTF-8).

These both define \glossariescaptionsirish but the *-noenc.ldf uses LaTeX accent commands:

\@ifpackageloaded{polyglossia}%  
{%  
  \newcommand*{\glossariescaptionsirish}{%  
    \renewcommand*{\glossaryname}{\textirish{Gluais}}%  
    \renewcommand*{\acronymname}{\textirish{Acrainmneacha}}%  
    \renewcommand*{\entryname}{\textirish{Ciall}}%  
    \renewcommand*{\descriptionname}{\textirish{Tuairisc}}%  
    \renewcommand*{\symbolname}{\textirish{Comhartha}}%  
    \renewcommand*{\glssymbolsgroupname}{\textirish{Comhartha\'{\i}}}%  
    \renewcommand*{\pagelistname}{\textirish{Leathanaigh}}%  
    \renewcommand*{\glsnumbersgroupname}{\textirish{Uimhreacha}}%  
  }%  
}%  
{%  
  \newcommand*{\glossariescaptionsirish}{%  
    \renewcommand*{\glossaryname}{Gluais}%  
    \renewcommand*{\acronymname}{Acrainmneacha}%  
    \renewcommand*{\entryname}{Ciall}%  
    \renewcommand*{\descriptionname}{Tuairisc}%  
    \renewcommand*{\symbolname}{Comhartha}%  
    \renewcommand*{\glssymbolsgroupname}{Comhartha\'{\i}}%  
    \renewcommand*{\pagelistname}{Leathanaigh}%  
    \renewcommand*{\glsnumbersgroupname}{Uimhreacha}%  
  }%  
}

whereas the *-utf8.ldf replaces the accent commands with the appropriate UTF-8 characters.

Top

1.4 Generating the Associated Glossary Files

This section is only applicable if you have chosen Options 2 or 3. You can ignore this section if you have chosen any of the other options. If you want to alphabetically sort your entries always remember to use the sort key if the name contains any LaTeX commands (except if you’re using bib2gls).

If this section seriously confuses you, and you can’t work out how to run external tools like makeglossaries or makeindex, you can try using the automake package option, described in §2.5 Sorting Options, but you will need TeX’s shell escape enabled. See also Incorporating makeglossaries or makeglossaries-lite or bib2gls into the document build.

In order to generate a sorted glossary with compact number lists, it is necessary to use an external indexing application as an intermediate step (unless you have chosen Option 1, which uses TeX to do the sorting or Option 5, which doesn’t perform any sorting). It is this application that creates the file containing the code required to typeset the glossary. If this step is omitted, the glossaries will not appear in your document. The two indexing applications that are most commonly used with LaTeX are makeindex and xindy. As from version 1.17, the glossaries package can be used with either of these applications. Previous versions were designed to be used with makeindex only. With the glossaries-extra package, you can also use bib2gls as the indexing application. (See the glossaries-extra and bib2gls user manuals for further details.) Note that xindy and bib2gls have much better multi-lingual support than makeindex, so xindy or bib2gls are recommended if you’re not writing in English. Commands that only have an effect when xindy is used are described in §11 Xindy (Option 3).

This is a multi-stage process, but there are methods of automating document compilation using applications such as latexmk and arara. With arara you can just add special comments to your document source:

% arara: pdflatex
% arara: makeglossaries
% arara: pdflatex
With latexmk you need to set up the required dependencies.

The glossaries package comes with the Perl script makeglossaries which will run makeindex or xindy on all the glossary files using a customized style file (which is created by \makeglossaries). See §1.4.1 Using the makeglossaries Perl Script for further details. Perl is stable, cross-platform, open source software that is used by a number of TeX-related applications (including xindy and latexmk). Most Unix-like operating systems come with a Perl interpreter. TeX Live also comes with a Perl interpreter. MiKTeX doesn’t come with a Perl interpreter so if you are a Windows MiKTeX user you will need to install Perl if you want to use makeglossaries or xindy. Further information is available at http://www.perl.org/about.html and MiKTeX and Perl scripts (and one Python script).

The advantages of using makeglossaries:

  • It automatically detects whether to use makeindex or xindy and sets the relevant application switches.
  • One call of makeglossaries will run makeindex/xindy for each glossary type.
  • If things go wrong, makeglossaries will scan the messages from makeindex or xindy and attempt to diagnose the problem in relation to the glossaries package. This will hopefully provide more helpful messages in some cases. If it can’t diagnose the problem, you will have to read the relevant transcript file and see if you can work it out from the makeindex or xindy messages.
  • If makeindex warns about multiple encap values, makeglossaries will detect this and attempt to correct the problem.1.6 This correction is only provided by makeglossaries when makeindex is used since xindy uses the order of the attributes list to determine which format should take precedence. (see §11.2 Locations and Number lists.)

As from version 4.16, the glossaries package also comes with a Lua script called makeglossaries-lite. This is a trimmed-down alternative to the makeglossaries Perl script. It doesn’t have some of the options that the Perl version has and it doesn’t attempt to diagnose any problems, but since modern TeX distributions come with LuaTeX (and therefore have a Lua interpreter) you don’t need to install anything else in order to use makeglossaries-lite so it’s an alternative to makeglossaries if you want to use Option 2 (makeindex).

If things go wrong and you can’t work out why your glossaries aren’t being generated correctly, you can use makeglossariesgui as a diagnostic tool. Once you’ve fixed the problem, you can then go back to using makeglossaries or makeglossaries-lite.

Whilst I strongly recommended that you use the makeglossaries Perl script or the makeglossaries-lite Lua script, it is possible to use the glossaries package without using those applications. However, note that some commands and package options have no effect if you explicitly run makeindex/xindy. These are listed in table 1.3.

If you are choosing not to use makeglossaries because you don’t want to install Perl, you will only be able to use makeindex as xindy also requires Perl. (Other useful Perl scripts include epstopdf and latexmk, so it’s well-worth the effort to install Perl.)

Note that if any of your entries use an entry that is not referenced outside the glossary, you will need to do an additional makeglossaries, makeindex or xindy run, as appropriate. For example, suppose you have defined the following entries:1.7

\newglossaryentry{citrusfruit}{name={citrus fruit},  
description={fruit of any citrus tree. (See also  
\gls{orange})}}  
\newglossaryentry{orange}{name={orange},  
description={an orange coloured fruit.}}

and suppose you have \gls{citrusfruit} in your document but don’t reference the orange entry, then the orange entry won’t appear in your glossary until you first create the glossary and then do another run of makeglossaries, makeindex or xindy. For example, if the document is called myDoc.tex, then you must do:


   
pdflatex myDoc
makeglossaries myDoc
pdflatex myDoc
makeglossaries myDoc
pdflatex myDoc

(In the case of Option 4, bib2gls will scan the description for instances of commands like \gls to ensure they are selected but an extra bib2gls call is required to ensure the locations are included, if locations lists are required. See the and bib2gls manual for further details.)

Likewise, an additional makeglossaries and LaTeX run may be required if the document pages shift with re-runs. For example, if the page numbering is not reset after the table of contents, the insertion of the table of contents on the second LaTeX run may push glossary entries across page boundaries, which means that the number lists in the glossary may need updating.

The examples in this document assume that you are accessing makeglossaries, xindy or makeindex via a terminal. Windows users can use the MSDOS Prompt which is usually accessed via the Start->All Programs menu or Start->All Programs->Accessories menu.

Alternatively, your text editor may have the facility to create a function that will call the required application. See Incorporating makeglossaries or makeglossaries-lite or bib2gls into the document build.

If any problems occur, remember to check the transcript files (e.g. glg or alg) for messages.


Table 1.3: Commands and package options that have no effect when using xindy or makeindex explicitly
Command or Package Option makeindex xindy
order=letter use -l use -M ord/letorder
order=word default default
xindy={language=lang,codename=code} N/A use -Llang-Ccode
\GlsSetXdyLanguage{lang} N/A use -Llang
\GlsSetXdyCodePage{code} N/A use -Ccode

Top

1.4.1 Using the makeglossaries Perl Script

The makeglossaries script picks up the relevant information from the auxiliary (aux) file and will either call xindy or makeindex, depending on the supplied information. Therefore, you only need to pass the document’s name without the extension to makeglossaries. For example, if your document is called myDoc.tex, type the following in your terminal:


   
pdflatex myDoc
makeglossaries myDoc
pdflatex myDoc

You may need to explicitly load makeglossaries into Perl:


   
perl makeglossaries myDoc

Windows users: TeX Live on Windows has its own internal Perl interpreter and provides makeglossaries.exe as a convenient wrapper for the makeglossaries Perl script. MiKTeX also provides a wrapper makeglossaries.exe but doesn’t provide a Perl interpreter, which is still required even if you run MiKTeX’s makeglossaries.exe, so with MiKTeX you’ll need to install Perl.1.8 There’s more information about this at http://tex.stackexchange.com/q/158796/19862 on the TeX.SX site.

The makeglossaries script attempts to fork the makeindex/xindy process using open() on the piped redirection 2>&1 | and parses the processor output to help diagnose problems. If this method fails makeglossaries will print an “Unable to fork” warning and will retry without redirection. If you run makeglossaries on an operating system that doesn’t support this form of redirection, then you can use the -Q switch to suppress this warning or you can use the -k switch to make makeglossaries automatically use the fallback method without attempting the redirection. Without this redirection, the -q (quiet) switch doesn’t work as well.

You can specify in which directory the aux, glo etc files are located using the -d switch. For example:


   
pdflatex -output-directory myTmpDir myDoc
makeglossaries -d myTmpDir myDoc

Note that makeglossaries assumes by default that makeindex/xindy is on your operating system’s path. If this isn’t the case, you can specify the full pathname using -mpath/to/makeindex⟩ for makeindex or -xpath/to/xindy⟩ for xindy.

As from makeglossaries v2.18, if you are using makeindex, there’s a check for makeindex’s multiple encap warning. This is where different encap values (location formats) are used on the same location for the same entry. For example:

\documentclass{article}  
\usepackage{glossaries}  
\makeglossaries  
\newglossaryentry{sample}{name={sample},description={an example}}  
\begin{document}  
\gls{sample}, \gls[format=textbf]{sample}.  
\printglossaries  
\end{document}

If you explicitly use makeindex, this will cause a warning and the location list will be “1, 1”. That is, the page number will be repeated with each format. As from v2.18, makeglossaries will check for this warning and, if found, will attempt to correct the problem by removing duplicate locations and retrying. There’s no similar check for xindy as xindy won’t produce any warning and will simply discard duplicates.

For a complete list of options do makeglossaries --help.

When upgrading the glossaries package, make sure you also upgrade your version of makeglossaries. The current version is 4.45.

Top

1.4.2 Using the makeglossaries-lite Lua Script

The Lua alternative to the makeglossaries Perl script requires a Lua interpreter, which should already be available if you have a modern TeX distribution that includes LuaTeX. Lua is a light-weight, cross-platform scripting language, but because it’s light-weight it doesn’t have the full-functionality of heavy-weight scripting languages, such as Perl. The makeglossaries-lite script is therefore limited by this and some of the options available to the makeglossaries Perl script aren’t available here. (In particular the -d option.)

Note that TeX Live on Unix-like systems creates a symbolic link called makeglossaries-lite (without the lua extension) to the actual makeglossaries-lite.lua script, so you may not need to supply the extension.

The makeglossaries-lite script can be invoked in the same way as makeglossaries. For example, if your document is called myDoc.tex, then do


   
makeglossaries-lite.lua myDoc

or


   
makeglossaries-lite myDoc

Some of the options available with the Perl makeglossaries script are also available with the Lua makeglossaries-lite script. For a complete list of available options, do


   
makeglossaries-lite.lua --help

Top

1.4.3 Using xindy explicitly (Option 3)

Xindy comes with TeX Live. It has also been added to MikTeX, but if you don’t have it installed, see How to use Xindy with MikTeX on TeX on StackExchange.

If you want to use xindy to process the glossary files, you must make sure you have used the xindy package option:

\usepackage[xindy]{glossaries}

This is required regardless of whether you use xindy explicitly or whether it’s called implicitly via applications such as makeglossaries. This causes the glossary entries to be written in raw xindy format, so you need to use -I xindy not -I tex.

To run xindy type the following in your terminal (all on one line):


   
xindy -L ⟨language⟩ -C ⟨encoding⟩ -I xindy -M ⟨style⟩ -t ⟨base⟩.glg -o ⟨base⟩.gls ⟨base⟩.glo

where ⟨language⟩ is the required language name, ⟨encoding⟩ is the encoding, ⟨base⟩ is the name of the document without the tex extension and ⟨style⟩ is the name of the xindy style file without the xdy extension. The default name for this style file is ⟨basexdy but can be changed via \setStyleFile{style}. You may need to specify the full path name depending on the current working directory. If any of the file names contain spaces, you must delimit them using double-quotes.

For example, if your document is called myDoc.tex and you are using UTF8 encoding in English, then type the following in your terminal:


   
xindy -L english -C utf8 -I xindy -M myDoc -t myDoc.glg -o myDoc.gls myDoc.glo

Note that this just creates the main glossary. You need to do the same for each of the other glossaries (including the list of acronyms if you have used the acronym package option), substituting glg, gls and glo with the relevant extensions. For example, if you have used the acronym package option, then you would need to do:


   
xindy -L english -C utf8 -I xindy -M myDoc -t myDoc.alg -o myDoc.acr myDoc.acn

For additional glossaries, the extensions are those supplied when you created the glossary with \newglossary.

Note that if you use makeglossaries instead, you can replace all those calls to xindy with just one call to makeglossaries:


   
makeglossaries myDoc

Note also that some commands and package options have no effect if you use xindy explicitly instead of using makeglossaries. These are listed in table 1.3.

Top

1.4.4 Using makeindex explicitly (Option 2)

If you want to use makeindex explicitly, you must make sure that you haven’t used the xindy package option or the glossary entries will be written in the wrong format. To run makeindex, type the following in your terminal:


   
makeindex -s ⟨style⟩.ist -t ⟨base⟩.glg -o ⟨base⟩.gls ⟨base⟩.glo

where ⟨base⟩ is the name of your document without the tex extension and ⟨styleist is the name of the makeindex style file. By default, this is ⟨baseist, but may be changed via \setStyleFile{style}. Note that there are other options, such as -l (letter ordering). See the makeindex manual for further details.

For example, if your document is called myDoc.tex, then type the following at the terminal:


   
makeindex -s myDoc.ist -t myDoc.glg -o myDoc.gls myDoc.glo

Note that this only creates the main glossary. If you have additional glossaries (for example, if you have used the acronym package option) then you must call makeindex for each glossary, substituting glg, gls and glo with the relevant extensions. For example, if you have used the acronym package option, then you need to type the following in your terminal:


   
makeindex -s myDoc.ist -t myDoc.alg -o myDoc.acr myDoc.acn

For additional glossaries, the extensions are those supplied when you created the glossary with \newglossary.

Note that if you use makeglossaries instead, you can replace all those calls to makeindex with just one call to makeglossaries:


   
makeglossaries myDoc

Note also that some commands and package options have no effect if you use makeindex explicitly instead of using makeglossaries. These are listed in table 1.3.

Top

1.5 Note to Front-End and Script Developers

The information needed to determine whether to use xindy, makeindex or bib2gls is stored in the aux file. This information can be gathered by a front-end, editor or script to make the glossaries where appropriate. This section describes how the information is stored in the auxiliary file.

Top

1.5.1 MakeIndex and Xindy

The file extensions used by each defined (but not ignored) glossary are given by


\@newglossary{label}{log}{out-ext}{in-ext}

where ⟨in-ext⟩ is the extension of the indexing application’s input file (the output file from the glossaries package’s point of view), ⟨out-ext⟩ is the extension of the indexing application’s output file (the input file from the glossaries package’s point of view) and ⟨log⟩ is the extension of the indexing application’s transcript file. The label for the glossary is also given. This isn’t required with makeindex, but with xindy it’s needed to pick up the associated language and encoding (see below). For example, the information for the default main glossary is written as:

\@newglossary{main}{glg}{gls}{glo}

If glossaries-extra’s hybrid method has been used (with \makeglossaries[sub-list]), then the sub-list of glossaries that need to be processed will be identified with:

\glsxtr@makeglossaries{list}

The indexing application’s style file is specified by


\@istfilename{filename}

The file extension indicates whether to use makeindex (ist) or xindy (xdy). Note that the glossary information is formatted differently depending on which indexing application is supposed to be used, so it’s important to call the correct one.

For example, with arara you can easily determine whether to run makeglossaries:

% arara: makeglossaries if found("aux", "@istfilename")
It’s more complicated if you want to explicitly run makeindex or xindy

Note that if you choose to explicitly call makeindex or xindy then the user will miss out on the diagnostic information and the encap-clash fix that makeglossaries also provides.

Word or letter ordering is specified by:


\@glsorder{order}

where ⟨order⟩ can be either word or letter.

If xindy should be used, the language and code page for each glossary is specified by


\@xdylanguage{label}{language}
\@gls@codepage{label}{code}

where ⟨label⟩ identifies the glossary, ⟨language⟩ is the root language (e.g. english) and ⟨code⟩ is the encoding (e.g. utf8). These commands are omitted if makeindex should be used.

If Option 1 has been used, the aux file will contain


\@gls@reference{type}{label}{location}

for every time an entry has been referenced.

Top

1.5.2 Entry Labels

If you need to gather labels for auto-completion, the writeglslabels package option will create a file containing the labels of all defined entries (regardless of whether or not the entry has been used in the document). The glossaries-extra package also provides docdef=atom, which will create the glsdefs file but will act like docdef=restricted.

Top

1.5.3 Bib2Gls

bib2glsIf Option 4 has been used, the aux file will contain one or more instances of

\glsxtr@resource{options}{basename}
where ⟨basename⟩ is the basename of the glstex file that needs to be created by bib2gls. If src= {bib list} isn’t present in ⟨options⟩ then ⟨basename⟩ also indicates the name of the associated bib file.

For example, with arara you can easily determine whether or not to run bib2gls:

% arara: bib2gls if found("aux", "glsxtr@resource")
(It gets more complicated if both \glsxtr@resource and \@istfilename are present as that indicates the hybrid record=alsoindex option.)

Remember that with bib2gls, the entries will never be defined on the first LaTeX call (because their definitions are contained in the glstex file created by bib2gls). You can also pick up labels from the records in aux file, which will be in the form:

\glsxtr@record{label}{h-prefix}{counter}{format}{loc}
or (with record=nameref)
\glsxtr@record@nameref{label}{href prefix}{counter}{format}{location}{title} {href anchor}{href value}
or (with \glssee)
\glsxtr@recordsee{label}{xr list}
You can also pick up the commands defined with \glsxtrnewglslike, which are added to the aux file for bib2gls’s benefit:
\@glsxtr@newglslike{label-prefix}{cs}
If \GlsXtrSetAltModifier is used, then the modifier is identified with:
\@glsxtr@altmodifier{character}
Label prefixes (for the \dgls set of commands) are identified with:
\@glsxtr@prefixlabellist{list}

Top

2. Package Options

This section describes the available glossaries package options. You may omit the =true for boolean options. (For example, acronym is equivalent to acronym=true).

The glossaries-extra package has additional options described in the glossaries-extra manual. The extension package also has some different default settings to the base package. Those that are available at the time of writing are included here. Future versions of glossaries-extra may have additional package options or new values for existing settings that aren’t listed here.

Note that ⟨key⟩=⟨value⟩ package options can’t be passed via the document class options. (This includes options where the ⟨value⟩ part may be omitted, such as acronym.) This is a general limitation not restricted to the glossaries package. Options that aren’t ⟨key⟩=⟨value⟩ (such as makeindex) may be passed via the document class options.

Top

2.1 General Options

nowarn

This suppresses all warnings generated by the glossaries package. Don’t use this option if you’re new to using glossaries as the warnings are designed to help detect common mistakes (such as forgetting to use \makeglossaries). Note that if you use debug with any value other than false it will override this option.

nolangwarn

This suppresses the warning generated by a missing language module.

noredefwarn

If you load glossaries with a class or another package that already defines glossary related commands, by default glossaries will warn you that it’s redefining those commands. If you are aware of the consequences of using glossaries with that class or package and you don’t want to be warned about it, use this option to suppress those warnings. Other warnings will still be issued unless you use the nowarn option described above. (This option is automatically switched on by glossaries-extra.)

debug={value}

Introduced in version 4.24. The default setting is debug=false. The following values are available: false, true, showtargets (v4.32+) and showaccsupp (v4.45+). If no value is given, debug=true is assumed.

The glossaries-extra package provides extra values showwrgloss, that may be used to show where the indexing is occurring, and all, which switches on all debugging options. See the glossaries-extra manual for further details.

All values other than debug=false switch on the debug mode (and will automatically cancel the nowarn option). The debug=showtargets option will additionally use:


\glsshowtarget{target name}

to show the hypertarget or hyperlink name when \glsdohypertarget is used by commands like \glstarget and when \glsdohyperlink is used by commands like \gls. In math mode or inner mode, this puts the target name in square brackets just before the link or anchor. In outer mode it uses:


\glsshowtargetouter{label}

which by default places the target name in the margin. The font is given by:


\glsshowtargetfont

The default definition is \ttfamily\small. This command is included in the definition of \glsshowtargetouter, so if you want to redefine that command remember to include the font. For example:

\renewcommand*{\glsshowtargetouter}[1]{%  
 {\glsshowtargetfont [#1]}}

Similarly, the debug=showaccsupp will add the accessibility support information using:


\glsshowaccsupp{options}{tag}{replacement text}

This internally uses \glsshowtarget. This option is provided for use with glossaries-accsupp.

The purpose of the debug mode can be demonstrated with the following example document:

\documentclass{article}
\usepackage{glossaries}
\newglossaryentry{sample1}{name={sample1},description={example}}
\newglossaryentry{sample2}{name={sample2},description={example}}
\glsadd{sample2}% <- does nothing here
\makeglossaries
\begin{document}
\gls{sample1}.
\printglossaries
\end{document}
In this case, only the sample1 entry has been indexed, even though \glsadd{sample2} appears in the source code. This is because \glsadd{sample2} has been used before the associated file is opened by \makeglossaries. Since the file isn’t open yet, the information can’t be written to it, which is why the sample2 entry doesn’t appear in the glossary.

Without \makeglossaries the indexing is suppressed with Options 2 and 3 but, other than that, commands like \gls behave as usual.

This situation doesn’t cause any errors or warnings as it’s perfectly legitimate for a user to want to use glossaries to format the entries (e.g. abbreviation expansion) but not display any lists of terms, abbreviations, symbols etc (or the user may prefer to use the unsorted Options 5 or 6). It’s also possible that the user may want to temporarily comment out \makeglossaries in order to suppress the indexing while working on a draft version to speed compilation, or the user may prefer to use Options 1 or 4 for indexing, which don’t use \makeglossaries.

Therefore \makeglossaries can’t be used to enable \newglossaryentry and commands like \gls and \glsadd. These commands must be enabled by default. (It does, however, enable the see key as that’s a more common problem. See below.)

The debug mode, enabled with the debug option,

\usepackage[debug]{glossaries}

will write information to the log file when the indexing can’t occur because the associated file isn’t open. The message is written in the form

Package glossaries Info: wrglossary(⟨type⟩)(⟨text⟩) on input line ⟨line number⟩.

where ⟨type⟩ is the glossary label and ⟨text⟩ is the line of text that would’ve been written to the associated file if it had been open. So if any entries haven’t appeared in the glossary but you’re sure you used commands like \glsadd or \glsaddall, try switching on the debug option and see if any information has been written to the log file.

savewrites={boolean}

This is a boolean option to minimise the number of write registers used by the glossaries package. The default is savewrites=false. With Options 2 and 3, one write register is required per (non-ignored) glossary and one for the style file.

With all options except Options 1 and 4, another write register is required if the docdefs file is needed to save document definitions. With both Options 1 and 4, no write registers are required (document definitions aren’t permitted and indexing information is written to the aux file). If you really need document definitions but you want to minimise the number of write registers then consider using docdef=restricted with glossaries-extra.

There are only a limited number of write registers, and if you have a large number of glossaries or if you are using a class or other packages that create a lot of external files, you may exceed the maximum number of available registers. If savewrites is set, the glossary information will be stored in token registers until the end of the document when they will be written to the external files.

This option can significantly slow document compilation and may cause the indexing to fail. Page numbers in the number list will be incorrect on page boundaries due to TeX’s asynchronous output routine. As an alternative, you can use the scrwfile package (part of the KOMA-Script bundle) and not use this option.

By way of comparison, sample-multi2.tex provided with bib2gls has a total of 15 glossaries. With Options 2 or 3, this would require 46 associated files and 16 write registers.2.1 With bib2gls, no write registers are required and there are only 10 associated files for that particular document (9 resource files and 1 transcript file).

If you want to use TeX’s \write18 mechanism to call makeindex or xindy from your document and use savewrites, you must create the external files with \glswritefiles before you call makeindex/xindy. Also set \glswritefiles to nothing or \relax before the end of the document to avoid rewriting the files. For example:

\glswritefiles  
\write18{makeindex -s \istfilename\space  
-t \jobname.glg -o \jobname.gls \jobname}  
\let\glswritefiles\relax

In general, this package option is best avoided.

translate={value}

This can take the following values:

true
If babel has been loaded and the translator package is installed, translator will be loaded and the translations will be provided by the translator package interface. You can modify the translations by providing your own dictionary. If the translator package isn’t installed and babel is loaded, the glossaries-babel package will be loaded and the translations will be provided using babel’s \addto\captionlanguage⟩ mechanism. If polyglossia has been loaded, glossaries-polyglossia will be loaded.
false
Don’t provide translations, even if babel or polyglossia has been loaded. (Note that babel provides the command \glossaryname so that will still be translated if you have loaded babel.)
babel
Don’t load the translator package. Instead load glossaries-babel.

I recommend you use translate=babel if you have any problems with the translations or with PDF bookmarks, but to maintain backward compatibility, if babel has been loaded the default is translate=true.

If translate is specified without a value, translate=true is assumed. If translate isn’t specified, translate=true is assumed if babel, polyglossia or translator have been loaded. Otherwise translate=false is assumed.

With glossaries-extra, if babel is detected then translate=babel is automatically passed to the base glossaries package.

See §1.3.1 Changing the Fixed Names for further details.

notranslate

This is equivalent to translate=false and may be passed via the document class options.

hyperfirst={boolean}

This is a boolean option that specifies whether each term has a hyperlink on first use. The default is hyperfirst=true (terms on first use have a hyperlink, unless explicitly suppressed using starred versions of commands such as \gls* or by identifying the glossary with nohypertypes, described above). Note that nohypertypes overrides hyperfirst=true. This option only affects commands that check the first use flag, such as the \gls-like commands (for example, \gls or \glsdisp), but not the \glstext-like commands (for example, \glslink or \glstext).

The hyperfirst setting applies to all glossary types (unless identified by nohypertypes or defined with \newignoredglossary). It can be overridden on an individual basis by explicitly setting the hyper key when referencing an entry (or by using the plus or starred version of the referencing command).

It may be that you only want to apply this to just the acronyms (where the first use explains the meaning of the acronym) but not for ordinary glossary entries (where the first use is identical to subsequent uses). In this case, you can use hyperfirst=false and apply \glsunsetall to all the regular (non-acronym) glossaries. For example:

 \usepackage[acronym,hyperfirst=false]{glossaries}  
 % acronym and glossary entry definitions  
 % at the end of the preamble  
 \glsunsetall[main]

Alternatively you can redefine the hook


\glslinkcheckfirsthyperhook

which is used by the commands that check the first use flag, such as \gls. Within the definition of this command, you can use \glslabel to reference the entry label and \glstype to reference the glossary type. You can also use \ifglsused to determine if the entry has been used. You can test if an entry is an acronym by checking if it has the long key set using \ifglshaslong. For example, to switch off the hyperlink on first use just for acronyms:

\renewcommand*{\glslinkcheckfirsthyperhook}{%  
 \ifglsused{\glslabel}{}%  
 {%  
   \ifglshaslong{\glslabel}{\setkeys{glslink}{hyper=false}}{}%  
 }%  
}

Note that this hook isn’t used by the commands that don’t check the first use flag, such as \glstext. (You can, instead, redefine \glslinkpostsetkeys, which is used by both the \gls-like and \glstext-like commands.)

The glossaries-extra package provides a method of disabling the first use hyperlink according to the entry’s associated category. For example, if you only to switch off the first use hyperlink for abbreviations and acronyms then you simply need to set the nohyperfirst attribute for the abbreviation and acronym categories. (Instead of using the nohyperfirst package option.) See the glossaries-extra manual for further details.

writeglslabels

This is a valueless option that will create a file called \jobname.glslabels at the end of the document. This file simply contains a list of all defined entry labels (including those in any ignored glossaries). It’s provided for the benefit of text editors that need to know labels for auto-completion. (See also glossaries-extra’s docdef=atom package option.)

Note that with bib2gls the file will only contain the entries that bib2gls has selected from the bib files.

undefaction={value} (glossaries-extra.sty)

The value may be one of:

error
generate an error if a referenced entry is undefined (default, and only available setting with just glossaries);
warn
only warn if a referenced entry is undefined (automatically activated with Option 4).

docdef={value} (glossaries-extra.sty)

This option governs the use of \newglossaryentry. Available values:

false
\newglossaryentry is not permitted in the document environment (default with glossaries-extra and for Option 1 with just the base glossaries package);
restricted
\newglossaryentry is only permitted in the document environment if it occurs before \printglossary (not available for some indexing options);
atom
as restricted but creates the docdefs file for use by atom (without the limitations of docdef=true);
true
\newglossaryentry is permitted in the document environment where it would normally be permitted by the base glossaries package. This will create the docdefs file if \newglossaryentry is found in the document environment.

Top

2.2 Sectioning, Headings and TOC Options

toc={boolean}

Add the glossaries to the table of contents. Note that an extra LaTeX run is required with this option. Alternatively, you can switch this function on and off using


\glstoctrue

and


\glstocfalse

The default value is toc=false for the base glossaries package and toc=true for glossaries-extra.

numberline={boolean}

When used with the above toc=true option, this will add \numberline{} in the final argument of \addcontentsline. This will align the table of contents entry with the numbered section titles. Note that this option has no effect if the toc option is omitted. If toc is used without numberline, the title will be aligned with the section numbers rather than the section titles.

section={value}

This option indicates the sectional unit to use for the glossary. The value should be the control sequence name without the leading backslash or following star (e.g. just chapter not \chapter or chapter*).

The default behaviour is for the glossary heading to use \chapter, if that command exists, or \section otherwise. The starred or unstarred form is determined by the numberedsection option.

Example:

\usepackage[section=subsection]{glossaries}

You can omit the value if you want to use \section, i.e. 

\usepackage[section]{glossaries}

is equivalent to

\usepackage[section=section]{glossaries}

You can change this value later in the document using


\setglossarysection{name}

where ⟨name⟩ is the sectional unit.

The start of each glossary adds information to the page header via


\glsglossarymark{glossary title}

By default this uses \@mkboth2.2 but you may need to redefine it. For example, to only change the right header:

\renewcommand{\glsglossarymark}[1]{\markright{#1}}

or to prevent it from changing the headers:

\renewcommand{\glsglossarymark}[1]{}

If you want \glsglossarymark to use \MakeUppercase in the header, use the ucmark option described below.

Occasionally you may find that another package defines \cleardoublepage when it is not required. This may cause an unwanted blank page to appear before each glossary. This can be fixed by redefining \glsclearpage:

\renewcommand*{\glsclearpage}{\clearpage}

ucmark={boolean}

This is a boolean option. The default is ucmark=false, unless memoir has been loaded, in which case the default is ucmark=true.

If set, \glsglossarymark uses \MakeTextUppercase2.3. You can test whether this option has been set or not using


\ifglsucmarktrue part\else false part\fi

For example:

\renewcommand{\glsglossarymark}[1]{%  
  \ifglsucmark  
    \markright{\MakeTextUppercase{#1}}%  
  \else  
    \markright{#1}%  
  \fi}

If memoir has been loaded and ucmark is set, then memoir’s \memUChead is used.

numberedsection={value}

The glossaries are placed in unnumbered sectional units by default, but this can be changed using numberedsection. This option can take one of the following values:

Top

2.3 Glossary Appearance Options

savenumberlist={boolean}

This is a boolean option that specifies whether or not to gather and store the number list for each entry. The default is savenumberlist=false. (See \glsentrynumberlist and \glsdisplaynumberlist in §9 Using Glossary Terms Without Links.) This is always true if you use Option 1.

If you use the record option (with either no value or record=only or record=nameref) then this package option has no effect. With bib2gls, the number lists are automatically saved with the default save-locations=true and save-loclist=true resource settings.

entrycounter={boolean}

This is a boolean option. (Default is entrycounter=false.) If set, each main (level 0) glossary entry will be numbered when using the standard glossary styles. This option creates the counter glossaryentry.

If you use this option, you can reference the entry number within the document using


\glsrefentry{label}

where ⟨label⟩ is the label associated with that glossary entry. The labelling systems uses ⟨prefix⟩⟨label⟩, where ⟨label⟩ is the entry’s label and ⟨prefix⟩ is given by


\GlsEntryCounterLabelPrefix

(which defaults to glsentry-).

If you use \glsrefentry, you must run LaTeX twice after creating the glossary files using makeglossaries, makeindex or xindy to ensure the cross-references are up-to-date.

counterwithin={value}

This is a ⟨key⟩=⟨value⟩ option where ⟨value⟩ is the name of a counter. If used, this option will automatically set entrycounter=true and the glossaryentry counter will be reset every time ⟨value⟩ is incremented.

The glossaryentry counter isn’t automatically reset at the start of each glossary, except when glossary section numbering is on and the counter used by counterwithin is the same as the counter used in the glossary’s sectioning command.

If you want the counter reset at the start of each glossary, you can modify the glossary preamble (\glossarypreamble) to use


\glsresetentrycounter

which sets glossaryentry to zero:

\renewcommand{\glossarypreamble}{%  
  \glsresetentrycounter  
}

or if you are using \setglossarypreamble, add it to each glossary preamble, as required. For example:

\setglossarypreamble[acronym]{%  
  \glsresetentrycounter  
  The preamble text here for the list of acronyms.  
}  
\setglossarypreamble{%  
  \glsresetentrycounter  
  The preamble text here for the main glossary.  
}

subentrycounter={boolean}

This is a boolean option. (Default is subentrycounter=false.) If set, each level 1 glossary entry will be numbered when using the standard glossary styles. This option creates the counter glossarysubentry. The counter is reset with each main (level 0) entry. Note that this package option is independent of entrycounter. You can reference the number within the document using \glsrefentry{label} where ⟨label⟩ is the label associated with the sub-entry.

style={value}

This is a ⟨key⟩=⟨value⟩ option. (Default is style=list, unless classicthesis has been loaded, in which case the default is style=index.) Its value should be the name of the glossary style to use. This key may only be used for styles defined in glossary-list, glossary-long, glossary-super or glossary-tree. Alternatively, you can set the style using


\setglossarystyle{style name}

(See §15 Glossary Styles for further details.)

nolong

This prevents the glossaries package from automatically loading glossary-long (which means that the longtable package also won’t be loaded). This reduces overhead by not defining unwanted styles and commands. Note that if you use this option, you won’t be able to use any of the glossary styles defined in the glossary-long package (unless you explicitly load glossary-long).

nosuper

This prevents the glossaries package from automatically loading glossary-super (which means that the supertabular package also won’t be loaded). This reduces overhead by not defining unwanted styles and commands. Note that if you use this option, you won’t be able to use any of the glossary styles defined in the glossary-super package (unless you explicitly load glossary-super).

nolist

This prevents the glossaries package from automatically loading glossary-list. This reduces overhead by not defining unwanted styles. Note that if you use this option, you won’t be able to use any of the glossary styles defined in the glossary-list package (unless you explicitly load glossary-list). Note that since the default style is list (unless classicthesis has been loaded), you will also need to use the style option to set the style to something else.

notree

This prevents the glossaries package from automatically loading glossary-tree. This reduces overhead by not defining unwanted styles. Note that if you use this option, you won’t be able to use any of the glossary styles defined in the glossary-tree package (unless you explicitly load glossary-tree). Note that if classicthesis has been loaded, the default style is index, which is provided by glossary-tree.

nostyles

This prevents all the predefined styles from being loaded. If you use this option, you need to load a glossary style package (such as glossary-mcols). Also if you use this option, you can’t use the style package option. Instead you must either use \setglossarystyle{style} or the style key in the optional argument to \printglossary. Example:

\usepackage[nostyles]{glossaries}  
\usepackage{glossary-mcols}  
\setglossarystyle{mcoltree}

nonumberlist

This option will suppress the associated number lists in the glossaries (see also §5 Number lists). Note that if you use Options 2 or 3 (makeindex or xindy) then the locations must still be valid. This package option merely prevents the number list from being displayed, but both makeindex and xindy still require a location or cross-reference for each term that’s indexed. Remember that number list includes any cross-references, so suppressing the number list will also hide the cross-references (see below).

seeautonumberlist

If you suppress the number lists with nonumberlist, described above, this will also suppress any cross-referencing information supplied by the see key in \newglossaryentry or \glssee. If you use seeautonumberlist, the see key will automatically implement nonumberlist=false for that entry. (Note this doesn’t affect \glssee.) For further details see §8 Cross-Referencing Entries.

counter={value}

This is a ⟨key⟩=⟨value⟩ option. (Default is counter=page.) The value should be the name of the default counter to use in the number lists (see §5 Number lists).

nopostdot={boolean}

This is a boolean option. If no value is specified, true is assumed. When set to true, this option suppresses the default post description dot used by some of the predefined styles.

The default setting is nopostdot=false for the base glossaries package and nopostdot=true for glossaries-extra.

The glossaries-extra package provides postdot, which is equivalent to nopostdot=false, and also postpunc, which allows you to choose a different punctuation character.

nogroupskip={boolean}

This is a boolean option. If no value is specified, true is assumed. When set to true, this option suppresses the default vertical gap between letter groups used by some of the predefined styles. The default setting is nogroupskip=false.

If you are using bib2gls without the --group (or -g) switch then you don’t need to use nogroupskip=true as there won’t be any letter groups.

stylemods={list} (glossaries-extra.sty)

Load the glossaries-extra-stylemods package and patch the predefined styles. The ⟨list⟩ argument is optional. If present, this will also load glossary-element.sty for each ⟨element⟩ in the comma-separated ⟨list⟩.

Top

2.4 Indexing Options

seenoindex={value}

Introduced in version 4.24, this option may take one of three values: error, warn or ignore. The see key automatically indexes the cross-referenced entry using \glsadd. This means that if this key is used in an entry definition before the relevant glossary file has been opened, the indexing can’t be performed. Since this is easy to miss, the glossaries package by default issues an error message if the see key is used before \makeglossaries. This option allows you to change the error into just a warning (seenoindex=warn) or ignore it (seenoindex=ignore) if, for example, you want to temporarily comment out \makeglossaries to speed up the compilation of a draft document by omitting the indexing.

esclocations={boolean}

This is a boolean option. The default is esclocations=true, which is needed for Options 2 and 3. With Option 1 \makenoidxglossaries changes it to esclocations=false. With Option 4 (bib2gls), this setting is ignored.

Both makeindex and xindy are fussy about the location formats (makeindex more so than xindy) so the glossaries package tries to ensure that special characters are escaped and allows for the location to be substituted for a format that’s more acceptable to the indexing application. This requires a bit of trickery to circumvent the problem posed by TeX’s asynchronous output routine, which can go wrong and also adds to the complexity of the document build.

If you’re sure that your locations will always expand to an acceptable format (or you’re prepared to post-process the glossary file before passing it to the relevant indexing application) then use esclocations=false to avoid the complex escaping of location values. (See “Writing information to associated files” in the documented code for further details.)

This isn’t an issue for Options 1 or 4 as the locations are written to the aux file so no syntax conversion is required.

indexonlyfirst={boolean}

This is a boolean option that specifies whether to only add information to the external glossary file on first use. The default is indexonlyfirst=false, which will add a line to the file every time one of the \gls-like or \glstext-like commands are used. Note that \glsadd will always add information to the external glossary file2.4 (since that’s the purpose of that command).

Resetting the first use flag with commands like \glsreset after an entry has been indexed will cause that entry to be indexed multiple times if it’s used again after the reset. Likewise unsetting the first use flag before an entry has been indexed will prevent it from being indexed (unless specifically indexed with \glsadd).

You can customise this by redefining


\glswriteentry{label}{wr-code}

where ⟨label⟩ is the entry’s label and ⟨wr-code⟩ is the code that writes the entry’s information to the external file. The default definition of \glswriteentry is:

\newcommand*{\glswriteentry}[2]{%  
  \ifglsindexonlyfirst  
    \ifglsused{#1}{}{#2}%  
  \else  
    #2%  
  \fi  
}

This checks the indexonlyfirst package option (using \ifglsindexonlyfirst) and does ⟨wr-code⟩ if this is false otherwise it only does ⟨wr-code⟩ of the entry hasn’t been used.

For example, suppose you only want to index the first use for entries in the acronym glossary and not in the main (or any other) glossary:

\renewcommand*{\glswriteentry}[2]{%  
 \ifthenelse{\equal{\glsentrytype{#1}}{acronym}}  
 {\ifglsused{#1}{}{#2}}%  
 {#2}%  
}

Here I’ve used \ifthenelse to ensure the arguments of \equal are fully expanded before the comparison is made.

With the glossaries-extra package it’s possible to only index first use for particular categories. For example, if you only want this enabled for abbreviations and acronyms then you can set the indexonlyfirst attribute for the abbreviation and acronym categories. (Instead of using the indexonlyfirst package option.) See the glossaries-extra manual for further details.

indexcrossrefs={boolean} (glossaries-extra.sty)

If true, this will automatically index any cross-referenced entries that haven’t been marked as used at the end of the document. Increases document build time. See glossaries-extra manual for further details.

autoseeindex={boolean} (glossaries-extra.sty)

If true, makes the see and seealso keys automatically index the cross-reference when the entry is defined (default, and the only option with just the base glossaries package).

record={value} (glossaries-extra.sty)

If not off, this option indicates that bib2gls is required. If the value is omitted, only is assumed. Permitted values:

off
bib2gls isn’t being used;
only
bib2gls is being used to fetch entries from a bib file, to sort the entries and collate the number lists, where the location information is the same as for Options 13;
nameref
like only but provides extra information that allows the associated title to be used instead of the location number;
alsoindex
a hybrid approach where bib2gls is begin used to fetch entries from a bib file but makeindex or xindy are used for the indexing. This requires a more complicated document build and isn’t recommended.

See glossaries-extra manual for further details.

equations={boolean} (glossaries-extra.sty)

If true, this option will cause the default location counter to automatically switch to equation when inside a numbered equation environment.

floats={boolean} (glossaries-extra.sty)

If true, this option will cause the default location counter to automatically switch to the corresponding counter when inside a float. (Remember that with floats it’s the \caption command that increments the counter so the location will be incorrect if an entry is indexed within the float before the caption.)

indexcounter (glossaries-extra.sty)

This valueless option is primarily intended for use with bib2gls and hyperref allowing the page location hyperlink to the relevant point in the page (rather than the top of the page). Unexpected results will occur with other indexing methods. See glossaries-extra manual for further details.

Top

2.5 Sorting Options

This section is mostly for Options 2 and 3. Only the sort and order options are applicable for Option 1.

With Options 46, only sort=none is applicable (and this is automatically implemented by record=only and record=nameref). With bib2gls, the sort method is provided in the optional argument of \GlsXtrLoadResources not with the sort package option. There’s no sorting with Options 5 and 6.

sanitizesort={boolean}

This is a boolean option that determines whether or not to sanitize the sort value when writing to the external glossary file. For example, suppose you define an entry as follows:

\newglossaryentry{hash}{name={\#},sort={#},  
 description={hash symbol}}

The sort value (#) must be sanitized before writing it to the glossary file, otherwise LaTeX will try to interpret it as a parameter reference. If, on the other hand, you want the sort value expanded, you need to switch off the sanitization. For example, suppose you do:

\newcommand{\mysortvalue}{AAA}  
\newglossaryentry{sample}{%  
  name={sample},  
  sort={\mysortvalue},  
  description={an example}}

and you actually want \mysortvalue expanded, so that the entry is sorted according to AAA, then use the package option sanitizesort=false.

The default for Options 2 and 3 is sanitizesort=true, and the default for Option 1 is sanitizesort=false.

sort={value}

If you use Options 2 or 3, this package option is the only way of specifying how to sort the glossaries. Only Option 1 allows you to specify sort methods for individual glossaries via the sort key in the optional argument of \printnoidxglossary. If you have multiple glossaries in your document and you are using Option 1, only use the package options sort=def or sort=use if you want to set this sort method for all your glossaries.

This is a ⟨key⟩=⟨value⟩ option where ⟨value⟩ may be one of the following:

Note that the group styles (such as listgroup) are incompatible with the sort=use and sort=def options.

The default is sort=standard. When the standard sort option is in use, you can hook into the sort mechanism by redefining:


\glsprestandardsort{sort cs}{type}{label}

where ⟨sort cs⟩ is a temporary control sequence that stores the sort value (which was either explicitly set via the sort key or implicitly set via the name key) before any escaping of the makeindex/xindy special characters is performed. By default \glsprestandardsort just does:


\glsdosanitizesort

which sanitizes ⟨sort cs⟩ if the sanitizesort package option is set (or does nothing if the package option sanitizesort=false is used).

The other arguments, ⟨type⟩ and ⟨label⟩, are the glossary type and the entry label for the current entry. Note that ⟨type⟩ will always be a control sequence, but ⟨label⟩ will be in the form used in the first argument of \newglossaryentry.

Redefining \glsprestandardsort won’t affect any entries that have already been defined and will have no effect at all if you are using sort=def or sort=use.

Example 1 (Mixing Alphabetical and Order of Definition Sorting)

Suppose I have three glossaries: main, acronym and notation, and let’s suppose I want the main and acronym glossaries to be sorted alphabetically, but the notation type should be sorted in order of definition.

For Option 1, the sort option can be used in \printnoidxglossary:

\printnoidxglossary[sort=word]  
\printnoidxglossary[type=acronym,sort=word]  
\printnoidxglossary[type=notation,sort=def]

For Options 2 or 3, I can set the sort to standard (which is the default, but can be explicitly set via the package option sort=standard), and I can either define all my main and acronym entries, then redefine \glsprestandardsort to set ⟨sort cs⟩ to an incremented integer, and then define all my notation entries. Alternatively, I can redefine \glsprestandardsort to check for the glossary type and only modify ⟨sort cs⟩ if ⟨type⟩ is notation.

The first option can be achieved as follows:

\newcounter{sortcount}  
\renewcommand{\glsprestandardsort}[3]{%  
  \stepcounter{sortcount}%  
  \edef#1{\glssortnumberfmt{\arabic{sortcount}}}%  
}

The second option can be achieved as follows:

\newcounter{sortcount}  
\renewcommand{\glsprestandardsort}[3]{%  
  \ifdefstring{#2}{notation}%  
  {%  
     \stepcounter{sortcount}%  
     \edef#1{\glssortnumberfmt{\arabic{sortcount}}}%  
  }%  
  {%  
     \glsdosanitizesort  
  }%  
}

(\ifdefstring is defined by the etoolbox package.) For a complete document, see the sample file sampleSort.tex.

____________________________

Example 2 (Customizing Standard Sort (Options 2 or 3))

Suppose you want a glossary of people and you want the names listed as ⟨first-name⟩ ⟨surname⟩ in the glossary, but you want the names sorted by ⟨surname⟩, ⟨first-name⟩. You can do this by defining a command called, say, \name{first-name}{surname} that you can use in the name key when you define the entry, but hook into the standard sort mechanism to temporarily redefine \name while the sort value is being set.

First, define two commands to set the person’s name:

\newcommand{\sortname}[2]{#2, #1}  
\newcommand{\textname}[2]{#1 #2}

and \name needs to be initialised to \textname:

\let\name\textname

Now redefine \glsprestandardsort so that it temporarily sets \name to \sortname and expands the sort value, then sets \name to \textname so that the person’s name appears as ⟨first-name⟩ ⟨surname⟩ in the text:

\renewcommand{\glsprestandardsort}[3]{%  
 \let\name\sortname  
 \edef#1{\expandafter\expandonce\expandafter{#1}}%  
 \let\name\textname  
 \glsdosanitizesort  
}

(The somewhat complicate use of \expandafter etc helps to protect fragile commands, but care is still needed.)

Now the entries can be defined:

\newglossaryentry{joebloggs}{name={\name{Joe}{Bloggs}},  
  description={some information about Joe Bloggs}}  
\newglossaryentry{johnsmith}{name={\name{John}{Smith}},  
  description={some information about John Smith}}

For a complete document, see the sample file samplePeople.tex.

____________________________

order={value}

This may take two values: word or letter. The default is word ordering.

Note that with Options 2 and 3, the order option has no effect if you don’t use makeglossaries.

If you use Option 1, this setting will be used if you use sort=standard in the optional argument of \printnoidxglossary:

\printnoidxglossary[sort=standard]

Alternatively, you can specify the order for individual glossaries:

\printnoidxglossary[sort=word]  
\printnoidxglossary[type=acronym,sort=letter]

With bib2gls, use the break-at option in \GlsXtrLoadResources instead of order.

makeindex

(Option 2) The glossary information and indexing style file will be written in makeindex format. If you use makeglossaries, it will automatically detect that it needs to call makeindex. If you don’t use makeglossaries, you need to remember to use makeindex not xindy. The indexing style file will been given a ist extension.

You may omit this package option if you are using Option 2 as this is the default. It’s available in case you need to override the effect of an earlier occurrence of xindy in the package option list.

xindy

(Option 3) The glossary information and indexing style file will be written in xindy format. If you use makeglossaries, it will automatically detect that it needs to call xindy. If you don’t use makeglossaries, you need to remember to use xindy not makeindex. The indexing style file will been given a xdy extension.

This package option may additionally have a value that is a ⟨key⟩=⟨value⟩ comma-separated list to override the language and codepage. For example:

\usepackage[xindy={language=english,codepage=utf8}]  
  {glossaries}

You can also specify whether you want a number group in the glossary. This defaults to true, but can be suppressed. For example:

\usepackage[xindy={glsnumbers=false}]{glossaries}

If no value is supplied to this package option (either simply writing xindy or writing xindy={}) then the language, codepage and number group settings are unchanged. See §11 Xindy (Option 3) for further details on using xindy with the glossaries package.

xindygloss

(Option 3) This is equivalent to xindy={} (that is, the xindy option without any value supplied) and may be used as a document class option. The language and code page can be set via \GlsSetXdyLanguage and \GlsSetXdyCodePage (see §11.1 Language and Encodings.)

xindynoglsnumbers

(Option 3) This is equivalent to xindy={glsnumbers=false} and may be used as a document class option.

automake={value}

This is option was introduced to version 4.08 as a boolean option. As from version 4.42 it may now take three values: false (default), true or immediate. If no option is supplied, immediate is assumed. The option automake=true will attempt to run makeindex or xindy using TeX’s \write18 mechanism at the end of the document. The option automake=immediate will attempt to run makeindex or xindy at the start of \makeglossaries using \immediate (before the glossary files have been opened).

In the case of automake=true, the associated files are created at the end of the document ready for the next LaTeX run. Since there is a possibility of commands such as \gls occurring on the last page of the document, it’s not possible to use \immediate to close the associated file or with \write18 since the writing of the final indexing lines may have been delayed. In certain situations this can mean that the \write18 fails. In such cases, you will need to use automake=immediate instead.

With automake=immediate, you will get a warning on the first LaTeX run as the associated glossary files don’t exist yet.

Since this mechanism can be a security risk, some TeX distributions disable it completely, in which case this option won’t have an effect. (If this option doesn’t appear to work, search the log file for “runsystem” and see if it is followed by “enabled” or “disabled”.)

Some distributions allow \write18 in a restricted mode. This mode has a limited number of trusted applications, which usually includes makeindex but may not include xindy. So if you have the restricted mode on, automake should work with makeindex but may not work with xindy.

However even in unrestricted mode this option may not work with xindy as xindy uses language names that don’t always correspond with \babel’s language names. (The makeglossaries script applies mappings to assist you.) Note that you still need at least two LaTeX runs to ensure the document is up-to-date with this setting.

Since this package option attempts to run the indexing application on every LaTeX run, its use should be considered a last resort for those who can’t work out how to incorporate the indexing application into their document build. The default value for this option is automake=false.

disablemakegloss

This valueless option indicates that \makeglossaries and \makenoidxglossaries should be disabled. This option is provided in the event that you have to use a class or package that disregards the advice in §1.1 Indexing Options and automatically performs \makeglossaries or \makenoidxglossaries but you don’t want this. (For example, you want to use a different indexing method or you want to disable indexing while working on a draft document.)

This option may be passed in the standard document class option list or passed using \PassOptionsToPackage before glossaries is loaded. Note that this does nothing if \makeglossaries or \makenoidxglossaries has already been used whilst enabled.

restoremakegloss

Cancels the effect of disablemakegloss. This option may be used in \setupglossaries. It issues a warning if \makeglossaries or \makenoidxglossaries has already been used whilst enabled. For example, suppose the class customclass.cls automatically loads glossaries and does \makeglossaries but you need an extra glossary, which has to be defined before \makeglossaries, then you can do:

\documentclass[disablemakegloss]{customclass}  
\newglossary*{functions}{Functions}  
\setupglossaries{restoremakegloss}  
\makeglossaries

or

\PassOptionsToPackage{disablemakegloss}{glossaries}  
\documentclass{customclass}  
\newglossary*{functions}{Functions}  
\setupglossaries{restoremakegloss}  
\makeglossaries

Note that restoring these commands doesn’t necessarily mean that they can be used. It just means that their normal behaviour given the current settings will apply. For example, if you use the record=only or record=nameref options with glossaries-extra then you can’t use \makeglossaries or \makenoidxglossaries regardless of restoremakegloss.

Top

2.6 Glossary Type Options

nohypertypes={list}

Use this option if you have multiple glossaries and you want to suppress the entry hyperlinks for a particular glossary or glossaries. The value of this option should be a comma-separated list of glossary types where \gls etc shouldn’t have hyperlinks by default. Make sure you enclose the value in braces if it contains any commas. Example:

\usepackage[acronym,nohypertypes={acronym,notation}]  
  {glossaries}  
\newglossary[nlg]{notation}{not}{ntn}{Notation}

The values must be fully expanded, so don’t try nohypertypes=\acronymtype. You may also use


\GlsDeclareNoHyperList{list}

instead or additionally. See §6 Links to Glossary Entries for further details.

nomain

This suppresses the creation of the main glossary and associated glo file, if unrequired. Note that if you use this option, you must create another glossary in which to put all your entries (either via the acronym (or acronyms) package option described in §2.7 Acronym and Abbreviation Options or via the symbols, numbers or index options described in §2.8 Other Options or via \newglossary described in §12 Defining New Glossaries).

If you don’t use the main glossary and you don’t use this option, makeglossaries will produce a warning.

Warning: File 'filename.glo' is empty.
Have you used any entries defined in glossary
'main'?
Remember to use package option 'nomain' if
you don't want to use the main glossary.
If you did actually want to use the main glossary and you see this warning, check that you have referenced the entries in that glossary via commands such as \gls.

symbols

This valueless option defines a new glossary type with the label symbols via

\newglossary[slg]{symbols}{sls}{slo}{\glssymbolsgroupname}

It also defines


\printsymbols[options]

which is a synonym for

\printglossary[type=symbols,options]

If you use Option 1, you need to use:

\printnoidxglossary[type=symbols,options]
to display the list of symbols.

Remember to use the nomain package option if you’re only interested in using this symbols glossary and don’t intend to use the main glossary.

The glossaries-extra package has a slightly modified version of this option which additionally provides \glsxtrnewsymbol as a convenient shortcut method for defining symbols. See the glossaries-extra manual for further details.

numbers

This valueless option defines a new glossary type with the label numbers via

\newglossary[nlg]{numbers}{nls}{nlo}{\glsnumbersgroupname}

It also defines


\printnumbers[options]

which is a synonym for

\printglossary[type=numbers,options]

If you use Option 1, you need to use:

\printnoidxglossary[type=numbers,options]
to display the list of numbers.

Remember to use the nomain package option if you’re only interested in using this numbers glossary and don’t intend to use the main glossary.

The glossaries-extra package has a slightly modified version of this option which additionally provides \glsxtrnewnumber as a convenient shortcut method for defining numbers. See the glossaries-extra manual for further details.

index

This valueless option defines a new glossary type with the label index via

\newglossary[ilg]{index}{ind}{idx}{\indexname}%

It also defines


\newterm[options]{term}

which is a synonym for

\newglossaryentry{term}[type=index,name={term},%
description=\nopostdesc,options]
and


\printindex[options]

which is a synonym for

\printglossary[type=index,options]

If you use Option 1, you need to use:

\printnoidxglossary[type=index,options]
to display this glossary.

Remember to use the nomain package option if you’re only interested in using this index glossary and don’t intend to use the main glossary. Note that you can’t mix this option with \index. Either use glossaries for the indexing or use a custom indexing package, such as makeidx, index or imakeidx. (You can, of course, load one of those packages and load glossaries without the index package option.)

Since the index isn’t designed for terms with descriptions, you might also want to disable the hyperlinks for this glossary using the package option nohypertypes=index or the command
\GlsDeclareNoHyperList{index}

The example file sample-index.tex illustrates the use of the index package option.

noglossaryindex

This valueless option switches off index if index has been passed implicitly (for example, through global document options). This option can’t be used in \setupglossaries.

Top

2.7 Acronym and Abbreviation Options

acronym={boolean}

If true, this creates a new glossary with the label acronym. This is equivalent to:

\newglossary[alg]{acronym}{acr}{acn}{\acronymname}

It will also define


\printacronyms[options]

that’s equivalent to

\printglossary[type=acronym,options]
(unless that command is already defined before the beginning of the document or the package option compatible-3.07 is used).

If you are using Option 1, you need to use

\printnoidxglossary[type=acronym,options]
to display the list of acronyms.

If the acronym package option is used, \acronymtype is set to acronym otherwise it is set to main.2.5 Entries that are defined using \newacronym are placed in the glossary whose label is given by \acronymtype, unless another glossary is explicitly specified.

Remember to use the nomain package option if you’re only interested in using this acronym glossary. (That is, you don’t intend to use the main glossary.)

The glossaries-extra extension package comes with an analogous abbreviations option, which creates a new glossary with the label abbreviations and sets the command \glsxtrabbrvtype to this. If the acronym option hasn’t also been used, then \acronymtype will be set to \glsxtrabbrvtype. This enables both \newacronym and \newabbreviation to use the same glossary.

Make sure you have at least v1.42 of glossaries-extra if you use the acronym (or acronyms) package option with the extension package to avoid a bug that interferes with the abbreviation style.

acronyms

This is equivalent to acronym=true and may be used in the document class option list.

abbreviations (glossaries-extra.sty)

This valueless option creates a new glossary type using:

\newglossary[glg-abr]{abbreviations}{gls-abr}{glo-abr}{\abbreviationsname}

The label can be accessed with \glsxtrabbrvtype, which is analogous to \acronymtype. See glossaries-extra manual for further details.

acronymlists={value}

By default, only the \acronymtype glossary is considered to be a list of acronyms. If you have other lists of acronyms, you can specify them as a comma-separated list in the value of acronymlists. For example, if you use the acronym package option but you also want the main glossary to also contain a list of acronyms, you can do:

\usepackage[acronym,acronymlists={main}]{glossaries}

No check is performed to determine if the listed glossaries exist, so you can add glossaries you haven’t defined yet. For example:

\usepackage[acronym,acronymlists={main,acronym2}]  
  {glossaries}  
\newglossary[alg2]{acronym2}{acr2}{acn2}%  
  {Statistical Acronyms}

You can use


\DeclareAcronymList{list}

instead of or in addition to the acronymlists option. This will add the glossaries given in ⟨list⟩ to the list of glossaries that are identified as lists of acronyms. To replace the list of acronym lists with a new list use:


\SetAcronymLists{list}

You can determine if a glossary has been identified as being a list of acronyms using:


\glsIfListOfAcronyms{label}{true part}{false part}

This option is incompatible with glossaries-extra’s abbreviation mechanism.

shortcuts

This option provides shortcut commands for acronyms. See §13 Acronyms and Other Abbreviations for further details. Alternatively you can use:


\DefineAcronymSynonyms

The glossaries-extra package provides additional shortcuts.

Top

2.7.1 Deprecated Acronym Style Options

The package options listed in this section are now deprecated but are kept for backward-compatibility. Use \setacronymstyle instead. See §13 Acronyms and Other Abbreviations for further details.

description

This option changes the definition of \newacronym to allow a description. This option may be replaced by

\setacronymstyle{long-short-desc}

or (with smallcaps)

\setacronymstyle{long-sc-short-desc}

or (with smaller)

\setacronymstyle{long-sm-short-desc}

or (with footnote)

\setacronymstyle{footnote-desc}

or (with footnote and smallcaps)

\setacronymstyle{footnote-sc-desc}

or (with footnote and smaller)

\setacronymstyle{footnote-sm-desc}

or (with dua)

\setacronymstyle{dua-desc}

smallcaps

This option changes the definition of \newacronym and the way that acronyms are displayed. This option may be replaced by:

\setacronymstyle{long-sc-short}

or (with description)

\setacronymstyle{long-sc-short-desc}

or (with description and footnote)

\setacronymstyle{footnote-sc-desc}

smaller

This option changes the definition of \newacronym and the way that acronyms are displayed.

If you use this option, you will need to include the relsize package or otherwise define \textsmaller or redefine \acronymfont.

This option may be replaced by:

\setacronymstyle{long-sm-short}

or (with description)

\setacronymstyle{long-sm-short-desc}

or (with description and footnote)

\setacronymstyle{footnote-sm-desc}

footnote

This option changes the definition of \newacronym and the way that acronyms are displayed. This option may be replaced by:

\setacronymstyle{footnote}

or (with smallcaps)

\setacronymstyle{footnote-sc}

or (with smaller)

\setacronymstyle{footnote-sm}

or (with description)

\setacronymstyle{footnote-desc}

or (with smallcaps and description)

\setacronymstyle{footnote-sc-desc}

or (with smaller and description)

\setacronymstyle{footnote-sm-desc}

dua

This option changes the definition of \newacronym so that acronyms are always expanded. This option may be replaced by:

\setacronymstyle{dua}

or (with description)

\setacronymstyle{dua-desc}

Top

2.8 Other Options

Other available options that don’t fit any of the above categories are described below.

accsupp (glossaries-extra.sty)

Load the glossaries-accsupp package.

prefix (glossaries-extra.sty)

Load the glossaries-prefix package.

nomissingglstext={boolean} (glossaries-extra.sty)

This option may be used to suppress the boilerplate text generated by \printglossary if the glossary file is missing.

compatible-2.07={boolean}

Compatibility mode for old documents created using version 2.07 or below.

compatible-3.07={boolean}

Compatibility mode for old documents created using version 3.07 or below.

kernelglossredefs={value}

As a legacy from the precursor glossary package, the standard glossary commands provided by the LaTeX kernel (\makeglossary and \glossary) are redefined in terms of the glossaries package’s commands. However, they were never documented in this user manual, and the conversion guide (“Upgrading from the glossary package to the glossaries package”) explicitly discourages their use.

The use of those kernel commands (instead of the appropriate commands documented in this user guide) are deprecated, and you will now get a warning if you try using them.

In the event that you require the original form of these kernel commands, for example, if you need to use the glossaries package with another class or package that also performs glossary-style indexing, then you can restore these commands to their previous definition (that is, their definitions prior to loading the glossaries package) with the package option kernelglossredefs=false. You may also need to use the nomain option in the event of file extension conflicts. (In which case, you must provide a new default glossary for use with the glossaries package.)

This option may take one of three values: true (redefine with warnings, default), false (restore previous definitions) or nowarn (redefine without warnings, not recommended).

The only glossary-related commands provided by the LaTeX kernel are \makeglossary and \glossary. Other packages or classes may provide additional glossary-related commands or environments that conflict with glossaries (such as \printglossary and theglossary). These non-kernel commands aren’t affected by this package option, and you will have to find some way to resolve the conflict if you require both glossary mechanisms. (The glossaries package will override the existing definitions of \printglossary and theglossary.)

In general, if possible, it’s best to stick with just one package that provides a glossary mechanism. (The glossaries package does check for the doc package and patches \PrintChanges.)

Top

2.9 Setting Options After the Package is Loaded

Some of the options described above may also be set after the glossaries package has been loaded using


\setupglossaries{key-val list}

The following package options can’t be used in \setupglossaries: xindy, xindygloss, xindynoglsnumbers, makeindex, nolong, nosuper, nolist, notree, nostyles, nomain, compatible-2.07, translate, notranslate, acronym. These options have to be set while the package is loading, except for the xindy sub-options which can be set using commands like \GlsSetXdyLanguage (see §11 Xindy (Option 3) for further details).

If you need to use this command, use it as soon as possible after loading glossaries otherwise you might end up using it too late for the change to take effect. For example, if you try changing the acronym styles (such as smallcaps) after you have started defining your acronyms, you are likely to get unexpected results. If you try changing the sort option after you have started to define entries, you may get unexpected results.

Top

3. Setting Up

In the preamble you need to indicate whether you want to use Option 1, Option 2 or Option 3. It’s not possible to mix these options within a document, although some combinations are possible with glossaries-extra. (For Options 4 and 5 see the bib2gls and glossaries-extra manuals.)

Top

3.1 Option 1

The command


\makenoidxglossaries

must be placed in the preamble. This sets up the internal commands required to make Option 1 work. If you omit \makenoidxglossaries none of the glossaries will be displayed.

Top

3.2 Options 2 and 3

The command


\makeglossaries

must be placed in the preamble in order to create the customised makeindex (ist) or xindy (xdy) style file (for Options 2 or 3, respectively) and to ensure that glossary entries are written to the appropriate output files. If you omit \makeglossaries none of the glossary files will be created.

If you are using glossaries-extra, \makeglossaries has an optional argument that allows you to have a hybrid of Options 1 or 2 or Options 1 or 3. See glossaries-extra manual for further details.

Note that some of the commands provided by the glossaries package must not be used after \makeglossaries as they are required when creating the customised style file. If you attempt to use those commands after \makeglossaries you will generate an error. Similarly, there are some commands that must not be used before \makeglossaries.

You can suppress the creation of the customised xindy or makeindex style file using


\noist

That this command must not be used after \makeglossaries.

Note that if you have a custom xdy file created when using glossaries version 2.07 or below, you will need to use the compatible-2.07 package option with it.

The default name for the customised style file is given by \jobnameist (Option 2) or \jobnamexdy (Option 3). This name may be changed using:


\setStyleFile{name}

where ⟨name⟩ is the name of the style file without the extension. Note that this command must not be used after \makeglossaries.

Each glossary entry is assigned a number list that lists all the locations in the document where that entry was used. By default, the location refers to the page number but this may be overridden using the counter package option. The default form of the location number assumes a full stop compositor (e.g. 1.2), but if your location numbers use a different compositor (e.g. 1-2) you need to set this using


\glsSetCompositor{symbol}

For example:

\glsSetCompositor{-}

This command must not be used after \makeglossaries.

If you use Option 3, you can have a different compositor for page numbers starting with an upper case alphabetical character using:


\glsSetAlphaCompositor{symbol}

This command has no effect if you use Option 2. For example, if you want number lists containing a mixture of A-1 and 2.3 style formats, then do:

\glsSetCompositor{.}\glsSetAlphaCompositor{-}

See §5 Number lists for further information about number lists.

Top

4. Defining Glossary Entries

All glossary entries must be defined before they are used, so it is better to define them in the preamble to ensure this. In fact, some commands such as \longnewglossaryentry may only be used in the preamble. See §4.8 Drawbacks With Defining Entries in the Document Environment for a discussion of the problems with defining entries within the document instead of in the preamble. (The glossaries-extra package has an option that provides a restricted form of document definitions that avoids some of the issues discussed in §4.8 Drawbacks With Defining Entries in the Document Environment.)

Option 1 enforces the preamble-only restriction on \newglossaryentry. Option 4 requires that definitions are provided in bib format. Option 5 requires either preamble-only definitions or the use of the glossaries-extra package option docdef=restricted.

Only those entries that are indexed in the document (using any of the commands described in §6 Links to Glossary Entries, §7 Adding an Entry to the Glossary Without Generating Text or §8 Cross-Referencing Entries) will appear in the glossary. See §10 Displaying a glossary to find out how to display the glossary.

New glossary entries are defined using the command:


\newglossaryentry{label}{key=value list}

This is a short command, so values in ⟨key-val list⟩ can’t contain any paragraph breaks. Take care to enclose values containing any commas (,) or equal signs (=) with braces to hide them from the key=value list parser.

If you have a long description that needs to span multiple paragraphs, use


\longnewglossaryentry{label}{key=value list}{long description}

instead. Note that this command may only be used in the preamble. Be careful of unwanted spaces. \longnewglossaryentry will remove trailing spaces in the description (via \unskip) but won’t remove leading spaces. This command also appends \nopostdesc to the end of the description, which suppresses the post-description hook. The glossaries-extra package provides a starred version of \longnewglossaryentry that doesn’t append either \unskip or \nopostdesc.

There are also commands that will only define the entry if it hasn’t already been defined:


\provideglossaryentry{label}{key=value list}

and


\longprovideglossaryentry{label}{key=value list}{long description}

(These are both preamble-only commands.)

For all the above commands, the first argument, ⟨label⟩, must be a unique label with which to identify this entry. This can’t contain any non-expandable commands or active characters. The reason for this restriction is that the label is used to construct internal commands that store the associated information (similarly to commands like \label) and therefore must be able to expand to a valid control sequence name.

Note that although an extended Latin character or other non-Latin character, such as é or ß, looks like a plain character in your .tex file, it’s actually a macro (an active character) and therefore can’t be used in the label. (This applies to LaTeX rather than XeLaTeX.) Also be careful of babel’s options that change certain punctuation characters (such as : or -) to active characters.

The second argument, ⟨key=value list⟩, is a ⟨key⟩=⟨value⟩ list that supplies the relevant information about this entry. There are two required fields: description and either name or parent. The description is set in the third argument of \longnewglossaryentry and \longprovideglossaryentry. With the other commands it’s set via the description key. As is typical with ⟨key⟩=⟨value⟩ lists, values that contain a comma or equal sign must be enclosed in braces. Available fields are listed below. Additional fields are provided by the supplementary packages glossaries-prefix (§17 Prefixes or Determiners) and glossaries-accsupp (§18 Accessibility Support) and also by glossaries-extra. You can also define your own custom keys (see §4.3 Additional Keys).

name
The name of the entry (as it will appear in the glossary). If this key is omitted and the parent key is supplied, this value will be the same as the parent’s name.

If the name key contains any commands, you must also use the sort key (described below) if you intend sorting the entries alphabetically, otherwise the entries can’t be sorted correctly.

description
A brief description of this term (to appear in the glossary). Within this value, you can use:


\nopostdesc

to suppress the description terminator for this entry. For example, if this entry is a parent entry that doesn’t require a description, you can do description={\nopostdesc}. If you want a paragraph break in the description use:


\glspar

or, better, use \longnewglossaryentry. However, note that not all glossary styles support multi-line descriptions. If you are using one of the tabular-like glossary styles that permit multi-line descriptions, use \newline not \\ if you want to force a line break.

With glossaries-extra, use \glsxtrnopostpunc instead of \nopostdesc to suppress the post-description punctuation.

parent
The label of the parent entry. Note that the parent entry must be defined before its sub-entries. See §4.5 Sub-Entries for further details.
descriptionplural
The plural form of the description, if required. If omitted, the value is set to the same as the description key.
text
How this entry will appear in the document text when using \gls (or one of its upper case variants). If this field is omitted, the value of the name key is used.
first
How the entry will appear in the document text on first use with \gls (or one of its upper case variants). If this field is omitted, the value of the text key is used. Note that if you use \glspl, \Glspl, \GLSpl, \glsdisp before using \gls, the firstplural value won’t be used with \gls.
plural
How the entry will appear in the document text when using \glspl (or one of its upper case variants). If this field is omitted, the value is obtained by appending \glspluralsuffix to the value of the text field. The default value of \glspluralsuffix is the letter “s”.
firstplural
How the entry will appear in the document text on first use with \glspl (or one of its upper case variants). If this field is omitted, the value is obtained from the plural key, if the first key is omitted, or by appending \glspluralsuffix to the value of the first field, if the first field is present. Note that if you use \gls, \Gls, \GLS, \glsdisp before using \glspl, the firstplural value won’t be used with \glspl.

Note: prior to version 1.13, the default value of firstplural was always taken by appending “s” to the first key, which meant that you had to specify both plural and firstplural, even if you hadn’t used the first key.

symbol
This field is provided to allow the user to specify an associated symbol. If omitted, the value is set to \relax. Note that not all glossary styles display the symbol.
symbolplural
This is the plural form of the symbol (as passed to \glsdisplay and \glsdisplayfirst by \glspl, \Glspl and \GLSpl). If omitted, the value is set to the same as the symbol key.
sort
This value indicates how this entry should be sorted. If omitted, the value is given by the name field unless one of the package options sort=def and sort=use have been used. In general, it’s best to use the sort key if the name contains commands (e.g. \ensuremath{\alpha}). You can also override the sort key by redefining \glsprestandardsort (see §2.5 Sorting Options).

The sort key shouldn’t be used with bib2gls. It has a system of fallbacks that allow different types of entries to obtain the sort value from the most relevant field. See the bib2gls manual for further details and see also bib2gls gallery: sorting.

Option 1 by default strips the standard LaTeX accents (that is, accents generated by core LaTeX commands) from the name key when it sets the sort key. So with Option 1:

\newglossaryentry{elite}{%  
  name={{\'e}lite},  
  description={select group of people}  
}

This is equivalent to:

\newglossaryentry{elite}{%  
  name={{\'e}lite},  
  description={select group of people},  
  sort={elite}  
}

Unless you use the package option sanitizesort=true, in which case it’s equivalent to:

\newglossaryentry{elite}{%  
  name={{\'e}lite},  
  description={select group of people},  
  sort={\'elite}  
}

This will place the entry before the “A” letter group since the sort value starts with a symbol.

Similarly if you use the inputenc package:

\newglossaryentry{elite}{%
  name={{é}lite},
  description={select group of people}
}
This is equivalent to
\newglossaryentry{elite}{%
  name=i{{é}lite},
  description={select group of people},
  sort=elite
}
Unless you use the package option sanitizesort=true, in which case it’s equivalent to:
\newglossaryentry{elite}{%
  name={{é}lite},
  description={select group of people},
  sort=élite
}
Again, this will place the entry before the “A” group.

With Options 2 and 3, the default value of sort will either be set to the name key (if sanitizesort=true) or it will set it to the expansion of the name key (if sanitizesort=false).

Take care with xindy (Option 3): if you have entries with the same sort value they will be treated as the same entry. If you use xindy and aren’t using the def or use sort methods, always use the sort key for entries where the name just consists of a control sequence (for example name={\alpha}).

Take care if you use Option 1 and the name contains fragile commands. You will either need to explicitly set the sort key or use the sanitizesort=true package option (unless you use the def or use sort methods).

type
This specifies the label of the glossary in which this entry belongs. If omitted, the default glossary is assumed unless \newacronym is used (see §13 Acronyms and Other Abbreviations).
user1, …, user6
Six keys provided for any additional information the user may want to specify. (For example, an associated dimension or an alternative plural or some other grammatical construct.) Alternatively, you can add new keys using \glsaddkey or \glsaddstoragekey (see §4.3 Additional Keys).
nonumberlist
A boolean key. If the value is missing or is true, this will suppress the number list just for this entry. Conversely, if you have used the package option nonumberlist, you can activate the number list just for this entry with nonumberlist=false. (See §5 Number lists.)
see
Cross-reference another entry. Using the see key will automatically add this entry to the glossary, but will not automatically add the cross-referenced entry. The referenced entry should be supplied as the value to this key. If you want to override the “see” tag, you can supply the new tag in square brackets before the label. For example see=[see also]{anotherlabel}. Note that if you have suppressed the number list, the cross-referencing information won’t appear in the glossary, as it forms part of the number list. You can override this for individual glossary entries using nonumberlist=false (see above). Alternatively, you can use the seeautonumberlist package option. For further details, see §8 Cross-Referencing Entries.

This key essentially provides a convenient shortcut that performs

\glssee[tag]{label}{xr-label list}
after the entry has been defined.

For Options 2 and 3, \makeglossaries must be used before any occurrence of \newglossaryentry that contains the see key. This key should not be used with entries defined in the document environment.

Since it’s useful to suppress the indexing while working on a draft document, consider using the seenoindex package option to warn or ignore the see key while \makeglossaries is commented out.

If you use the see key, you may want to consider using the glossaries-extra package which additionally provides a seealso and alias key. If you want to avoid the automatic indexing triggered by the see key, consider using Option 4.

seealso
This key is only available with glossaries-extra and is similar to see but it doesn’t allow for the optional tag. The glossaries-extra package provides \seealsoname and seealso={list} is essentially like see=[\seealsoname]list⟩ (Options 3 and 4 may treat these differently).
alias
This key is only available with glossaries-extra and is another form of cross-referencing. An entry can be aliased to another entry with alias={label}. This behaves like see={label} but also alters the behaviour of commands like \gls so that they index the entry given by ⟨label⟩ instead of the original entry. More variations with the key are available with bib2gls.
category
This key is only available with glossaries-extra and is used to assign a category to the entry. The value should be a label that can be used to identify the category. See glossaries-extra manual for further details.

The following keys are reserved for \newacronym (see §13 Acronyms and Other Abbreviations) and also for \newabbreviation (see the glossaries-extra manual): long, longplural, short and shortplural.

There are also special internal field names used by bib2gls. See the bib2gls manual for further details.

The supplementary packages glossaries-prefix (§17 Prefixes or Determiners) and glossaries-accsupp (§18 Accessibility Support) provide additional keys.

Avoid using any of the \gls-like or \glstext-like commands within the text, first, short or long keys (or their plural equivalent) or any other key that you plan to access through those commands. (For example, the symbol key if you intend to use \glssymbol.) Otherwise you end up with nested links, which can cause complications and they won’t work with the case-changing commands. You can use them within the value of keys that won’t be accessed through those commands. For example, the description key if you don’t use \glsdesc. Additionally, they’ll confuse the entry formatting commands, such as \glslabel.

Note that if the name starts with non-Latin character, you must group the character, otherwise it will cause a problem for commands like \Gls and \Glspl. For example:

\newglossaryentry{elite}{name={{\'e}lite},  
description={select group or class}}

Note that the same applies if you are using the inputenc package:

\newglossaryentry{elite}{name={{é}lite},
description={select group or class}}
(This doesn’t apply for XeLaTeX documents using the fontspec package. For further details, see the “UTF-8” section in the mfirstuc user manual.)

Note that in both of the above examples, you will also need to supply the sort key if you are using Option 2 whereas xindy (Option 3) is usually able to sort non-Latin characters correctly. Option 1 discards accents from standard LaTeX extended Latin characters unless you use the sanitizesort=true.

Top

4.1 Plurals

You may have noticed from above that you can specify the plural form when you define a term. If you omit this, the plural will be obtained by appending


\glspluralsuffix

to the singular form. This command defaults to the letter “s”. For example:

\newglossaryentry{cow}{name=cow,description={a fully grown  
female of any bovine animal}}

defines a new entry whose singular form is “cow” and plural form is “cows”. However, if you are writing in archaic English, you may want to use “kine” as the plural form, in which case you would have to do:

\newglossaryentry{cow}{name=cow,plural=kine,  
description={a fully grown female of any bovine animal}}

If you are writing in a language that supports multiple plurals (for a given term) then use the plural key for one of them and one of the user keys to specify the other plural form. For example:

\newglossaryentry{cow}{%  
  name=cow,%  
  description={a fully grown female of any bovine animal  
               (plural cows, archaic plural kine)},%  
  user1={kine}}

You can then use \glspl{cow} to produce “cows” and \glsuseri{cow} to produce “kine”. You can, of course, define an easy to remember synonym. For example:

\let\glsaltpl\glsuseri

Then you don’t have to remember which key you used to store the second plural. Alternatively, you can define your own keys using \glsaddkey, described in §4.3 Additional Keys.

If you are using a language that usually forms plurals by appending a different letter, or sequence of letters, you can redefine \glspluralsuffix as required. However, this must be done before the entries are defined. For languages that don’t form plurals by simply appending a suffix, all the plural forms must be specified using the plural key (and the firstplural key where necessary).

Top

4.2 Other Grammatical Constructs

You can use the six user keys to provide alternatives, such as participles. For example:

\let\glsing\glsuseri  
\let\glsd\glsuserii  
\newcommand*{\ingkey}{user1}  
\newcommand*{\edkey}{user2}  
\newcommand*{\newword}[3][]{%  
  \newglossaryentry{#2}{%  
   name={#2},%  
   description={#3},%  
   \edkey={#2ed},%  
   \ingkey={#2ing},#1%  
  }%  
}

With the above definitions, I can now define terms like this:

\newword{play}{to take part in activities for enjoyment}  
\newword[\edkey={ran},\ingkey={running}]{run}{to move fast using  
the legs}

and use them in the text:

Peter is \glsing{play} in the park today.  
Jane \glsd{play} in the park yesterday.  
Peter and Jane \glsd{run} in the park last week.

Alternatively, you can define your own keys using \glsaddkey, described below in §4.3 Additional Keys.

Top

4.3 Additional Keys

You can now also define your own custom keys using the commands described in this section. There are two types of keys: those for use within the document and those to store information used behind the scenes by other commands.

For example, if you want to add a key that indicates the associated unit for a term, you might want to reference this unit in your document. In this case use \glsaddkey described in §4.3.1 Document Keys. If, on the other hand, you want to add a key to indicate to a glossary style or acronym style that this entry should be formatted differently to other entries, then you can use \glsaddstoragekey described in §4.3.2 Storage Keys.

In both cases, a new command ⟨no link cs⟩ will be defined that can be used to access the value of this key (analogous to commands such as \glsentrytext). This can be used in an expandable context (provided any fragile commands stored in the key have been protected). The new keys must be added using \glsaddkey or \glsaddstoragekey before glossary entries are defined.

Top

4.3.1 Document Keys

A custom key that can be used in the document is defined using:


\glsaddkey{key}{default value}{no link cs}{no link ucfirst cs}{link cs}{link ucfirst cs}{link allcaps cs}

where:

key
is the new key to use in \newglossaryentry (or similar commands such as \longnewglossaryentry);
default value
is the default value to use if this key isn’t used in an entry definition (this may reference the current entry label via \glslabel, but you will have to switch on expansion via the starred version of \glsaddkey and protect fragile commands);
no link cs
is the control sequence to use analogous to commands like \glsentrytext;
no link ucfirst cs
is the control sequence to use analogous to commands like \Glsentrytext;
link cs
is the control sequence to use analogous to commands like \glstext;
link ucfirst cs
is the control sequence to use analogous to commands like \Glstext;
link allcaps cs
is the control sequence to use analogous to commands like \GLStext.

The starred version of \glsaddkey switches on expansion for this key. The unstarred version doesn’t override the current expansion setting.

Example 3 (Defining Custom Keys)

Suppose I want to define two new keys, ed and ing, that default to the entry text followed by “ed” and “ing”, respectively. The default value will need expanding in both cases, so I need to use the starred form:

 % Define "ed" key:  
 \glsaddkey*  
  {ed}% key  
  {\glsentrytext{\glslabel}ed}% default value  
  {\glsentryed}% command analogous to \glsentrytext  
  {\Glsentryed}% command analogous to \Glsentrytext  
  {\glsed}% command analogous to \glstext  
  {\Glsed}% command analogous to \Glstext  
  {\GLSed}% command analogous to \GLStext  
 % Define "ing" key:  
 \glsaddkey*  
  {ing}% key  
  {\glsentrytext{\glslabel}ing}% default value  
  {\glsentrying}% command analogous to \glsentrytext  
  {\Glsentrying}% command analogous to \Glsentrytext  
  {\glsing}% command analogous to \glstext  
  {\Glsing}% command analogous to \Glstext  
  {\GLSing}% command analogous to \GLStext

Now I can define some entries:

 % No need to override defaults for this entry:  
 \newglossaryentry{jump}{name={jump},description={}}  
 % Need to override defaults on these entries:  
 \newglossaryentry{run}{name={run},%  
   ed={ran},%  
   ing={running},%  
   description={}}  
 \newglossaryentry{waddle}{name={waddle},%  
   ed={waddled},%  
   ing={waddling},%  
   description={}}

These entries can later be used in the document:

The dog \glsed{jump} over the duck.  
The duck was \glsing{waddle} round the dog.  
The dog \glsed{run} away from the duck.

For a complete document, see the sample file sample-newkeys.tex.

____________________________

Top

4.3.2 Storage Keys

A custom key that can be used for simply storing information is defined using:


\glsaddstoragekey{key}{default value}{no link cs}

where the arguments are as the first three arguments of \glsaddkey, described above in §4.3.1 Document Keys.

This is essentially the same as \glsaddkey except that it doesn’t define the additional commands. You can access or update the value of your new field using the commands described in §16.3 Fetching and Updating the Value of a Field.

Example 4 (Defining Custom Storage Key (Acronyms and Initialisms))

Suppose I want to define acronyms and other forms of abbreviations, such as initialisms, but I want them all in the same glossary and I want the acronyms on first use to be displayed with the short form followed by the long form in parentheses, but the opposite way round for other forms of abbreviations. (The glossaries-extra package provides a simpler way of achieving this.)

Here I can define a new key that determines whether the term is actually an acronym rather than some other form of abbreviation. I’m going to call this key abbrtype (since type already exists):

\glsaddstoragekey  
 {abbrtype}% key/field name  
 {word}% default value if not explicitly set  
 {\abbrtype}% custom command to access the value if required

Now I can define a style that looks up the value of this new key to determine how to display the full form:

\newacronymstyle  
 {mystyle}% style name  
 {% Use the generic display  
   \ifglshaslong{\glslabel}{\glsgenacfmt}{\glsgenentryfmt}%  
 }  
 {% Put the long form in the description  
   \renewcommand*{\GenericAcronymFields}{%  
       description={\the\glslongtok}}%  
   % For the full format, test the value of the "abbrtype" key.  
   % If it's set to "word" put the short form first with  
   % the long form in brackets.  
   \renewcommand*{\genacrfullformat}[2]{%  
    \ifglsfieldeq{##1}{abbrtype}{word}  
    {% is a proper acronym  
      \protect\firstacronymfont{\glsentryshort{##1}}##2\space  
       (\glsentrylong{##1})%  
    }  
    {% is another form of abbreviation  
     \glsentrylong{##1}##2\space  
      (\protect\firstacronymfont{\glsentryshort{##1}})%  
    }%  
  }%  
  % first letter upper case version:  
   \renewcommand*{\Genacrfullformat}[2]{%  
    \ifglsfieldeq{##1}{abbrtype}{word}  
    {% is a proper acronym  
      \protect\firstacronymfont{\Glsentryshort{##1}}##2\space  
       (\glsentrylong{##1})%  
    }  
    {% is another form of abbreviation  
     \Glsentrylong{##1}##2\space  
      (\protect\firstacronymfont{\glsentryshort{##1}})%  
    }%  
  }%  
  % plural  
   \renewcommand*{\genplacrfullformat}[2]{%  
    \ifglsfieldeq{##1}{abbrtype}{word}  
    {% is a proper acronym  
      \protect\firstacronymfont{\glsentryshortpl{##1}}##2\space  
       (\glsentrylong{##1})%  
    }  
    {% is another form of abbreviation  
     \glsentrylongpl{##1}##2\space  
      (\protect\firstacronymfont{\glsentryshortpl{##1}})%  
    }%  
  }%  
  % plural and first letter upper case  
  \renewcommand*{\Genplacrfullformat}[2]{%  
    \ifglsfieldeq{##1}{abbrtype}{word}  
    {% is a proper acronym  
      \protect\firstacronymfont{\Glsentryshortpl{##1}}##2\space  
       (\glsentrylong{##1})%  
    }  
    {% is another form of abbreviation  
     \Glsentrylongpl{##1}##2\space  
      (\protect\firstacronymfont{\glsentryshortpl{##1}})%  
    }%  
  }%  
  % Just use the short form as the name part in the glossary:  
  \renewcommand*{\acronymentry}[1]{%  
     \acronymfont{\glsentryshort{##1}}}%  
  % Sort by the short form:  
  \renewcommand*{\acronymsort}[2]{##1}%  
  % Just use the surrounding font for the short form:  
  \renewcommand*{\acronymfont}[1]{##1}%  
  % Same for first use:  
  \renewcommand*{\firstacronymfont}[1]{\acronymfont{##1}}%  
  % Default plural suffix if the plural isn't explicitly set  
  \renewcommand*{\acrpluralsuffix}{\glspluralsuffix}%  
 }

Remember that the new style needs to be set before defining any terms:

\setacronymstyle{mystyle}

Since it’s a bit confusing to use \newacronym for something that’s not technically an acronym, let’s define a new command for initialisms:

\newcommand*{\newinitialism}[4][]{%  
  \newacronym[abbrtype=initialism,#1]{#2}{#3}{#4}%  
}

Now the entries can all be defined:

\newacronym{radar}{radar}{radio detecting and ranging}  
\newacronym{laser}{laser}{light amplification by stimulated  
emission of radiation}  
\newacronym{scuba}{scuba}{self-contained underwater breathing  
apparatus}  
\newinitialism{dsp}{DSP}{digital signal processing}  
\newinitialism{atm}{ATM}{automated teller machine}

On first use, \gls{radar} will produce “radar (radio detecting and ranging)” but \gls{dsp} will produce “DSP (digital signal processing)”.

For a complete document, see the sample file sample-storage-abbr.tex.

____________________________

In the above example, if \newglossaryentry is explicitly used (instead of through \newacronym) the abbrtype key will be set to its default value of “word” but the \ifglshaslong test in the custom acronym style will be false (since the long key hasn’t been set) so the display style will switch to that given by \glsgenentryfmt and they’ll be no test performed on the abbrtype field.

Example 5 (Defining Custom Storage Key (Acronyms and Non-Acronyms with Descriptions))

The previous example can be modified if the description also needs to be provided. Here I’ve changed “word” to “acronym”:

\glsaddstoragekey  
 {abbrtype}% key/field name  
 {acronym}% default value if not explicitly set  
 {\abbrtype}% custom command to access the value if required

This may seem a little odd for non-abbreviated entries defined using \newglossaryentry directly, but \ifglshaslong can be used to determine whether or not to reference the value of this new abbrtype field.

The new acronym style has a minor modification that forces the user to specify a description. In the previous example, the line:

   \renewcommand*{\GenericAcronymFields}{%  
     description={\the\glslongtok}}%

needs to be changed to:

   \renewcommand*{\GenericAcronymFields}{}%

Additionally, to accommodate the change in the default value of the abbrtype key, all instances of

    \ifglsfieldeq{##1}{abbrtype}{word}

need to be changed to:

    \ifglsfieldeq{##1}{abbrtype}{acronym}

Once this new style has been set, the new acronyms can be defined using the optional argument to set the description:

\newacronym[description={system for detecting the position and  
speed of aircraft, ships, etc}]{radar}{radar}{radio detecting  
and ranging}

No change is required for the definition of \newinitialism but again the optional argument is required to set the description:

\newinitialism[description={mathematical manipulation of an  
information signal}]{dsp}{DSP}{digital signal processing}

We can also accommodate contractions in a similar manner to the initialisms:

\newcommand*{\newcontraction}[4][]{%  
  \newacronym[abbrtype=contraction,#1]{#2}{#3}{#4}%  
}

The contractions can similarly been defined using this new command:

\newcontraction[description={front part of a ship below the  
deck}]{focsle}{fo'c's'le}{forecastle}

Since the custom acronym style just checks if abbrtype is acronym, the contractions will be treated the same as the initialisms, but the style could be modified by a further test of the abbrtype value if required.

To test regular non-abbreviated entries, I’ve also defined a simple word:

\newglossaryentry{apple}{name={apple},description={a fruit}}

Now for a new glossary style that provides information about the abbreviation (in addition to the description):

\newglossarystyle  
 {mystyle}% style name  
 {% base it on the "list" style  
   \setglossarystyle{list}%  
   \renewcommand*{\glossentry}[2]{%  
     \item[\glsentryitem{##1}%  
          \glstarget{##1}{\glossentryname{##1}}]  
       \ifglshaslong{##1}%  
       { (\abbrtype{##1}: \glsentrylong{##1})\space}{}%  
       \glossentrydesc{##1}\glspostdescription\space ##2}%  
 }

This uses \ifglshaslong to determine whether or not the term is an abbreviation. If it has an abbreviation, the full form is supplied in parentheses and \abbrtype (defined by \glsaddstoragekey earlier) is used to indicate the type of abbreviation.

With this style set, the apple entry is simply displayed in the glossary as

apple
a fruit.

but the abbreviations are displayed in the form

laser
(acronym: light amplification by stimulated emission of radiation) device that creates a narrow beam of intense light.

(for acronyms) or

DSP
(initialism: digital signal processing) mathematical manipulation of an information signal.

(for initalisms) or

fo’c’s’le
(contraction: forecastle) front part of a ship below the deck.

(for contractions).

For a complete document, see sample-storage-abbr-desc.tex.

____________________________

Top

4.4 Expansion

When you define new glossary entries expansion is performed by default, except for the name, description, descriptionplural, symbol, symbolplural and sort keys (these keys all have expansion suppressed via \glssetnoexpandfield).

You can switch expansion on or off for individual keys using


\glssetexpandfield{field}

or


\glssetnoexpandfield{field}

respectively, where ⟨field⟩ is the field tag corresponding to the key. In most cases, this is the same as the name of the key except for those listed in table 4.1.


Table 4.1: Key to Field Mappings
Key Field
sort sortvalue
firstplural firstpl
description desc
descriptionplural descplural
user1 useri
user2 userii
user3 useriii
user4 useriv
user5 userv
user6 uservi
longplural longpl
shortplural shortpl

Any keys that haven’t had the expansion explicitly set using \glssetexpandfield or \glssetnoexpandfield are governed by


\glsexpandfields

and


\glsnoexpandfields

If your entries contain any fragile commands, I recommend you switch off expansion via \glsnoexpandfields. (This should be used before you define the entries.)

Top

4.5 Sub-Entries

As from version 1.17, it is possible to specify sub-entries. These may be used to order the glossary into categories, in which case the sub-entry will have a different name to its parent entry, or it may be used to distinguish different definitions for the same word, in which case the sub-entries will have the same name as the parent entry. Note that not all glossary styles support hierarchical entries and may display all the entries in a flat format. Of the styles that support sub-entries, some display the sub-entry’s name whilst others don’t. Therefore you need to ensure that you use a suitable style. (See §15 Glossary Styles for a list of predefined styles.) As from version 3.0, level 1 sub-entries are automatically numbered in the predefined styles if you use the subentrycounter package option (see §2.3 Glossary Appearance Options for further details).

Note that the parent entry will automatically be added to the glossary if any of its child entries are used in the document. If the parent entry is not referenced in the document, it will not have a number list. Note also that makeindex has a restriction on the maximum sub-entry depth.

Top

4.5.1 Hierarchical Categories

To arrange a glossary with hierarchical categories, you need to first define the category and then define the sub-entries using the relevant category entry as the value of the parent key.

Example 6 (Hierarchical Categories—Greek and Roman Mathematical Symbols)

Suppose I want a glossary of mathematical symbols that are divided into Greek letters and Roman letters. Then I can define the categories as follows:

\newglossaryentry{greekletter}{name={Greek letters},  
description={\nopostdesc}}  
\newglossaryentry{romanletter}{name={Roman letters},  
description={\nopostdesc}}

Note that in this example, the category entries don’t need a description so I have set the descriptions to \nopostdesc. This gives a blank description and suppresses the description terminator.

I can now define my sub-entries as follows:

\newglossaryentry{pi}{name={\ensuremath{\pi}},sort={pi},  
description={ratio of the circumference of a circle to  
the diameter},  
parent=greekletter}  
\newglossaryentry{C}{name={\ensuremath{C}},sort={C},  
description={Euler's constant},  
parent=romanletter}

For a complete document, see the sample file sampletree.tex.

____________________________

Top

4.5.2 Homographs

Sub-entries that have the same name as the parent entry, don’t need to have the name key. For example, the word “glossary” can mean a list of technical words or a collection of glosses. In both cases the plural is “glossaries”. So first define the parent entry:

\newglossaryentry{glossary}{name=glossary,  
description={\nopostdesc},  
plural={glossaries}}

Again, the parent entry has no description, so the description terminator needs to be suppressed using \nopostdesc.

Now define the two different meanings of the word:

\newglossaryentry{glossarylist}{  
description={list of technical words},  
sort={1},  
parent={glossary}}  
\newglossaryentry{glossarycol}{  
description={collection of glosses},  
sort={2},  
parent={glossary}}

Note that if I reference the parent entry, the location will be added to the parent’s number list, whereas if I reference any of the child entries, the location will be added to the child entry’s number list. Note also that since the sub-entries have the same name, the sort key is required unless you are using the sort=use or sort=def package options (see §2.5 Sorting Options). You can use the subentrycounter package option to automatically number the first-level child entries. See §2.3 Glossary Appearance Options for further details.

In the above example, the plural form for both of the child entries is the same as the parent entry, so the plural key was not required for the child entries. However, if the sub-entries have different plurals, they will need to be specified. For example:

\newglossaryentry{bravo}{name={bravo},  
description={\nopostdesc}}  
\newglossaryentry{bravocry}{description={cry of approval  
(pl.\ bravos)},  
sort={1},  
plural={bravos},  
parent=bravo}  
\newglossaryentry{bravoruffian}{description={hired  
ruffian or killer (pl.\ bravoes)},  
sort={2},  
plural={bravoes},  
parent=bravo}

Top

4.6 Loading Entries From a File

You can store all your glossary entry definitions in another file and use:


\loadglsentries[type]{filename}

where ⟨filename⟩ is the name of the file containing all the \newglossaryentry or \longnewglossaryentry commands. The optional argument ⟨type⟩ is the name of the glossary to which those entries should belong, for those entries where the type key has been omitted (or, more specifically, for those entries whose type has been specified by \glsdefaulttype, which is what \newglossaryentry uses by default).

This is a preamble-only command. You may also use \input to load the file but don’t use \include. If you find that your file is becoming unmanageably large, you may want to consider switching to bib2gls and use an application such as JabRef to manage the entry definitions.

If you want to use \AtBeginDocument to \input all your entries automatically at the start of the document, add the \AtBeginDocument command before you load the glossaries package (and babel, if you are also loading that) to avoid the creation of the glsdefs file and any associated problems that are caused by defining commands in the document environment. (See §4.8 Drawbacks With Defining Entries in the Document Environment.)

Example 7 (Loading Entries from Another File)

Suppose I have a file called myentries.tex which contains:

\newglossaryentry{perl}{type=main,  
name={Perl},  
description={A scripting language}}  
\newglossaryentry{tex}{name={\TeX},  
description={A typesetting language},sort={TeX}}  
\newglossaryentry{html}{type=\glsdefaulttype,  
name={html},  
description={A mark up language}}

and suppose in my document preamble I use the command:

\loadglsentries[languages]{myentries}

then this will add the entries tex and html to the glossary whose type is given by languages, but the entry perl will be added to the main glossary, since it explicitly sets the type to main.

____________________________

Note: if you use \newacronym (see §13 Acronyms and Other Abbreviations) the type is set as type=\acronymtype unless you explicitly override it. For example, if my file myacronyms.tex contains:

\newacronym{aca}{aca}{a contrived acronym}

then (supposing I have defined a new glossary type called altacronym)

\loadglsentries[altacronym]{myacronyms}

will add aca to the glossary type acronym, if the package option acronym has been specified, or will add aca to the glossary type altacronym, if the package option acronym is not specified.4.1

If you have used the acronym package option, there are two possible solutions to this problem:

  1. Change myacronyms.tex so that entries are defined in the form:
    \newacronym[type=\glsdefaulttype]{aca}{aca}{a  
    contrived acronym}

    and do:

    \loadglsentries[altacronym]{myacronyms}

  2. Temporarily change \acronymtype to the target glossary:
    \let\orgacronymtype\acronymtype  
    \renewcommand{\acronymtype}{altacronym}  
    \loadglsentries{myacronyms}  
    \let\acronymtype\orgacronymtype

Note that only those entries that have been used in the text will appear in the relevant glossaries. Note also that \loadglsentries may only be used in the preamble.

Remember that you can use \provideglossaryentry rather than \newglossaryentry. Suppose you want to maintain a large database of acronyms or terms that you’re likely to use in your documents, but you may want to use a modified version of some of those entries. (Suppose, for example, one document may require a more detailed description.) Then if you define the entries using \provideglossaryentry in your database file, you can override the definition by simply using \newglossaryentry before loading the file. For example, suppose your file (called, say, terms.tex) contains:

\provideglossaryentry{mallard}{name=mallard,  
 description={a type of duck}}

but suppose your document requires a more detailed description, you can do:

\usepackage{glossaries}  
\makeglossaries  
\newglossaryentry{mallard}{name=mallard,  
 description={a dabbling duck where the male has a green head}}  
\loadglsentries{terms}

Now the mallard definition in the terms.tex file will be ignored.

Top

4.7 Moving Entries to Another Glossary

As from version 3.02, you can move an entry from one glossary to another using:


\glsmoveentry{label}{target glossary label}

where ⟨label⟩ is the unique label identifying the required entry and ⟨target glossary label⟩ is the unique label identifying the glossary in which to put the entry.

Note that no check is performed to determine the existence of the target glossary. If you want to move an entry to a glossary that’s skipped by \printglossaries, then define an ignored glossary with \newignoredglossary. (See §12 Defining New Glossaries.)

Unpredictable results may occur if you move an entry to a different glossary from its parent or children.

Top

4.8 Drawbacks With Defining Entries in the Document Environment

Originally, \newglossaryentry (and \newacronym) could only be used in the preamble. I reluctantly removed this restriction in version 1.13, but there are issues with defining commands in the document environment instead of the preamble, which is why the restriction is maintained for newer commands. This restriction is also reimposed for \newglossaryentry by the new Option 1. (The glossaries-extra package automatically reimposes this restriction for Options 2 and 3 but provides a package option to allow document definitions.)

Top

4.8.1 Technical Issues

  1. If you define an entry mid-way through your document, but subsequently shuffle sections around, you could end up using an entry before it has been defined.
  2. Entry information is required when displaying the glossary. If this occurs at the start of the document, but the entries aren’t defined until later, then the entry details are being looked up before the entry has been defined.
  3. If you use a package, such as babel, that makes certain characters active at the start of the document environment, there will be a problem if those characters have a special significance when defining glossary entries. These characters include the double-quote " character, the exclamation mark ! character, the question mark ? character, and the pipe | character. They must not be active when defining a glossary entry where they occur in the sort key (and they should be avoided in the label if they may be active at any point in the document). Additionally, the comma , character and the equals = character should not be active when using commands that have ⟨key⟩=⟨value⟩ arguments.

To overcome the first two problems, as from version 4.0 the glossaries package modifies the definition of \newglossaryentry at the beginning of the document environment so that the definitions are written to an external file (\jobnameglsdefs) which is then read in at the start of the document on the next run. The entry will then only be defined in the document environment if it doesn’t already exist. This means that the entry can now be looked up in the glossary, even if the glossary occurs at the beginning of the document.

There are drawbacks to this mechanism: if you modify an entry definition, you need a second run to see the effect of your modification; this method requires an extra \newwrite, which may exceed TeX’s maximum allocation; unexpected expansion issues could occur; the see key isn’t stored, which means it can’t be added to the glsdefs file when it’s created at the end of the document (and therefore won’t be present on subsequent runs).

The glossaries-extra package provides a setting (but only for Options 2 and 3) that allows \newglossaryentry to occur in the document environment but doesn’t create the glsdefs file. This circumvents some problems but it means that you can’t display any of the glossaries before all the entries have been defined (so it’s all right if all the glossaries are at the end of the document but not if any occur in the front matter).

Top

4.8.2 Good Practice Issues

The above section covers technical issues that can cause your document to have compilation errors or produce incorrect output. This section focuses on good writing practice. The main reason cited by users wanting to define entries within the document environment rather than in the preamble is that they want to write the definition as they type in their document text. This suggests a “stream of consciousness” style of writing that may be acceptable in certain literary genres but is inappropriate for factual documents.

When you write technical documents, regardless of whether it’s a PhD thesis or an article for a journal or proceedings, you must plan what you write in advance. If you plan in advance, you should have a fairly good idea of the type of terminology that your document will contain, so while you are planning, create a new file with all your entry definitions. If, while you’re writing your document, you remember another term you need, then you can switch over to your definition file and add it. Most text editors have the ability to have more than one file open at a time. The other advantage to this approach is that if you forget the label, you can look it up in the definition file rather than searching through your document text to find the definition.

Top

5. Number lists

Each entry in the glossary has an associated number list. By default, these numbers refer to the pages on which that entry has been indexed (using any of the commands described in §6 Links to Glossary Entries and §7 Adding an Entry to the Glossary Without Generating Text). The number list can be suppressed using the nonumberlist package option, or an alternative counter can be set as the default using the counter package option. The number list is also referred to as the location list.

Number lists are more common with indexes rather than glossaries (although you can use the glossaries package for indexes as well). However, the glossaries package makes use of makeindex or xindy to hierarchically sort and collate the entries since they are readily available with most modern TeX distributions. Since these are both designed as indexing applications they both require that terms either have a valid location or a cross-reference. Even if you use nonumberlist, the locations must still be provided and acceptable to the indexing application or they will cause an error during the indexing stage, which will interrupt the document build. However, if you’re not interested in the locations, each entry only needs to be indexed once, so consider using indexonlyfirst, which can improve the document build time by only indexing the first use of each term.

The \glsaddall command (see §7 Adding an Entry to the Glossary Without Generating Text), which is used to automatically index all entries, iterates over all defined entries and does \glsadd{label} for each entry (where ⟨label⟩ is that entry’s label). This means that \glsaddall automatically adds the same location to every entry’s number list, which looks weird if the number list hasn’t been suppressed.

With Option 4, the indexing is performed by bib2gls, which was specifically designed for the glossaries-extra package. So it will allow any location format, and its selection=all option will select all entries without adding an unwanted location to the number list. If bib2gls can deduce a numerical value for a location, it will attempt to form a range over consecutive locations, otherwise it won’t try to form a range and the location will just form an individual item in the list. Option 1 also allows any location but it doesn’t form ranges.

Top

5.1 Encap Values

Each location in the number list is encapsulated with a command formed from the encap value. By default this is the \glsnumberformat command, which corresponds to the encap glsnumberformat, but this may be overridden using the format key in the optional argument to commands like \gls. (See §6 Links to Glossary Entries.) For example, you may want the location to appear in bold to indicate the principle use of a term or symbol. If the encap starts with an open parenthesis ( this signifies the start of a range and if the encap starts with close parenthesis ) this signifies the end of a range. These must always occur in matching pairs.

The glossaries package provides the command \glsignore which ignores its argument. This is the format used by \glsaddallunused to suppress the location, which works fine as long as no other locations are added to the number list. For example, if you use \gls{sample} on page 2 then reset the first use flag and then use \glsaddallunused on page 10, the number list for sample will be 2, \glsignore{10} which will result in “2, ” which has a spurious comma.

This isn’t a problem with bib2gls because you’d use selection=all instead of \glsaddallunused, but even if you explicitly had, for example, \gls[format=glsignore]{label} for some reason, bib2gls will recognise glsignore as a special encap indicating an ignored location, so it will select the entry but not add that location to the number list. It’s a problem for all the other options (except Option 5, which doesn’t perform any indexing).

Complications can arise if you use different encap values for the same location. For example, suppose on page 10 you have both the default glsnumberformat and textbf encaps. While it may seem apparent that textbf should override glsnumberformat in this situation, the indexing application may not know it. This is therefore something you need to be careful about if you use the format key or if you use a command that implicitly sets it.

In the case of xindy, it only accepts one encap (according to the order of precedence given in the xindy module) and discards the others for identical locations (for the same entry). This can cause a problem if a discarded location forms the start or end of a range.

In the case of makeindex, it accepts different encaps for the same location, but warns about it. This leads to a number list with the same location repeated in different formats. If you use the makeglossaries Perl script with Option 2 it will detect makeindex’s warning and attempt to fix the problem, ensuring that the glsnumberformat encap always has the least precedence unless it includes a range identifier. Other conflicting encaps will have the last one override earlier ones for the same location with range identifiers taking priority.

No discard occurs with Option 1 so again you get the same location repeated in different formats. With Option 4, bib2gls will discard according to order of precedence, giving priority to start and end range encaps. (See the bib2gls manual for further details.)

Top

5.2 Locations

Neither Option 1 nor Option 4 care about the location syntax as long as it’s valid LaTeX code (and doesn’t contain fragile commands). In both cases, the indexing is performed by writing a line to the aux file. The write operation is deferred to avoid the problems associated with TeX’s asynchronous output routine. (See, for example, Finding if you’re on an odd or an even page for more details on this issue.) Unfortunately Options 2 and 3 are far more problematic and need some complex code to deal with awkward locations.

If you know that your locations will always expand to a format acceptable to your chosen indexing application then use the package option esclocations=false to bypass this operation. This setting only affects Options 2 and 3 as the problem doesn’t arise with the other indexing options.

Both makeindex and xindy are fussy about the syntax of the locations. In the case of makeindex, only the numbering system obtained with \arabic, \roman, \Roman, \alph and \Alph or composites formed from them with the same separator (set with \glsSetCompositor{char}) are accepted. (makeindex won’t accept an empty location.) In the case of xindy, you can define your own location classes, but if the location contains a robust command then the leading backslash must be escaped. The glossaries package tries to do this, but it’s caught between two conflicting requirements:

  1. The location must be fully expanded before \ can be converted to \\ (there’s no point converting \thepage to \\thepage);
  2. The page number can’t be expanded until the deferred write operation (so \c@page mustn’t expand in the previous step but \the\c@page mustn’t be converted to \\the\\c@page and \number\c@page mustn’t be converted to \\number\\c@page etc).

There’s a certain amount of trickery needed to deal with this conflict and the code requires the location to be in a form that doesn’t embed the counter’s internal register in commands like \value. For example, suppose you have a robust command called \tallynum that takes a number as the argument and an expandable command called \tally that converts a counter name into the associated register or number to pass to \tallynum. Let’s suppose that this command is used to represent the page number:

\renewcommand{\thepage}{\tally{page}}

Now let’s suppose that a term is indexed at the beginning of page 2 at the end of a paragraph that started on page 1. With xindy, the location \tally{page} needs to be written to the file as \\tallynum{2}. If it’s written as \tallynum{2} then xindy will interpret \t as the character “t” (which means the location would appear as “tallynum2”). So glossaries tries to expand \thepage without expanding \c@page and then escapes all the backslashes, except for the page counter’s internal command. The following definitions of \tally will work:

The following don’t work:

If you have a numbering system where \cs name{page} expands to \internal cs name\c@page (for example, if \tally{page} expands to \tallynum\c@page) then you need to use:


\glsaddprotectedpagefmt{internal cs name}

Note that the backslash must be omitted from ⟨internal cs name⟩ and the corresponding command must be able to process a count register as the (sole) argument.

For example, suppose you have a style samplenum that is implemented as follows:

 \newcommand*{\samplenum}[1]{%  
   \expandafter\@samplenum\csname c@#1\endcsname}  
 \newcommand*{\@samplenum}[1]{\two@digits{#1}}

(That is, it displays the value of the counter as a two-digit number.) Then to ensure the location is correct for entries in page-spanning paragraphs, you need to do:

\glsaddprotectedpagefmt{@samplenum}

(If you are using a different counter for the location, such as section or equation, you don’t need to worry about this provided the inner command is expandable.)

If the inner macro (as given by \internal cs name⟩) contains non-expandable commands then you may need to redefine \glsinternal cs namepage after using

\glsaddprotectedpagefmt{internal cs name}
This command doesn’t take any arguments as the location is assumed to be given by \c@page because that’s the only occasion this command should be used. For example, suppose now my page counter format uses small caps Roman numerals:
 \newcommand*{\samplenum}[1]{%  
  \expandafter\@samplenum\csname c@#1\endcsname}  
 \newcommand*{\@samplenum}[1]{\textsc{\romannumeral#1}}

Again, the inner macro needs to be identified using:

\glsaddprotectedpagefmt{@samplenum}

However, since \textsc isn’t fully expandable, the location is written to the file as \textsc  {i} (for page 1), \textsc  {ii} (for page 2), etc. This format may cause a problem for the indexing application (particularly for makeindex). To compensate for this, the \glsinternal cs namepage command may be redefined so that it expands to a format that’s acceptable to the indexing application. For example:

\renewcommand*{\gls@samplenumpage}{\romannumeral\c@page}

While this modification means that the number list in the glossary won’t exactly match the format of the page numbers (displaying lower case Roman numbers instead of small cap Roman numerals) this method will at least work correctly for both makeindex and xindy. If you are using xindy, the following definition:

 \renewcommand*{\gls@samplenumpage}{%  
   \glsbackslash\string\textsc{\romannumeral\c@page}}

combined with

\GlsAddXdyLocation{romansc}{:sep "\string\textsc\glsopenbrace"  
 "roman-numbers-lowercase" :sep "\glsclosebrace"}

will now have lowercase Roman numerals in the location list (see §11.2 Locations and Number lists for further details on that command). Take care of the backslashes. The location (which ends up in the :locref attribute) needs \\ but the location class (identified with \GlsAddXdyLocation) just has a single backslash. Note that this example will cause problems if your locations should be hyperlinks.

Another possibility that may work with both makeindex and xindy is to redefine \glsinternal cs namepage (\gls@samplenumpage in this example) to just expand to the decimal page number (\number\c@page) and redefine \glsnumberformat to change the displayed format:

\renewcommand*{\gls@samplenumpage}{\number\c@page}  
\renewcommand*{\glsnumberformat}[1]{\textsc{\romannumeral#1}}

If you redefine \glsinternal cs namepage, you must make sure that \c@page is expanded when it’s written to the file. (So don’t, for example, hide \c@page inside a robust command.)

The mechanism that allows this to work temporarily redefines \the and \number while it processes the location. If this causes a problem you can disallow it using


\glswrallowprimitivemodsfalse

but you will need to find some other way to ensure the location expands correctly.

Top

5.3 Range Formations

Both makeindex and xindy (Options 2 and 3) concatenate a sequence of 3 or more consecutive pages into a range. With xindy (Option 3) you can vary the minimum sequence length using \GlsSetXdyMinRangeLength {n} where ⟨n⟩ is either an integer or the keyword none which indicates that there should be no range formation (see §11.2 Locations and Number lists for further details).

Note that \GlsSetXdyMinRangeLength must be used before \makeglossaries and has no effect if \noist is used.

With both makeindex and xindy (Options 2 and 3), you can replace the separator and the closing number in the range using:


\glsSetSuffixF{suffix}


\glsSetSuffixFF{suffix}

where the former command specifies the suffix to use for a 2 page list and the latter specifies the suffix to use for longer lists. For example:

\glsSetSuffixF{f.}  
\glsSetSuffixFF{ff.}

Note that if you use xindy (Option 3), you will also need to set the minimum range length to 1 if you want to change these suffixes:

\GlsSetXdyMinRangeLength{1}

Note that if you use the hyperref package, you will need to use \nohyperpage in the suffix to ensure that the hyperlinks work correctly. For example:

\glsSetSuffixF{\nohyperpage{f.}}  
\glsSetSuffixFF{\nohyperpage{ff.}}

Note that \glsSetSuffixF and \glsSetSuffixFF must be used before \makeglossaries and have no effect if \noist is used.

It’s also possible to concatenate a sequence of consecutive locations into a range or have suffixes with Option 4, but with bib2gls these implicit ranges can’t be merged with explicit ranges (created with the ( and ) encaps). See the bib2gls manual for further details.

Option 1 doesn’t form ranges. However, with this option you can iterate over an entry’s number list using:


\glsnumberlistloop{label}{handler cs}{xr handler cs}

where ⟨label⟩ is the entry’s label and ⟨handler cs⟩ is a handler control sequence of the form:


handler cs{prefix}{counter}{format}{location}

where ⟨prefix⟩ is the hyperref prefix, ⟨counter⟩ is the name of the counter used for the location, ⟨format⟩ is the format used to display the location (e.g. textbf) and ⟨location⟩ is the location. The third argument is the control sequence to use for any cross-references in the list. This handler should have the syntax:


xr handler cs[tag]{xr list}

where ⟨tag⟩ is the cross-referenced text (e.g. “see”) and ⟨xr list⟩ is a comma-separated list of labels. (This actually has a third argument but it’s always empty when used with Option 1.)

For example, if on page 12 I have used

\gls[format=textbf]{apple}

and on page 18 I have used

\gls[format=emph]{apple}

then

\glsnumberlistloop{apple}{\myhandler}

will be equivalent to:

\myhandler{}{page}{textbf}{12}%  
\myhandler{}{page}{emph}{18}%

There is a predefined handler that’s used to display the number list in the glossary:


\glsnoidxdisplayloc{prefix}{counter}{format}{location}

The predefined handler used for the cross-references in the glossary is:


\glsseeformat[tag]{xr list}{location}

which is described in §8.1 Customising Cross-reference Text.

\glsnumberlistloop is not available for Options 2 and 3.

Top

5.4 Style Hook

As from version 4.24, there’s a hook that’s used near the end of \writeist before the file is closed. You can set the code to be performed then using:


\GlsSetWriteIstHook{code}

If you want the ⟨code⟩ to write any information to the file, you need to use


\write\glswrite{style information}

Make sure you use the correct format within ⟨style information⟩. For example, if you are using makeindex:

\GlsSetWriteIstHook{%  
 \write\glswrite{page_precedence "arnAR"}%  
 \write\glswrite{line_max 80}%  
}

This changes the page type precedence and the maximum line length used by makeindex.

Remember that if you switch to xindy, this will no longer be valid code.

Top

6. Links to Glossary Entries

Once you have defined a glossary entry using \newglossaryentry (§4 Defining Glossary Entries) or \newacronym (see §13 Acronyms and Other Abbreviations), you can refer to that entry in the document using one of the commands listed in §6.1 The \gls-Like Commands (First Use Flag Queried) or §6.2 The \glstext-Like Commands (First Use Flag Not Queried). The text which appears at that point in the document when using one of these commands is referred to as the link text (even if there are no hyperlinks). These commands also add a line to an external file that is used to generate the relevant entry in the glossary. This information includes an associated location that is added to the number list for that entry. By default, the location refers to the page number. For further information on number lists, see §5 Number lists. These external files need to be post-processed by makeindex or xindy unless you have chosen Options 1 or 4. If you don’t use \makeglossaries these external files won’t be created. (Options 1 and 4 write the information to the aux file.)

Note that repeated use of these commands for the same entry can cause the number list to become quite long, which may not be particular helpful to the reader. In this case, you can use the non-indexing commands described in §9 Using Glossary Terms Without Links or you can use the supplemental glossaries-extra package, which provides a means to suppress the automated indexing of the commands listed in this chapter.

I strongly recommend that you don’t use the commands defined in this chapter in the arguments of sectioning or caption commands or any other command that has a moving argument.

Aside from problems with expansion issues, PDF bookmarks and possible nested hyperlinks in the table of contents (or list of whatever) any use of the commands described in §6.1 The \gls-Like Commands (First Use Flag Queried) will have their first use flag unset when they appear in the table of contents (or list of whatever).

The above warning is particularly important if you are using the glossaries package in conjunction with the hyperref package. Instead, use one of the expandable commands listed in §9 Using Glossary Terms Without Links (such as \glsentrytext but not the non-expandable case changing versions like \Glsentrytext). Alternatively, provide an alternative via the optional argument to the sectioning/caption command or use hyperref’s \texorpdfstring. Examples:

\chapter{An overview of \glsentrytext{perl}}  
\chapter[An overview of Perl]{An overview of \gls{perl}}  
\chapter{An overview of \texorpdfstring{\gls{perl}}{Perl}}

If you want to retain the formatting that’s available through commands like \acrshort (for example, if you are using one of the small caps styles), then you might want to consider the glossaries-extra package which provides commands for this purpose.

If you want the link text to produce a hyperlink to the corresponding entry details in the glossary, you should load the hyperref package before the glossaries package. That’s what I’ve done in this document, so if you see a hyperlinked term, such as link text, you can click on the word or phrase and it will take you to a brief description in this document’s glossary.

If you use the hyperref package, I strongly recommend you use pdflatex rather than latex to compile your document, if possible. The DVI format of LaTeX has limitations with the hyperlinks that can cause a problem when used with the glossaries package. Firstly, the DVI format can’t break a hyperlink across a line whereas PDFLaTeX can. This means that long glossary entries (for example, the full form of an acronym) won’t be able to break across a line with the DVI format. Secondly, the DVI format doesn’t correctly size hyperlinks in subscripts or superscripts. This means that if you define a term that may be used as a subscript or superscript, if you use the DVI format, it won’t come out the correct size.

These are limitations of the DVI format not of the glossaries package.

It may be that you only want terms in certain glossaries to have hyperlinks, but not for other glossaries. In this case, you can use the package option nohypertypes to identify the glossary lists that shouldn’t have hyperlinked link text. See §2.1 General Options for further details.

The way the link text is displayed depends on


\glstextformat{text}

For example, to make all link text appear in a sans-serif font, do:

\renewcommand*{\glstextformat}[1]{\textsf{#1}}

Further customisation can be done via \defglsentryfmt or by redefining \glsentryfmt. See §6.3 Changing the format of the link text for further details.

Each entry has an associated conditional referred to as the first use flag. Some of the commands described in this chapter automatically unset this flag and can also use it to determine what text should be displayed. These types of commands are the \gls-like commands and are described in §6.1 The \gls-Like Commands (First Use Flag Queried). The commands that don’t reference or change the first use flag are \glstext-like commands and are described in §6.2 The \glstext-Like Commands (First Use Flag Not Queried). See §14 Unsetting and Resetting Entry Flags for commands that unset (mark the entry as having been used) or reset (mark the entry as not used) the first use flag without referencing the entries.

The \gls-like and \glstext-like commands all take a first optional argument that is a comma-separated list of ⟨key⟩=⟨value⟩ options, described below. They also have a star-variant, which inserts hyper=false at the start of the list of options and a plus-variant, which inserts hyper=true at the start of the list of options. For example \gls*{sample} is the same as \gls[hyper=false]{sample} and \gls+{sample} is the same as \gls[hyper=true]{sample}, whereas just \gls{sample} will use the default hyperlink setting which depends on a number of factors (such as whether the entry is in a glossary that has been identified in the nohypertypes list). You can override the hyper key in the variant’s optional argument, for example, \gls*[hyper=true]{sample} but this creates redundancy and is best avoided. The glossaries-extra package provides the option to add a third custom variant.

Avoid nesting these commands. For example don’t do \glslink{label}{\gls{label2}} as this is likely to cause problems. By implication, this means that you should avoid using any of these commands within the text, first, short or long keys (or their plural equivalent) or any other key that you plan to access through these commands. (For example, the symbol key if you intend to use \glssymbol.)

The keys listed below are available for the optional argument. The glossaries-extra package provides additional keys. (See the glossaries-extra manual for further details.)

hyper
This is a boolean key which can be used to enable/disable the hyperlink to the relevant entry in the glossary. If this key is omitted, the value is determined by current settings, as indicated above. For example, when used with a \gls-like command, if this is the first use and the hyperfirst=false package option has been used, then the default value is hyper=false. The hyperlink can be forced on using hyper=true unless the hyperlinks have been suppressed using \glsdisablehyper. You must load the hyperref package before the glossaries package to ensure the hyperlinks work.
format
This specifies how to format the associated location number for this entry in the glossary. This value is equivalent to the makeindex encap value, and (as with \index) the value needs to be the name of a command without the initial backslash. As with \index, the characters ( and ) can also be used to specify the beginning and ending of a number range and they must be in matching pairs. (For example, \gls[format={(}]{sample} on one page to start the range and later \gls[format={)}]{sample} to close the range.) Again as with \index, the command should be the name of a command which takes an argument (which will be the associated location). Be careful not to use a declaration (such as bfseries) instead of a text block command (such as textbf) as the effect is not guaranteed to be localised. If you want to apply more than one style to a given entry (e.g. bold and italic) you will need to create a command that applies both formats, e.g. 
\newcommand*{\textbfem}[1]{\textbf{\emph{#1}}}

and use that command.

In this document, the standard formats refer to the standard text block commands such as \textbf or \emph or any of the commands listed in table 6.1. You can combine a range and format using (format⟩ to start the range and )format⟩ to end the range. The ⟨format⟩ part must match. For example, format={(emph} and format={)emph}.

If you use xindy instead of makeindex, you must specify any non-standard formats that you want to use with the format key using \GlsAddXdyAttribute{name}. So if you use xindy with the above example, you would need to add:

\GlsAddXdyAttribute{textbfem}

See §11 Xindy (Option 3) for further details.

If you are using hyperlinks and you want to change the font of the hyperlinked location, don’t use \hyperpage (provided by the hyperref package) as the locations may not refer to a page number. Instead, the glossaries package provides number formats listed in table 6.1. These commands are designed to work with the particular location formats created by makeindex and xindy and shouldn’t be used in other contexts.


Table 6.1: Predefined Hyperlinked Location Formats
hyperrm serif hyperlink
hypersf sans-serif hyperlink
hypertt monospaced hyperlink
hyperbf bold hyperlink
hypermd medium weight hyperlink
hyperit italic hyperlink
hypersl slanted hyperlink
hyperup upright hyperlink
hypersc small caps hyperlink
hyperemph emphasized hyperlink

Note that if the \hyperlink command hasn’t been defined, the hyperxx⟩ formats are equivalent to the analogous textxx⟩ font commands (and hyperemph is equivalent to emph). If you want to make a new format, you will need to define a command which takes one argument and use that. For example, if you want the location number to be in a bold sans-serif font, you can define a command called, say, \hyperbsf:

\newcommand{\hyperbsf}[1]{\textbf{\hypersf{#1}}}

and then use hyperbsf as the value for the format key.6.1

When defining a custom location format command that uses one of the \hyperxx commands, make sure that the argument of \hyperxx is just the location. Any formatting must be outside of \hyperxx (as in the above \hyperbfsf example).

Remember that if you use xindy, you will need to add this to the list of location attributes:

\GlsAddXdyAttribute{hyperbsf}

counter
This specifies which counter to use for this location. This overrides the default counter used by this entry. (See also §5 Number lists.)
local
This is a boolean key that only makes a difference when used with \gls-like commands that change the entry’s first use flag. If local=true, the change to the first use flag will be localised to the current scope. The default is local=false.
noindex
This is a boolean key that suppresses the indexing. Only available with glossaries-extra.
hyperoutside
This is a boolean key that determines whether to put the hyperlink outside of \glstextformat. Only available with glossaries-extra.
wrgloss
This key determines whether to index before (wrgloss=before) or after (wrgloss=after) the link text. Only available with glossaries-extra.
textformat
This key identifies the name of the control sequence to encapsulate the link text instead of the default \glstextformat. Only available with glossaries-extra.
prefix
This key locally redefines \glolinkprefix to the given value. Only available with glossaries-extra.
thevalue
This key explicitly sets the location. Only available with glossaries-extra.
theHvalue
This key explicitly sets the hyperlink location. Only available with glossaries-extra.

The link text isn’t scoped by default with just the base glossaries package. Any unscoped declarations in the link text may affect subsequent text.

Top

6.1 The \gls-Like Commands (First Use Flag Queried)

This section describes the commands that unset (mark as used) the first use flag on completion, and in most cases they use the current state of the flag to determine the text to be displayed. As described above, these commands all have a star-variant (hyper=false) and a plus-variant (hyper=true) and have an optional first argument that is a ⟨key⟩=⟨value⟩ list. These commands use \glsentryfmt or the equivalent definition provided by \defglsentryfmt to determine the automatically generated text and its format (see §6.3 Changing the format of the link text).

Apart from \glsdisp, the commands described in this section also have a final optional argument ⟨insert⟩ which may be used to insert material into the automatically generated text.

Since the commands have a final optional argument, take care if you actually want to display an open square bracket after the command when the final optional argument is absent. Insert an empty set of braces {} immediately before the opening square bracket to prevent it from being interpreted as the final argument. For example:

\gls{sample} {}[Editor's comment]

Don’t use any of the \gls-like or \glstext-like commands in the ⟨insert⟩ argument.

Take care using these commands within commands or environments that are processed multiple times as this can confuse the first use flag query and state change. This includes frames with overlays in beamer and the tabularx environment provided by tabularx. The glossaries package automatically deals with this issue in amsmath’s align environment. You can apply a patch to tabularx by placing the following command (new to v4.28) in the preamble:


\glspatchtabularx

This does nothing if tabularx hasn’t been loaded. There’s no patch available for beamer. See §14 Unsetting and Resetting Entry Flags for more details.


\gls[options]{label}[insert]

This command typically determines the link text from the values of the text or first keys supplied when the entry was defined using \newglossaryentry. However, if the entry was defined using \newacronym and \setacronymstyle was used, then the link text will usually be determined from the long or short keys.

There are two upper case variants:


\Gls[options]{label}[insert]

and


\GLS[options]{label}[insert]

which make the first letter of the link text or all the link text upper case, respectively. For the former, the uppercasing of the first letter is performed by \makefirstuc.

The first letter uppercasing command \makefirstuc has limitations which must be taken into account if you use \Gls or any of the other commands that convert the first letter to uppercase.

The upper casing is performed as follows:

(Note the use of the sort key in the above examples.)

There are hundreds of LaTeX packages that altogether define thousands of commands with various syntax and it’s impossible for mfirstuc to take them all into account. The above rules are quite simplistic and are designed for link text that starts with a text-block command (such as \emph) or a command that produces a character (such as \oe). This means that if your link text starts with something that doesn’t adhere to mfirstuc’s assumptions then things are likely to go wrong.

For example, starting with a math-shift symbol:

\newglossaryentry{sample}{  
  name={$a$},  
  sort={a},  
  description={an example}  
}

This falls into case 2 above, so the link text will be set to

\MakeUppercase $a$

This attempts to uppercase the math-shift $, which will go wrong. In this case it’s not appropriate to perform any case-changing, but it may be that you want to use \Gls programmatically without checking if the text contains any maths. In this case, the simplest solution is to insert an empty brace at the start:

\newglossaryentry{sample}{  
  name={{}$a$},  
  sort={a},  
  description={an example}  
}

Now the link text will be set to

\MakeUppercase{}$a$

and the \uppercase becomes harmless.

Another issue occurs when the link text starts with a command followed by an argument (case 1) but the argument is a label, identifier or something else that shouldn’t have a case-change. A common example is when the link text starts with one of the commands described in this chapter. (But you haven’t done that, have you? What with the warning not to do it at the beginning of the chapter.) Or when the link text starts with one of the non-linking commands described in §9 Using Glossary Terms Without Links. For example:

\newglossaryentry{sample}{name={sample},description={an example}}  
\newglossaryentry{sample2}{  
   name={\glsentrytext{sample} two},  
   sort={sample two},  
   description={another example}  
 }

Now the link text will be set to:

\glsentrytext{\MakeUppercase sample} two

This will generate an error because there’s no entry with the label “\MakeUppercase sample”. The best solution here is to write the term out in the text field and use the command in the name field. If you don’t use \glsname anywhere in your document, you can use \gls in the name field:

\newglossaryentry{sample}{name={sample},description={an example}}  
\newglossaryentry{sample2}{  
   name={\gls{sample} two},  
   sort={sample two},  
   text={sample two},  
   description={another example}  
 }

If the link text starts with a command that has an optional argument or with multiple arguments where the actual text isn’t in the first argument, then \makefirstuc will also fail. For example:

\newglossaryentry{sample}{  
 name={\textcolor{blue}{sample} phrase},  
 sort={sample phrase},  
 description={an example}}

Now the link text will be set to:

\textcolor{\MakeUppercase blue}{sample} phrase

This won’t work because \MakeUppercase blue isn’t a recognised colour name. In this case you will have to define a helper command where the first argument is the text. For example:

\newglossaryentry{sample}{  
\newcommand*{\blue}[1]{\textcolor{blue}{#1}}  
 name={\blue{sample} phrase},  
 sort={sample phrase},  
 description={an example}}

In fact, since the whole design ethos of LaTeX is the separation of content and style, it’s better to use a semantic command. For example:

\newglossaryentry{sample}{  
\newcommand*{\keyword}[1]{\textcolor{blue}{#1}}  
 name={\keyword{sample} phrase},  
 sort={sample phrase},  
 description={an example}}

For further details see the mfirstuc user manual.

There are plural forms that are analogous to \gls:


\glspl[options]{label}[insert]


\Glspl[options]{label}[insert]


\GLSpl[options]{label}[insert]

These typically determine the link text from the plural or firstplural keys supplied when the entry was defined using \newglossaryentry or, if the entry is an abbreviation and \setacronymstyle was used, from the longplural or shortplural keys.

Be careful when you use glossary entries in math mode especially if you are using hyperref as it can affect the spacing of subscripts and superscripts. For example, suppose you have defined the following entry:

\newglossaryentry{Falpha}{name={F_\alpha},  
description=sample}

and later you use it in math mode:

$\gls{Falpha}^2$

This will result in Fα2 instead of Fα2. In this situation it’s best to bring the superscript into the hyperlink using the final ⟨insert⟩ optional argument:

$\gls{Falpha}[^2]$


\glsdisp[options]{label}{link text}

This behaves in the same way as the above commands, except that the ⟨link text⟩ is explicitly set. There’s no final optional argument as any inserted material can be added to the ⟨link text⟩ argument.

Don’t use any of the \gls-like or \glstext-like commands in the ⟨link text⟩ argument of \glsdisp.

Top

6.2 The \glstext-Like Commands (First Use Flag Not Queried)

This section describes the commands that don’t change or reference the first use flag. As described above, these commands all have a star-variant (hyper=false) and a plus-variant (hyper=true) and have an optional first argument that is a ⟨key⟩=⟨value⟩ list. These commands also don’t use \glsentryfmt or the equivalent definition provided by \defglsentryfmt (see §6.3 Changing the format of the link text). Additional commands for abbreviations are described in §13 Acronyms and Other Abbreviations.

Apart from \glslink, the commands described in this section also have a final optional argument ⟨insert⟩ which may be used to insert material into the automatically generated text. See the caveat above in §6.1 The \gls-Like Commands (First Use Flag Queried).


\glslink[options]{label}{link text}

This command explicitly sets the link text as given in the final argument.

Don’t use any of the \gls-like or \glstext-like commands in the argument of \glslink. By extension, this means that you can’t use them in the value of fields that are used to form link text.


\glstext[options]{label}[insert]

This command always uses the value of the text key as the link text.

There are also analogous commands:


\Glstext[options]{text}[insert]


\GLStext[options]{text}[insert]

These convert the first character or all the characters to uppercase, respectively. See the note on \Gls above for details on the limitations of converting the first letter to upper case.

There’s no equivalent command for title-casing, but you can use the more generic command \glsentrytitlecase in combination with \glslink. For example:

\glslink{sample}{\glsentrytitlecase{sample}{text}}

(See §9 Using Glossary Terms Without Links.)


\glsfirst[options]{label}[insert]

This command always uses the value of the first key as the link text.

There are also analogous uppercasing commands:


\Glsfirst[options]{text}[insert]


\GLSfirst[options]{text}[insert]

The value of the first key (and firstplural key) doesn’t necessarily match the text produced by \gls (or \glspl) on first use as the link text used by \gls may be modified through commands like \defglsentry. (Similarly, the value of the text and plural keys don’t necessarily match the link text used by \gls or \glspl on subsequent use.)


\glsplural[options]{label}[insert]

This command always uses the value of the plural key as the link text.

There are also analogous uppercasing commands:


\Glsplural[options]{text}[insert]


\GLSplural[options]{text}[insert]


\glsfirstplural[options]{label}[insert]

This command always uses the value of the firstplural key as the link text.

There are also analogous uppercasing commands:


\Glsfirstplural[options]{text}[insert]


\GLSfirstplural[options]{text}[insert]


\glsname[options]{label}[insert]

This command always uses the value of the name key as the link text. Note that this may be different from the values of the text or first keys. In general it’s better to use \glstext or \glsfirst instead of \glsname.

There are also analogous uppercasing commands:


\Glsname[options]{text}[insert]


\GLSname[options]{text}[insert]

In general it’s best to avoid \Glsname with acronyms. Instead, consider using \Acrlong, \Acrshort or \Acrfull.


\glssymbol[options]{label}[insert]

This command always uses the value of the symbol key as the link text.

There are also analogous uppercasing commands:


\Glssymbol[options]{text}[insert]


\GLSsymbol[options]{text}[insert]


\glsdesc[options]{label}[insert]

This command always uses the value of the description key as the link text.

There are also analogous uppercasing commands:


\Glsdesc[options]{text}[insert]


\GLSdesc[options]{text}[insert]

If you want the title case version you can use

\glslink{sample}{\glsentrytitlecase{sample}{desc}}


\glsuseri[options]{label}[insert]

This command always uses the value of the user1 key as the link text.

There are also analogous uppercasing commands:


\Glsuseri[options]{text}[insert]


\GLSuseri[options]{text}[insert]


\glsuserii[options]{text}[insert]

This command always uses the value of the user2 key as the link text.

There are also analogous uppercasing commands:


\Glsuserii[options]{text}[insert]


\GLSuserii[options]{text}[insert]


\glsuseriii[options]{text}[insert]

This command always uses the value of the user3 key as the link text.

There are also analogous uppercasing commands:


\Glsuseriii[options]{text}[insert]


\GLSuseriii[options]{text}[insert]


\glsuseriv[options]{text}[insert]

This command always uses the value of the user4 key as the link text.

There are also analogous uppercasing commands:


\Glsuseriv[options]{text}[insert]


\GLSuseriv[options]{text}[insert]


\glsuserv[options]{text}[insert]

This command always uses the value of the user5 key as the link text.

There are also analogous uppercasing commands:


\Glsuserv[options]{text}[insert]


\GLSuserv[options]{text}[insert]


\glsuservi[options]{text}[insert]

This command always uses the value of the user6 key as the link text.

There are also analogous uppercasing commands:


\Glsuservi[options]{text}[insert]


\GLSuservi[options]{text}[insert]

Top

6.3 Changing the format of the link text

The glossaries-extra package provides ways of altering the format according to the category. See the glossaries-extra manual for further details.

The default format of the link text for the \gls-like commands is governed by6.3:


\glsentryfmt

This may be redefined but if you only want the change the display style for a given glossary, then you need to use


\defglsentryfmt[type]{definition}

instead of redefining \glsentryfmt. The optional first argument ⟨type⟩ is the glossary type. This defaults to \glsdefaulttype if omitted. The second argument is the entry format definition.

Note that \glsentryfmt is the default display format for entries. Once the display format has been changed for an individual glossary using \defglsentryfmt, redefining \glsentryfmt won’t have an effect on that glossary, you must instead use \defglsentryfmt again. Note that glossaries that have been identified as lists of acronyms (via the package option acronymlists or the command \DeclareAcronymList, see §2.7 Acronym and Abbreviation Options) use \defglsentryfmt to set their display style.

Within the ⟨definition⟩ argument of \defglsentryfmt, or if you want to redefine \glsentryfmt, you may use the following commands:


\glslabel

This is the label of the entry being referenced. As from version 4.08, you can also access the glossary entry type using:


\glstype

This is defined using \edef so the replacement text is the actual glossary type rather than \glsentrytype{\glslabel}.


\glscustomtext

This is the custom text supplied in \glsdisp. It’s always empty for \gls, \glspl and their upper case variants. (You can use etoolbox’s \ifdefempty to determine if \glscustomtext is empty.)


\glsinsert

The custom text supplied in the final optional argument to \gls, \glspl and their upper case variants.


\glsifplural{true text}{false text}

If \glspl, \Glspl or \GLSpl was used, this command does ⟨true text⟩ otherwise it does ⟨false text⟩.


\glscapscase{no case}{first uc}{all caps}

If \gls, \glspl or \glsdisp were used, this does ⟨no case⟩. If \Gls or \Glspl were used, this does ⟨first uc⟩. If \GLS or \GLSpl were used, this does ⟨all caps⟩.


\glsifhyperon{hyper true}{hyper false}

This will do ⟨hyper true⟩ if the hyperlinks are on for the current reference, otherwise it will do ⟨hyper false⟩. The hyperlink may be off even if it wasn’t explicitly switched off with the hyper key or the use of a starred command. It may be off because the hyperref package hasn’t been loaded or because \glsdisablehyper has been used or because the entry is in a glossary type that’s had the hyperlinks switched off (using nohypertypes) or because it’s the first use and the hyperlinks have been suppressed on first use.

The \glsifhyperon command should be used instead of \glsifhyper, which is now deprecated (and no longer documented).

If you want to know if the command used to reference this entry was used with the star or plus variant, you can use:


\glslinkvar{unmodified}{star}{plus}

This will do ⟨unmodified⟩ if the unmodified version was used, or will do ⟨star⟩ if the starred version was used, or will do ⟨plus⟩ if the plus version was used. Note that this doesn’t take into account if the hyper key was used to override the default setting, so this command shouldn’t be used to guess whether or not the hyperlink is on for this reference.

Note that you can also use commands such as \ifglsused within the definition of \glsentryfmt (see §14 Unsetting and Resetting Entry Flags).

The commands \glslabel, \glstype, \glsifplural, \glscapscase, \glsinsert and \glscustomtext are typically updated at the start of the \gls-like and \glstext-like commands so they can usually be accessed in the hook user commands, such as \glspostlinkhook and \glslinkpostsetkeys.

This means that using commands like \gls within the fields that are accessed using the \gls-like or \glstext-like commands (such as the first, text, long or short keys) will cause a problem. The entry formatting performed by \glsentryfmt and related commands isn’t scoped (otherwise if would cause problems for \glspostlinkhook which may need to look ahead as well as look behind). This means that any nested commands will, at the very least, change the label stored in \glslabel.

If you only want to make minor modifications to \glsentryfmt, you can use


\glsgenentryfmt

This uses the above commands to display just the first, text, plural or firstplural keys (or the custom text) with the insert text appended.

Alternatively, if you want to change the entry format for abbreviations (defined via \newacronym) you can use:


\glsgenacfmt

This uses the values from the long, short, longplural and shortplural keys, rather than using the text, plural, first and firstplural keys. The first use singular text is obtained via:


\genacrfullformat{label}{insert}

instead of from the first key, and the first use plural text is obtained via:


\genplacrfullformat{label}{insert}

instead of from the firstplural key. In both cases, ⟨label⟩ is the entry’s label and ⟨insert⟩ is the insert text provided in the final optional argument of commands like \gls. The default behaviour is to do the long form (or plural long form) followed by ⟨insert⟩ and a space and the short form (or plural short form) in parentheses, where the short form is in the argument of \firstacronymfont. There are also first letter upper case versions:


\Genacrfullformat{label}{insert}

and


\Genplacrfullformat{label}{insert}

By default these perform a protected expansion on their no-case-change equivalents and then use \makefirstuc to convert the first character to upper case. If there are issues caused by this expansion, you will need to redefine those commands to explicitly use commands like \Glsentrylong (which is what the predefined acronym styles, such as long-short, do). Otherwise, you only need to redefine \genacrfullformat and \genplacrfullformat to change the behaviour of \glsgenacfmt. See §13 Acronyms and Other Abbreviations for further details on changing the style of acronyms.

Note that \glsentryfmt (or the formatting given by \defglsentryfmt) is not used by the \glstext-like commands.

As from version 4.16, both the \gls-like and \glstext-like commands use


\glslinkpostsetkeys

after the ⟨options⟩ are set. This macro does nothing by default but can be redefined. (For example, to switch off the hyperlink under certain conditions.) This version also introduces


\glspostlinkhook

which is done after the link text has been displayed and also after the first use flag has been unset (see example 25).

Example 8 (Custom Entry Display in Text)

Suppose you want a glossary of measurements and units, you can use the symbol key to store the unit:

\newglossaryentry{distance}{name=distance,  
description={The length between two points},  
symbol={km}}

and now suppose you want \gls{distance} to produce “distance (km)” on first use, then you can redefine \glsentryfmt as follows:

\renewcommand*{\glsentryfmt}{%  
  \glsgenentryfmt  
  \ifglsused{\glslabel}{}{\space (\glsentrysymbol{\glslabel})}%  
}

(Note that I’ve used \glsentrysymbol rather than \glssymbol to avoid nested hyperlinks.)

Note also that all of the link text will be formatted according to \glstextformat (described earlier). So if you do, say:

\renewcommand{\glstextformat}[1]{\textbf{#1}}  
\renewcommand*{\glsentryfmt}{%  
  \glsgenentryfmt  
  \ifglsused{\glslabel}{}{\space(\glsentrysymbol{\glslabel})}%  
}

then \gls{distance} will produce “distance (km)”.

For a complete document, see the sample file sample-entryfmt.tex.

____________________________

Example 9 (Custom Format for Particular Glossary)

Suppose you have created a new glossary called notation and you want to change the way the entry is displayed on first use so that it includes the symbol, you can do:

\defglsentryfmt[notation]{\glsgenentryfmt  
 \ifglsused{\glslabel}{}{\space  
   (denoted \glsentrysymbol{\glslabel})}}

Now suppose you have defined an entry as follows:

\newglossaryentry{set}{type=notation,  
  name=set,  
  description={A collection of objects},  
  symbol={$S$}  
}

The first time you reference this entry it will be displayed as: “set (denoted S)” (assuming \gls was used).

Alternatively, if you expect all the symbols to be set in math mode, you can do:

\defglsentryfmt[notation]{\glsgenentryfmt  
 \ifglsused{\glslabel}{}{\space  
   (denoted $\glsentrysymbol{\glslabel}$)}}

and define entries like this:

\newglossaryentry{set}{type=notation,  
  name=set,  
  description={A collection of objects},  
  symbol={S}  
}

____________________________

Remember that if you use the symbol key, you need to use a glossary style that displays the symbol, as many of the styles ignore it.

Top

6.4 Enabling and disabling hyperlinks to glossary entries

If you load the hyperref or html packages prior to loading the glossaries package, the \gls-like and \glstext-like commands will automatically have hyperlinks to the relevant glossary entry, unless the hyper option has been switched off (either explicitly or through implicit means, such as via the nohypertypes package option).

You can disable or enable links using:


\glsdisablehyper

and


\glsenablehyper

respectively. The effect can be localised by placing the commands within a group. Note that you should only use \glsenablehyper if the commands \hyperlink and \hypertarget have been defined (for example, by the hyperref package).

You can disable just the first use links using the package option hyperfirst=false. Note that this option only affects the \gls-like commands that recognise the first use flag.

Example 10 (First Use With Hyperlinked Footnote Description)

Suppose I want the first use to have a hyperlink to the description in a footnote instead of hyperlinking to the relevant place in the glossary. First I need to disable the hyperlinks on first use via the package option hyperfirst=false:

\usepackage[hyperfirst=false]{glossaries}

Now I need to redefine \glsentryfmt (see §6.3 Changing the format of the link text):

\renewcommand*{\glsentryfmt}{%  
  \glsgenentryfmt  
  \ifglsused{\glslabel}{}{\footnote{\glsentrydesc{\glslabel}}}%  
}

Now the first use won’t have hyperlinked text, but will be followed by a footnote. See the sample file sample-FnDesc.tex for a complete document.

____________________________

Note that the hyperfirst option applies to all defined glossaries. It may be that you only want to disable the hyperlinks on first use for glossaries that have a different form on first use. This can be achieved by noting that since the entries that require hyperlinking for all instances have identical first and subsequent text, they can be unset via \glsunsetall (see §14 Unsetting and Resetting Entry Flags) so that the hyperfirst option doesn’t get applied.

Example 11 (Suppressing Hyperlinks on First Use Just For Acronyms)

Suppose I want to suppress the hyperlink on first use for acronyms but not for entries in the main glossary. I can load the glossaries package using:

\usepackage[hyperfirst=false,acronym]{glossaries}

Once all glossary entries have been defined I then do:

\glsunsetall[main]

____________________________

For more complex requirements, you might find it easier to switch off all hyperlinks via \glsdisablehyper and put the hyperlinks (where required) within the definition of \glsentryfmt (see §6.3 Changing the format of the link text) via \glshyperlink (see §9 Using Glossary Terms Without Links).

Example 12 (Only Hyperlink in Text Mode Not Math Mode)

This is a bit of a contrived example, but suppose, for some reason, I only want the \gls-like commands to have hyperlinks when used in text mode, but not in math mode. I can do this by adding the glossary to the list of nohypertypes and redefining \glsentryfmt:

\GlsDeclareNoHyperList{main}  
\renewcommand*{\glsentryfmt}{%  
  \ifmmode  
    \glsgenentryfmt  
  \else  
    \glsifhyperon  
    {\glsgenentryfmt}% hyperlink already on  
    {\glshyperlink[\glsgenentryfmt]{\glslabel}}%  
  \fi  
}

Note that this doesn’t affect the \glstext-like commands, which will have the hyperlinks off unless they’re forced on using the plus variant.

See the sample file sample-nomathhyper.tex for a complete document.

____________________________

Example 13 (One Hyper Link Per Entry Per Chapter)

Here’s a more complicated example that will only have the hyperlink on the first time an entry is used per chapter. This doesn’t involve resetting the first use flag. Instead it adds a new key using \glsaddstoragekey (see §4.3.2 Storage Keys) that keeps track of the chapter number that the entry was last used in:

\glsaddstoragekey{chapter}{0}{\glschapnum}

This creates a new user command called \glschapnum that’s analogous to \glsentrytext. The default value for this key is 0. I then define my glossary entries as usual.

Next I redefine the hook \glslinkpostsetkeys (see §6.3 Changing the format of the link text) so that it determines the current chapter number (which is stored in \currentchap using \edef). This value is then compared with the value of the entry’s chapter key that I defined earlier. If they’re the same, this entry has already been used in this chapter so the hyperlink is switched off using xkeyval’s \setkeys command. If the chapter number isn’t the same, then this entry hasn’t been used in the current chapter. The chapter field is updated using \glsfieldxdef (§16.3 Fetching and Updating the Value of a Field) provided the user hasn’t switched off the hyperlink. (This test is performed using \glsifhyperon.

\renewcommand*{\glslinkpostsetkeys}{%  
 \edef\currentchap{\arabic{chapter}}%  
 \ifnum\currentchap=\glschapnum{\glslabel}\relax  
  \setkeys{glslink}{hyper=false}%  
 \else  
  \glsifhyperon{\glsfieldxdef{\glslabel}{chapter}{\currentchap}}{}%  
 \fi  
}

Note that this will be confused if you use \gls etc when the chapter counter is 0. (That is, before the first \chapter.)

See the sample file sample-chap-hyperfirst.tex for a complete document.

____________________________

Top

7. Adding an Entry to the Glossary Without Generating Text

It is possible to add a line to the glossary file without generating any text at that point in the document using:


\glsadd[options]{label}

This is similar to the \glstext-like commands, only it doesn’t produce any text (so therefore, there is no hyper key available in ⟨options⟩ but all the other options that can be used with \glstext-like commands can be passed to \glsadd). For example, to add a page range to the glossary number list for the entry whose label is given by set:

\glsadd[format=(]{set}  
Lots of text about sets spanning many pages.  
\glsadd[format=)]{set}

To add all entries that have been defined, use:


\glsaddall[options]

The optional argument is the same as for \glsadd, except there is also a key types which can be used to specify which glossaries to use. This should be a comma-separated list. For example, if you only want to add all the entries belonging to the list of acronyms (specified by the glossary type \acronymtype) and a list of notation (specified by the glossary type notation) then you can do:

\glsaddall[types={\acronymtype,notation}]

If you are using bib2gls with glossaries-extra, you can’t use \glsaddall. Instead use the selection=all resource option to select all entries in the given bib files.

Note that \glsadd and \glsaddall add the current location to the number list. In the case of \glsaddall, all entries in the glossary will have the same location in the number list. If you want to use \glsaddall, it’s best to suppress the number list with the nonumberlist package option. (See sections 2.3 and 5.)

There is now a variation of \glsaddall that skips any entries that have already been used:


\glsaddallunused[list]

This command uses \glsadd[format=glsignore] which will ignore this location in the number list. The optional argument ⟨list⟩ is a comma-separated list of glossary types. If omitted, it defaults to the list of all defined glossaries.

If you want to use \glsaddallunused, it’s best to place the command at the end of the document to ensure that all the commands you intend to use have already been used. Otherwise you could end up with a spurious comma or dash in the location list.

With glossaries-extra and bib2gls, glsignore indicates an “ignored location” which influences selection but isn’t added to the location list. In this case, if you use selection=all then only those entries that have been explicitly indexed in the document will have location lists. The other entries that were selected as a result of selection=all won’t have location lists.

Base glossaries package only:

\documentclass{article}  
\usepackage{glossaries}  
\makeglossaries  
\newglossaryentry{cat}{name={cat},description={feline}}  
\newglossaryentry{dog}{name={dog},description={canine}}  
\begin{document}  
\gls{cat}.  
\printglossaries  
\glsaddallunused % <- make sure dog is also listed  
\end{document}

Corresponding glossaries-extra and bib2gls document code:

\documentclass{article}  
\usepackage[record]{glossaries-extra}  
\GlsXtrLoadResources[src={entries},selection=all]  
\begin{document}  
\gls{cat}.  
\printunsrtglossaries  
\end{document}

With the file entries.bib:

@entry{cat,name={cat},description={feline}}  
@entry{dog,name={dog},description={canine}}

Example 14 (Dual Entries)

The example file sample-dual.tex makes use of \glsadd to allow for an entry that should appear both in the main glossary and in the list of acronyms. This example sets up the list of acronyms using the acronym package option:

\usepackage[acronym]{glossaries}

A new command is then defined to make it easier to define dual entries:

\newcommand*{\newdualentry}[5][]{%  
  \newglossaryentry{main-#2}{name={#4},%  
  text={#3\glsadd{#2}},%  
  description={#5},%  
  #1  
  }%  
  \newacronym{#2}{#3\glsadd{main-#2}}{#4}%  
}

This has the following syntax:

\newdualentry[options]{label}{abbrv}{long}{description}
You can then define a new dual entry:
\newdualentry{svm}% label  
  {SVM}% abbreviation  
  {support vector machine}% long form  
  {Statistical pattern recognition technique}% description

Now you can reference the acronym with \gls{svm} or you can reference the entry in the main glossary with \gls{main-svm}.

Note that with bib2gls, there are special dual entry types that implement this behaviour. That is, if an entry is referenced then its corresponding dual entry will automatically be selected as well. So there is less need for \glsadd with bib2gls. (Although it can still be useful, as shown in Option 6.)

____________________________

Top

8. Cross-Referencing Entries

You must use \makeglossaries (Options 2 or 3) or \makenoidxglossaries (Option 1) before defining any terms that cross-reference entries. If any of the terms that you have cross-referenced don’t appear in the glossary, check that you have put \makeglossaries/\makenoidxglossaries before all entry definitions. The glossaries-extra package provides better cross-reference handling.

There are several ways of cross-referencing entries in the glossary:

  1. You can use commands such as \gls in the entries description. For example:
    \newglossaryentry{apple}{name=apple,  
    description={firm, round fruit. See also \gls{pear}}}

    Note that with this method, if you don’t use the cross-referenced term in the main part of the document, you will need two runs of makeglossaries:

    
         
    pdflatex filename
    makeglossaries filename
    pdflatex filename
    makeglossaries filename
    pdflatex filename

  2. As described in §4 Defining Glossary Entries, you can use the see key when you define the entry. For example:
    \newglossaryentry{MaclaurinSeries}{name={Maclaurin  
    series},  
    description={Series expansion},  
    see={TaylorsTheorem}}

    Note that in this case, the entry with the see key will automatically be added to the glossary, but the cross-referenced entry won’t. You therefore need to ensure that you use the cross-referenced term with the commands described in §6 Links to Glossary Entries or §7 Adding an Entry to the Glossary Without Generating Text.

    The “see” tag is produce using \seename, but can be overridden in specific instances using square brackets at the start of the see value. For example:

    \newglossaryentry{MaclaurinSeries}{name={Maclaurin  
    series},  
    description={Series expansion},  
    see=[see also]{TaylorsTheorem}}

    Take care if you want to use the optional argument of commands such as \newacronym or \newterm as the value will need to be grouped. For example:

    \newterm{seal}  
    \newterm[see={[see also]seal}]{sea lion}

    Similarly if the value contains a list. For example:

    \glossaryentry{lemon}{  
      name={lemon},  
      description={Yellow citrus fruit}  
    }  
    \glossaryentry{lime}  
    {  
      name={lime},  
      description={Green citrus fruit}  
    }  
    \glossaryentry{citrus}  
    {  
      name={citrus},  
      description={Plant in the Rutaceae family},  
      see={lemon,lime}  
    }

  3. After you have defined the entry, use


    \glssee[tag]{label}{xr label list}

    where ⟨xr label list⟩ is a comma-separated list of entry labels to be cross-referenced, ⟨label⟩ is the label of the entry doing the cross-referencing and ⟨tag⟩ is the “see” tag. (The default value of ⟨tag⟩ is \seename.) For example:

    \glssee[see also]{series}{FourierSeries,TaylorsTheorem}

    Note that this automatically adds the entry given by ⟨label⟩ to the glossary but doesn’t add the cross-referenced entries (specified by ⟨xr label list⟩) to the glossary.

In both cases 2 and 3 above, the cross-referenced information appears in the number list, whereas in case 1, the cross-referenced information appears in the description. (See the sample-crossref.tex example file that comes with this package.) This means that in cases 2 and 3, the cross-referencing information won’t appear if you have suppressed the number list. In this case, you will need to activate the number list for the given entries using nonumberlist=false. Alternatively, if you just use the see key instead of \glssee, you can automatically activate the number list using the seeautonumberlist package option.

Top

8.1 Customising Cross-reference Text

When you use either the see key or the command \glssee, the cross-referencing information will be typeset in the glossary according to:


\glsseeformat[tag]{label-list}{location}

The default definition of \glsseeformat is:
\emph{tag} \glsseelist{label-list}
Note that the location is always ignored.8.1 For example, if you want the tag to appear in bold, you can do:8.2

\renewcommand*{\glsseeformat}[3][\seename]{\textbf{#1}  
 \glsseelist{#2}}

The list of labels is dealt with by \glsseelist, which iterates through the list and typesets each entry in the label. The entries are separated by


\glsseesep

or (for the last pair)


\glsseelastsep

These default to “,\space” and “\space\andname\space” respectively. The list entry text is displayed using:


\glsseeitemformat{label}

This defaults to \glsentrytext{label}.8.3 For example, to make the cross-referenced list use small caps:

\renewcommand{\glsseeitemformat}[1]{%  
  \textsc{\glsentrytext{#1}}}

You can use \glsseeformat and \glsseelist in the main body of the text, but they won’t automatically add the cross-referenced entries to the glossary. If you want them added with that location, you can do:

Some information (see also  
\glsseelist{FourierSeries,TaylorsTheorem}%  
\glsadd{FourierSeries}\glsadd{TaylorsTheorem}).

Top

9. Using Glossary Terms Without Links

The commands described in this section display entry details without adding any information to the glossary. They don’t use \glstextformat, they don’t have any optional arguments, they don’t affect the first use flag and, apart from \glshyperlink, they don’t produce hyperlinks.

Commands that aren’t expandable will be ignored by PDF bookmarks, so you will need to provide an alternative via hyperref’s \texorpdfstring if you want to use them in sectioning commands. (This isn’t specific to the glossaries package.) See the hyperref documentation for further details. All the commands that convert the first letter to upper case aren’t expandable. The other commands depend on whether their corresponding keys were assigned non-expandable values.

If you want to title case a field, you can use:


\glsentrytitlecase{label}{field}

where ⟨label⟩ is the label identifying the glossary entry, ⟨field⟩ is the field label (see table 4.1). For example:

\glsentrytitlecase{sample}{desc}

(If you want title-casing in your glossary style, you might want to investigate the glossaries-extra package.) This command will trigger an error if the entry is undefined.

Note that this command has the same limitations as \makefirstuc which is used by commands like \Gls and \Glsentryname to upper-case the first letter (see the notes about \Gls in §6.1 The \gls-Like Commands (First Use Flag Queried)).


\glsentryname{label}


\Glsentryname{label}

These commands display the name of the glossary entry given by ⟨label⟩, as specified by the name key. \Glsentryname makes the first letter upper case. Neither of these commands check for the existence of ⟨label⟩. The first form \glsentryname is expandable (unless the name contains unexpandable commands). Note that this may be different from the values of the text or first keys. In general it’s better to use \glsentrytext or \glsentryfirst instead of \glsentryname.

In general it’s best to avoid \Glsentryname with abbreviations. Instead, consider using \Glsentrylong, \Glsentryshort or \Glsentryfull.


\glossentryname{label}

This is like \glsnamefont{\glsentryname{label}} but also checks for the existence of ⟨label⟩. This command is not expandable. It’s used in the predefined glossary styles, so if you want to change the way the name is formatted in the glossary, you can redefine \glsnamefont to use the required fonts. For example:

\renewcommand*{\glsnamefont}[1]{\textmd{\sffamily #1}}


\Glossentryname{label}

This is like \glossentryname but makes the first letter of the name upper case.


\glsentrytext{label}


\Glsentrytext{label}

These commands display the subsequent use text for the glossary entry given by ⟨label⟩, as specified by the text key. \Glsentrytext makes the first letter upper case. The first form is expandable (unless the text contains unexpandable commands). The second form is not expandable. Neither checks for the existence of ⟨label⟩.


\glsentryplural{label}


\Glsentryplural{label}

These commands display the subsequent use plural text for the glossary entry given by ⟨label⟩, as specified by the plural key. \Glsentryplural makes the first letter upper case. The first form is expandable (unless the value of that key contains unexpandable commands). The second form is not expandable. Neither checks for the existence of ⟨label⟩.


\glsentryfirst{label}


\Glsentryfirst{label}

These commands display the first use text for the glossary entry given by ⟨label⟩, as specified by the first key. \Glsentryfirst makes the first letter upper case. The first form is expandable (unless the value of that key contains unexpandable commands). The second form is not expandable. Neither checks for the existence of ⟨label⟩.


\glsentryfirstplural{label}


\Glsentryfirstplural{label}

These commands display the plural form of the first use text for the glossary entry given by ⟨label⟩, as specified by the firstplural key. \Glsentryfirstplural makes the first letter upper case. The first form is expandable (unless the value of that key contains unexpandable commands). The second form is not expandable. Neither checks for the existence of ⟨label⟩.


\glsentrydesc{label}


\Glsentrydesc{label}

These commands display the description for the glossary entry given by ⟨label⟩.

\Glsentrydesc makes the first letter upper case. The first form is expandable (unless the value of that key contains unexpandable commands). The second form is not expandable. Neither checks for the existence of ⟨label⟩.


\glossentrydesc{label}

This is like \glsentrydesc{label} but also checks for the existence of ⟨label⟩. This command is not expandable. It’s used in the predefined glossary styles to display the description.


\Glossentrydesc{label}

This is like \glossentrydesc but converts the first letter to upper case. This command is not expandable.


\glsentrydescplural{label}


\Glsentrydescplural{label}

These commands display the plural description for the glossary entry given by ⟨label⟩. \Glsentrydescplural makes the first letter upper case. The first form is expandable (unless the value of that key contains unexpandable commands). The second form is not expandable. Neither checks for the existence of ⟨label⟩.


\glsentrysymbol{label}


\Glsentrysymbol{label}

These commands display the symbol for the glossary entry given by ⟨label⟩.

\Glsentrysymbol makes the first letter upper case. The first form is expandable (unless the value of that key contains unexpandable commands). The second form is not expandable. Neither checks for the existence of ⟨label⟩.


\glsletentryfield{cs}{label}{field}

This command doesn’t display anything. It merely fetches the value associated with the given field (where the available field names are listed in table 4.1) and stores the result in the control sequence ⟨cs⟩. For example, to store the description for the entry whose label is “apple” in the control sequence \tmp:

\glsletentryfield{\tmp}{apple}{desc}


\glossentrysymbol{label}

This is like \glsentrysymbol{label} but also checks for the existence of ⟨label⟩. This command is not expandable. It’s used in some of the predefined glossary styles to display the symbol.


\Glossentrysymbol{label}

This is like \glossentrysymbol but converts the first letter to upper case. This command is not expandable.


\glsentrysymbolplural{label}


\Glsentrysymbolplural{label}

These commands display the plural symbol for the glossary entry given by ⟨label⟩. \Glsentrysymbolplural makes the first letter upper case. The first form is expandable (unless the value of that key contains unexpandable commands). The second form is not expandable. Neither checks for the existence of ⟨label⟩.


\glsentryuseri{label}


\Glsentryuseri{label}


\glsentryuserii{label}


\Glsentryuserii{label}


\glsentryuseriii{label}


\Glsentryuseriii{label}


\glsentryuseriv{label}


\Glsentryuseriv{label}


\glsentryuserv{label}


\Glsentryuserv{label}


\glsentryuservi{label}


\Glsentryuservi{label}

These commands display the value of the user keys for the glossary entry given by ⟨label⟩. The lower case forms are expandable (unless the value of the key contains unexpandable commands). The commands beginning with an upper case letter convert the first letter of the required value to upper case and are not expandable. None of these commands check for the existence of ⟨label⟩.


\glshyperlink[link text]{label}

This command provides a hyperlink to the glossary entry given by ⟨labelbut does not add any information to the glossary file. The link text is given by \glsentrytext{label} by default9.1, but can be overridden using the optional argument. Note that the hyperlink will be suppressed if you have used \glsdisablehyper or if you haven’t loaded the hyperref package.

If you use \glshyperlink, you need to ensure that the relevant entry has been added to the glossary using any of the commands described in §6 Links to Glossary Entries or §7 Adding an Entry to the Glossary Without Generating Text otherwise you will end up with an undefined link.

The next two commands are only available with Option 1 or with the savenumberlist package option:


\glsentrynumberlist{label}


\glsdisplaynumberlist{label}

Both display the number list for the entry given by ⟨label⟩. When used with Option 1 a rerun is required to ensure this list is up-to-date, when used with Options 2 or 3 a run of makeglossaries (or makeindex/xindy) followed by one or two runs of LaTeX is required.

The first command, \glsentrynumberlist, simply displays the number list as is. The second command, \glsdisplaynumberlist, formats the list using:


\glsnumlistsep

as the separator between all but the last two elements and


\glsnumlistlastsep

between the final two elements. The defaults are ,␣ and ␣\&␣, respectively.

\glsdisplaynumberlist is fairly experimental. It works with Option 1, but for Options 2 or 3 it only works when the default counter format is used (that is, when the format key is set to glsnumberformat). This command will only work with hyperref if you choose Option 1. If you try using this command with Options 2 or 3 and hyperref, \glsentrynumberlist will be used instead.

For further information see “Displaying entry details without adding information to the glossary” in the documented code (glossaries-code.pdf).

Top

10. Displaying a glossary

All defined glossaries may be displayed using:

Option 1:
(Must be used with \makenoidxglossaries in the preamble.)


\printnoidxglossaries

Options 2 and 3:
(Must be used with \makeglossaries in the preamble.)


\printglossaries

These commands will display all the glossaries in the order in which they were defined.

Note that, in the case of Options 2 and 3, no glossaries will appear until you have either used the Perl script makeglossaries or Lua script makeglossaries-lite or have directly used makeindex or xindy (as described in §1.4 Generating the Associated Glossary Files).

While the external files are missing, \printglossary will just do \null for each missing glossary to assist dictionary style documents that just use \glsaddall without inserting any text. This use of \null ensures that all indexing information is written before the final page is shipped out. Once the external files are present \null will no longer be used. This can cause a spurious blank page on the first LaTeX run before the glossary files have been created. Once these files are present, \null will no longer be used and so shouldn’t cause interference for the final document.

If you use glossaries-extra, it will insert a heading and boilerplate text when the external files are missing. The extension package also provides \printunsrtglossaries as an alternative. See the glossaries-extra manual for further details.

If the glossary still does not appear after you re-LaTeX your document, check the makeindex/xindy log files to see if there is a problem. With Option 1, you just need two LaTeX runs to make the glossaries appear, but you may need further runs to make the number lists up-to-date.

An individual glossary can be displayed using:

Option 1:


\printnoidxglossary[options]

(Must be used with \makenoidxglossaries in the preamble.)

Options 2 and 3:


\printglossary[options]

(Must be used with \makeglossaries in the preamble.)

where ⟨options⟩ is a ⟨key⟩=⟨value⟩ list of options. (Again, when the associated external file is missing, \null is inserted into the document.)

The following keys are available:

type
The value of this key specifies which glossary to print. If omitted, the default glossary is assumed. For example, to print the list of acronyms:
\printglossary[type=\acronymtype]

Note that you can’t display an ignored glossary, so don’t try setting type to the name of a glossary that was defined using \newignoredglossary, described in §12 Defining New Glossaries. (You can display an ignored glossary with \printunsrtglossary provided by glossaries-extra.)

title
This is the glossary’s title (overriding the title specified when the glossary was defined).
toctitle
This is the title to use for the table of contents (if the toc package option has been used). It may also be used for the page header, depending on the page style. If omitted, the value of title is used.
style
This specifies which glossary style to use for this glossary, overriding the effect of the style package option or \glossarystyle.
numberedsection
This specifies whether to use a numbered section for this glossary, overriding the effect of the numberedsection package option. This key has the same syntax as the numberedsection package option, described in §2.2 Sectioning, Headings and TOC Options.
nonumberlist
This is a boolean key. If true (nonumberlist=true) the numberlist is suppressed for this glossary. If false (nonumberlist=false) the numberlist is displayed for this glossary.
nogroupskip
This is a boolean key. If true the vertical gap between groups is suppressed for this glossary.
nopostdot
This is a boolean key. If true the full stop after the description is suppressed for this glossary.
entrycounter
This is a boolean key. Behaves similar to the package option of the same name. The corresponding package option must be used to make \glsrefentry work correctly.
subentrycounter
This is a boolean key. Behaves similar to the package option of the same name. If you want to set both entrycounter and subentrycounter, make sure you specify entrycounter first. The corresponding package option must be used to make \glsrefentry work correctly.
sort
This key is only available for Option 1. Possible values are: word (word order), letter (letter order), standard (word or letter ordering taken from the order package option), use (order of use), def (order of definition) nocase (case-insensitive) or case (case-sensitive). Note that the word and letter comparisons (that is, everything other than sort=use and sort=def) just use a simple character code comparison. For a locale-sensitive sort, you must use either xindy (Option 3) or bib2gls (Option 4). Note that bib2gls provides many other sort options.

If you use the use or def values make sure that you select a glossary style that doesn’t have a visual indicator between groups, as the grouping no longer makes sense. Consider using the nogroupskip option.

The word and letter order sort methods use datatool’s \dtlwordindexcompare and \dtlletterindexcompare handlers. The case-insensitive sort method uses datatool’s \dtlicompare handler. The case-sensitive sort method uses datatool’s \dtlcompare handler. See the datatool documentation for further details.

If you don’t get an error with sort=use and sort=def but you do get an error with one of the other sort options, then you probably need to use the sanitizesort=true package option or make sure none of the entries have fragile commands in their sort field.

label={label}
This key is only available with glossaries-extra and labels the glossary with \label{label}. This is an alternative to the package option numberedsection=autolabel
target
This is a boolean key only available with glossaries-extra, which can be used to switch off the automatic hypertarget for each entry. (This refers to the target used by commands like \gls and \glslink.)

This option is useful with \printunsrtglossary as it allows the same list (of sub-list) of entries to be displayed multiple times without causing duplicate hypertarget names.

prefix={prefix}
This key is only available with glossaries-extra and provides another way of avoiding duplicate hypertarget names is to use a different prefix for those names. This locally redefines \glolinkprefix but note this will also affect the target for any entry referenced within the glossary with commands like \gls, \glslink or \glshypertarget.
targetnameprefix={prefix}
This key is only available with glossaries-extra. This is similar to the prefix option, but it alters the prefix of the hypertarget anchors without changing \glslinkprefix (so it won’t change the hyperlinks for any entries referenced in the glossary).

By default, the glossary is started either by \chapter* or by \section*, depending on whether or not \chapter is defined. This can be overridden by the section package option or the \setglossarysection command. Numbered sectional units can be obtained using the numberedsection package option. Each glossary sets the page header via the command


\glsglossarymark{title}

If this mechanism is unsuitable for your chosen class file or page style package, you will need to redefine \glsglossarymark. Further information about these options and commands is given in §2.2 Sectioning, Headings and TOC Options.

Information can be added to the start of the glossary (after the title and before the main body of the glossary) by redefining


\glossarypreamble

For example:

\renewcommand{\glossarypreamble}{Numbers in italic  
indicate primary definitions.}

This needs to be done before the glossary is displayed.

If you want a different preamble per glossary you can use


\setglossarypreamble[type]{preamble text}

If ⟨type⟩ is omitted, \glsdefaulttype is used.

For example:

\setglossarypreamble{Numbers in italic  
indicate primary definitions.}

This will print the given preamble text for the main glossary, but not have any preamble text for any other glossaries.

There is an analogous command to \glossarypreamble called


\glossarypostamble

which is placed at the end of each glossary.

Example 15 (Switch to Two Column Mode for Glossary)

Suppose you are using the superheaderborder style10.1, and you want the glossary to be in two columns, but after the glossary you want to switch back to one column mode, you could do:

\renewcommand*{\glossarysection}[2][]{%  
  \twocolumn[{\chapter*{#2}}]%  
  \setlength\glsdescwidth{0.6\linewidth}%  
  \glsglossarymark{\glossarytoctitle}%  
}  
\renewcommand*{\glossarypostamble}{\onecolumn}

____________________________

Within each glossary, each entry name is formatted according to


\glsnamefont{name}

which takes one argument: the entry name. This command is always used regardless of the glossary style. By default, \glsnamefont simply displays its argument in whatever the surrounding font happens to be. This means that in the list-like glossary styles (defined in the glossary-list style file) the name will appear in bold, since the name is placed in the optional argument of \item, whereas in the tabular styles (defined in the glossary-long and glossary-super style files) the name will appear in the normal font. The hierarchical glossary styles (defined in the glossary-tree style file) also set the name in bold.

If you want to change the font for the description, or if you only want to change the name font for some types of entries but not others, you might want to consider using the glossaries-extra package.

Example 16 (Changing the Font Used to Display Entry Names in the Glossary)

Suppose you want all the entry names to appear in medium weight small caps in your glossaries, then you can do:

\renewcommand{\glsnamefont}[1]{\textsc{\mdseries #1}}

____________________________

Top

11. Xindy (Option 3)

If you want to use xindy to sort the glossary, you must use the package option xindy:

\usepackage[xindy]{glossaries}

This ensures that the glossary information is written in xindy syntax.

§1.4 Generating the Associated Glossary Files covers how to use the external indexing application, and §5.2 Locations covers the issues involved in the location syntax. This section covers the commands provided by the glossaries package that allow you to adjust the xindy style file (xdy) and parameters.

To assist writing information to the xindy style file, the glossaries package provides the following commands:


\glsopenbrace


\glsclosebrace

which produce an open and closing brace. (This is needed because \{ and \} don’t expand to a simple brace character when written to a file.) Similarly, you can write a percent character using:


\glspercentchar

and a tilde character using:


\glstildechar

For example, a newline character is specified in a xindy style file using ~n so you can use \glstildechar n to write this correctly (or you can do \string~n). A backslash can be written to a file using


\glsbackslash

In addition, if you are using a package that makes the double quote character active (e.g. ngerman) you can use:


\glsquote{text}

which will produce "text". Alternatively, you can use \string" to write the double-quote character. This document assumes that the double quote character has not been made active, so the examples just use " for clarity.

If you want greater control over the xindy style file than is available through the LaTeX commands provided by the glossaries package, you will need to edit the xindy style file. In which case, you must use \noist to prevent the style file from being overwritten by the glossaries package. For additional information about xindy, read the xindy documentation. I’m sorry I can’t provide any assistance with writing xindy style files. If you need help, I recommend you ask on the xindy mailing list (http://xindy.sourceforge.net/mailing-list.html).

Top

11.1 Language and Encodings

When you use xindy, you need to specify the language and encoding used (unless you have written your own custom xindy style file that defines the relevant alphabet and sort rules). If you use makeglossaries, this information is obtained from the document’s auxiliary (aux) file. The makeglossaries script attempts to find the root language given your document settings, but in the event that it gets it wrong or if xindy doesn’t support that language, then you can specify the required language using:


\GlsSetXdyLanguage[glossary type]{language}

where ⟨language⟩ is the name of the language. The optional argument can be used if you have multiple glossaries in different languages. If ⟨glossary type⟩ is omitted, it will be applied to all glossaries, otherwise the language setting will only be applied to the glossary given by ⟨glossary type⟩.

If the inputenc package is used, the encoding will be obtained from the value of \inputencodingname. Alternatively, you can specify the encoding using:


\GlsSetXdyCodePage{code}

where ⟨code⟩ is the name of the encoding. For example:

\GlsSetXdyCodePage{utf8}

Note that you can also specify the language and encoding using the package option xindy={language=lang,codepage=code}. For example:

\usepackage[xindy={language=english,codepage=utf8}]{glossaries}

If you write your own custom xindy style file that includes the language settings, you need to set the language to nothing:

\GlsSetXdyLanguage{}

(and remember to use \noist to prevent the style file from being overwritten).

The commands \GlsSetXdyLanguage and \GlsSetXdyCodePage have no effect if you don’t use makeglossaries. If you call xindy without makeglossaries you need to remember to set the language and encoding using the -L and -C switches.

Top

11.2 Locations and Number lists

If you use xindy, the glossaries package needs to know which counters you will be using in the number list in order to correctly format the xindy style file. Counters specified using the counter package option or the ⟨counter⟩ option of \newglossary are automatically taken care of, but if you plan to use a different counter in the counter key for commands like \glslink, then you need to identify these counters before \makeglossaries using:


\GlsAddXdyCounters{counter list}

where ⟨counter list⟩ is a comma-separated list of counter names.

The most likely attributes used in the format key (textrm, hyperrm etc) are automatically added to the xindy style file, but if you want to use another attribute, you need to add it using:


\GlsAddXdyAttribute{name}

where ⟨name⟩ is the name of the attribute, as used in the format key.

Take care if you have multiple instances of the same location with different formats. The duplicate locations will be discarded according to the order in which the attributes are listed. Consider defining semantic commands to use for primary references. For example:

\newcommand*{\primary}[1]{\textbf{#1}}  
\GlsAddXdyAttribute{primary}

Then in the document:

A \gls[format=primary]{duck} is an aquatic bird.  
There are lots of different types of \gls{duck}.

This will give the format=primary instance preference over the next use that doesn’t use the format key.

Example 17 (Custom Font for Displaying a Location)

Suppose I want a bold, italic, hyperlinked location. I first need to define a command that will do this:

\newcommand*{\hyperbfit}[1]{\textit{\hyperbf{#1}}}

but with xindy, I also need to add this as an allowed attribute:

\GlsAddXdyAttribute{hyperbfit}

Now I can use it in the optional argument of commands like \gls:

Here is a \gls[format=hyperbfit]{sample} entry.

(where sample is the label of the required entry).

____________________________

Note that \GlsAddXdyAttribute has no effect if \noist is used or if \makeglossaries is omitted. \GlsAddXdyAttribute must be used before \makeglossaries. Additionally, \GlsAddXdyCounters must come before \GlsAddXdyAttribute.

If the location numbers include formatting commands, then you need to add a location style in the appropriate format using


\GlsAddXdyLocation[prefix-location]{name}{definition}

where ⟨name⟩ is the name of the format and ⟨definition⟩ is the xindy definition. The optional argument ⟨prefix-location⟩ is needed if \theHcounter⟩ either isn’t defined or is different from \thecounter⟩. Be sure to also read §5.2 Locations for some issues that you may encounter.

Note that \GlsAddXdyLocation has no effect if \noist is used or if \makeglossaries is omitted. \GlsAddXdyLocation must be used before \makeglossaries.

Example 18 (Custom Numbering System for Locations)

Suppose I decide to use a somewhat eccentric numbering system for sections where I redefine \thesection as follows:

\renewcommand*{\thesection}{[\thechapter]\arabic{section}}

If I haven’t done counter=section in the package option, I need to specify that the counter will be used as a location number:

\GlsAddXdyCounters{section}

Next I need to add the location style (\thechapter is assumed to be the standard \arabic{chapter}):

\GlsAddXdyLocation{section}{:sep "[" "arabic-numbers" :sep "]"  
  "arabic-numbers"  
}

Note that if I have further decided to use the hyperref package and want to redefine \theHsection as:

\renewcommand*{\theHsection}{\thepart.\thesection}  
\renewcommand*{\thepart}{\Roman{part}}

then I need to modify the \GlsAddXdyLocation code above to:

\GlsAddXdyLocation["roman-numbers-uppercase"]{section}{:sep "["  
  "arabic-numbers" :sep "]" "arabic-numbers"  
}

Since \Roman will result in an empty string if the counter is zero, it’s a good idea to add an extra location to catch this:

\GlsAddXdyLocation{zero.section}{:sep "["  
  "arabic-numbers" :sep "]" "arabic-numbers"  
}

This example is illustrated in the sample file samplexdy2.tex.

____________________________

Example 19 (Locations as Dice)

Suppose I want a rather eccentric page numbering system that’s represented by the number of dots on dice. The stix package provides \dicei, …, \dicevi that represent the six sides of a die. I can define a command that takes a number as its argument. If the number is less than seven, the appropriate \dicen⟩ command is used otherwise it does \dicevi the required number of times with the leftover in a final \dicen⟩. For example, the number 16 is represented by \dicevi\dicevi\diceiv (6 + 6 + 4 = 16). I’ve called this command \tallynum to match the example given earlier in §5.2 Locations:

\newrobustcmd{\tallynum}[1]{%  
 \ifnum\number#1<7  
  $\csname dice\romannumeral#1\endcsname$%  
 \else  
  $\dicevi$%  
  \expandafter\tallynum\expandafter{\numexpr#1-6}%  
 \fi  
}

Here’s the counter command:

newcommand{\tally}[1]{\tallynum{\arabic{#1}}}

The page counter representation (\thepage) needs to be changed to use this command:

\renewcommand*{\thepage}{\tally{page}}

The \tally command expands to \tallynum {number} so this needs a location class that matches this format:

\GlsAddXdyLocation{tally}{%  
 :sep "\string\tallynum\space\glsopenbrace"  
 "arabic-numbers"  
 :sep "\glsclosebrace"  
}

The space between \tallynum and {number} is significant to xindy so \space is required.

Note that \GlsAddXdyLocation{name}{definition} will define commands in the form:


\glsXcounterXname{Hprefix}{location}

for each counter that has been identified either by the counter package option, the ⟨counter⟩ option for \newglossary or in the argument of \GlsAddXdyCounters. The first argument ⟨Hprefix⟩ is only relevant when used with the hyperref package and indicates that \theHcounter⟩ is given by \Hprefix.\thecounter⟩.

The sample file samplexdy.tex, which comes with the glossaries package, uses the default page counter for locations, and it uses the default \glsnumberformat and a custom \hyperbfit format. A new xindy location called tallynum, as illustrated above, is defined to make the page numbers appear as dice. In order for the location numbers to hyperlink to the relevant pages, I need to redefine the necessary \glsXcounterXformat⟩ commands:

\renewcommand{\glsXpageXglsnumberformat}[2]{%  
 \linkpagenumber#2%  
}  
\renewcommand{\glsXpageXhyperbfit}[2]{%  
 \textbf{\em\linkpagenumber#2}%  
}  
\newcommand{\linkpagenumber}[2]{\hyperlink{page.#2}{#1{#2}}}

Note that the second argument of \glsXpageXglsnumberformat is in the format \tallynum{n} so the line

 \linkpagenumber#2%

does

 \linkpagenumber\tallynum{number}
so \tallynum is the first argument of \linkpagenumber and ⟨number⟩ is the second argument.

____________________________

This method is very sensitive to the internal definition of the location command.

Example 20 (Locations as Words not Digits)

Suppose I want the page numbers written as words rather than digits and I use the fmtcount package to do this. I can redefine \thepage as follows:

\renewcommand*{\thepage}{\Numberstring{page}}

This used to get expanded to \protect \Numberstringnum {n} where ⟨n⟩ is the Arabic page number. This means that I needed to define a new location with the form:

\GlsAddXdyLocation{Numberstring}{:sep "\string\protect\space  
  \string\Numberstringnum\space\glsopenbrace"  
  "arabic-numbers" :sep "\glsclosebrace"}

and if I’d used the \linkpagenumber command from the previous example, it would need three arguments (the first being \protect):

\newcommand{\linkpagenumber}[3]{\hyperlink{page.#3}{#1#2{#3}}}

The internal definition of \Numberstring has since changed so that it now expands to \Numberstringnum  {n} (no \protect). This means that the location class definition must be changed to:

\GlsAddXdyLocation{Numberstring}{% no \protect now!  
  :sep "\string\Numberstringnum\space\glsopenbrace"  
  "arabic-numbers" :sep "\glsclosebrace"}

and \linkpagenumber goes back to only two arguments:

\newcommand{\linkpagenumber}[2]{\hyperlink{page.#2}{#1{#2}}}

The other change is that \Numberstring uses

\the\value{counter}
instead of
\expandafter\the\csname c@counter\endcsname
so it hides \c@page from the location escaping mechanism (see §5.2 Locations). This means that the page number may be incorrect if the indexing occurs during the output routine.

A more recent change to fmtcount (v3.03) now puts three instances of \expandafter before \the\value which no longer hides \c@page from the location escaping mechanism, so the page numbers should once more be correct. Further changes to the fmtcount package may cause a problem again.

When dealing with custom formats where the internal definitions are outside of your control and liable to change, it’s best to provide a wrapper command.

Instead of directly using \Numberstring in the definition of \thepage, I can provide a custom command in the same form as the earlier \tally command:

\newcommand{\customfmt}[1]{\customfmtnum{\arabic{#1}}}  
\newrobustcmd{\customfmtnum}[1]{\Numberstringnum{#1}}

This ensures that the location will always be written to the indexing file in the form:

:locref "{}{\\customfmtnum {n}}"
So the location class can be defined as:
\GlsAddXdyLocation{customfmt}{  
 :sep "\string\customfmtnum\space\glsopenbrace"  
 "arabic-numbers"  
 :sep "\glsclosebrace"}

The sample file samplexdy3.tex illustrates this.

____________________________

In the number list, the locations are sorted according to the list of provided location classes. The default ordering is: roman-page-numbers (i, ii, …), arabic-page-numbers (1, 2, …), arabic-section-numbers (for example, 1.1 if the compositor is a full stop or 1-1 if the compositor is a hyphen11.1), alpha-page-numbers (a, b, …), Roman-page-numbers (I, II, …), Alpha-page-numbers (A, B, …), Appendix-page-numbers (for example, A.1 if the Alpha compositor is a full stop or A-1 if the Alpha compositor is a hyphen11.2), user defined location names (as specified by \GlsAddXdyLocation in the order in which they were defined), and finally see (cross-referenced entries).11.3 This ordering can be changed using:


\GlsSetXdyLocationClassOrder{location names}

where each location name is delimited by double quote marks and separated by white space. For example:

\GlsSetXdyLocationClassOrder{  
  "arabic-page-numbers"  
  "arabic-section-numbers"  
  "roman-page-numbers"  
  "Roman-page-numbers"  
  "alpha-page-numbers"  
  "Alpha-page-numbers"  
  "Appendix-page-numbers"  
  "see"  
}

(Remember to add "seealso" if you’re using glossaries-extra.)

Note that \GlsSetXdyLocationClassOrder has no effect if \noist is used or if \makeglossaries is omitted. \GlsSetXdyLocationClassOrder must be used before \makeglossaries.

If a number list consists of a sequence of consecutive numbers, the range will be concatenated. The number of consecutive locations that causes a range formation defaults to 2, but can be changed using:


\GlsSetXdyMinRangeLength{n}

For example:

\GlsSetXdyMinRangeLength{3}

The argument may also be the keyword none, to indicate that there should be no range formations. See the xindy manual for further details on range formations.

Note that \GlsSetXdyMinRangeLength has no effect if \noist is used or if \makeglossaries is omitted. \GlsSetXdyMinRangeLength must be used before \makeglossaries.

See also §5.3 Range Formations.

Top

11.3 Glossary Groups

The glossary is divided into groups according to the first letter of the sort key. The glossaries package also adds a number group by default, unless you suppress it in the xindy package option. For example:

\usepackage[xindy={glsnumbers=false}]{glossaries}

Any entry that doesn’t go in one of the letter groups or the number group is placed in the default group. If you want xindy to sort the number group numerically (rather than by a string sort) then you need to use xindy’s numeric-sort module:

\GlsAddXdyStyle{numeric-sort}

If you don’t use glsnumbers=false, the default behaviour is to locate the number group before the “A” letter group. If you are not using a Roman alphabet, you need to change this using:


\GlsSetXdyFirstLetterAfterDigits{letter}

where ⟨letter⟩ is the first letter of your alphabet. Take care if you’re using inputenc as non-ASCII characters are actually active characters that expand. (This isn’t a problem with the native UTF-8 engines and fontspec.) The starred form will sanitize the argument to prevent expansion. Alternatively you can use:


\GlsSetXdyNumberGroupOrder{relative location}

to change the default

:before \string"letter\string"
to ⟨relative location⟩. For example:
\GlsSetXdyNumberGroupOrder{:after \string"Z\string"}

will put the number group after the “Z” letter group. Again take care of active characters. There’s a starred version that sanitizes the argument (so don’t use \string in it).

\GlsSetXdyNumberGroupOrder*{:after "Ö"}

Note that these commands have no effect if \noist is used or if \makeglossaries is omitted. \GlsSetXdyFirstLetterAfterDigits must be used before \makeglossaries.

Top

12. Defining New Glossaries

A new glossary can be defined using:


\newglossary[log-ext]{name}{in-ext}{out-ext} {title}[counter]

where ⟨name⟩ is the label to assign to this glossary. The arguments ⟨in-ext⟩ and ⟨out-ext⟩ specify the extensions to give to the input and output files for that glossary, ⟨title⟩ is the default title for this new glossary and the final optional argument ⟨counter⟩ specifies which counter to use for the associated number lists (see also §5 Number lists). The first optional argument specifies the extension for the makeindex (Option 2) or xindy (Option 3) transcript file (this information is only used by makeglossaries which picks up the information from the auxiliary file). If you use Option 1, the ⟨log-ext⟩, ⟨in-ext⟩ and ⟨out-ext⟩ arguments are ignored.

The glossary label ⟨name⟩ must not contain any active characters. It’s generally best to stick with just characters that have category code 11 (typically the non-extended Latin characters for standard LaTeX).

There is also a starred version (new to v4.08):


\newglossary*{name}{title}[counter]

which is equivalent to

\newglossary[name-glg]{name}{name-gls}{name-glo}{title}[counter]
or you can also use:


\altnewglossary{name}{tag}{title}[counter]

which is equivalent to

\newglossary[tag-glg]{name}{tag-gls}{tag-glo}{title}[counter]

It may be that you have some terms that are so common that they don’t need to be listed. In this case, you can define a special type of glossary that doesn’t create any associated files. This is referred to as an “ignored glossary” and it’s ignored by commands that iterate over all the glossaries, such as \printglossaries. To define an ignored glossary, use


\newignoredglossary{name}

where ⟨name⟩ is the name of the glossary (as above). This glossary type will automatically be added to the nohypertypes list, since there are no hypertargets for the entries in an ignored glossary. (The sample file sample-entryfmt.tex defines an ignored glossary.)

The glossaries-extra package provides a starred version of this command that allows hyperlinks (since ignored glossaries can be useful with bib2gls). There is also an analogous \provideignoredglossary command.

You can test if a glossary is an ignored one using:


\ifignoredglossary{name}{true}{false}

This does ⟨true⟩ if ⟨name⟩ was defined as an ignored glossary, otherwise it does ⟨false⟩.

Note that the main (default) glossary is automatically created as:

\newglossary{main}{gls}{glo}{\glossaryname}

so it can be identified by the label main (unless the nomain package option is used). Using the acronym package option is equivalent to:

\newglossary[alg]{acronym}{acr}{acn}{\acronymname}

so it can be identified by the label acronym. If you are not sure whether the acronym option has been used, you can identify the list of acronyms by the command \acronymtype which is set to acronym, if the acronym option has been used, otherwise it is set to main. Note that if you are using the main glossary as your list of acronyms, you need to declare it as a list of acronyms using the package option acronymlists.

The symbols package option creates a new glossary with the label symbols using:

\newglossary[slg]{symbols}{sls}{slo}{\glssymbolsgroupname}

The numbers package option creates a new glossary with the label numbers using:

\newglossary[nlg]{numbers}{nls}{nlo}{\glsnumbersgroupname}

The index package option creates a new glossary with the label index using:

\newglossary[ilg]{index}{ind}{idx}{\indexname}

Options 2 and 3: all glossaries must be defined before \makeglossaries to ensure that the relevant output files are opened.

See §1.3.1 Changing the Fixed Names if you want to redefine \glossaryname, especially if you are using babel or translator. (Similarly for \glssymbolsgroupname and \glsnumbersgroupname.) If you want to redefine \indexname, just follow the advice in How to change LaTeX’s “fixed names”.

Top

13. Acronyms and Other Abbreviations

The glossaries-extra package provides superior abbreviation handling. You may want to consider using that package instead of the commands described here.

Note that although this chapter uses the term “acronym”, you can also use the commands described here for initialisms or contractions (as in the case of some of the examples given below). If the glossary title is no longer applicable (for example, it should be “Abbreviations” rather than “Acronyms”) then you can change the title either by redefining \acronymname (see §1.3 Multi-Lingual Support) or by using the title in the optional argument of \printglossary (or \printacronyms). Alternatively consider using the glossaries-extra package’s abbreviations option instead.

You may have noticed in §4 Defining Glossary Entries that when you specify a new entry, you can specify alternate text to use when the term is first used in the document. This provides a useful means to define abbreviations. For convenience, the glossaries package defines the command:


\newacronym[key-val list]{label}{abbrv}{long}

This uses \newglossaryentry to create an entry with the given label in the glossary given by \acronymtype. You can specify a different glossary using the type key within the optional argument. The \newacronym command also uses the long, longplural, short and shortplural keys in \newglossaryentry to store the long and abbreviated forms and their plurals.

Note that the same restrictions on the entry ⟨label⟩ in \newglossaryentry also apply to \newacronym (see §4 Defining Glossary Entries).

If you haven’t identified the specified glossary type as a list of acronyms (via the package option acronymlists or the command \DeclareAcronymList, see §2.7 Acronym and Abbreviation Options) \newacronym will add it to the list and reset the display style for that glossary via \defglsentryfmt. If you have a mixture of acronyms and regular entries within the same glossary, care is needed if you want to change the display style: you must first identify that glossary as a list of acronyms and then use \defglsentryfmt (not redefine \glsentryfmt) before defining your entries.

The optional argument {key-val list} allows you to specify additional information. Any key that can be used in the second argument of \newglossaryentry can also be used here in ⟨key-val list⟩. For example, description (when used with one of the styles that require a description, described in §13.1 Changing the Abbreviation Style) or you can override plural forms of ⟨abbrv⟩ or ⟨long⟩ using the shortplural or longplural keys. For example:

\newacronym[longplural={diagonal matrices}]%  
  {dm}{DM}{diagonal matrix}

If the first use uses the plural form, \glspl{dm} will display: diagonal matrices (DMs). If you want to use the longplural or shortplural keys, I recommend you use \setacronymstyle to set the display style rather than using one of the pre-version 4.02 acronym styles.

As with plural and firstplural, if longplural is missing, it’s obtained by appended \glspluralsuffix to the singular form. The short plural shortplural is obtained (is not explicitly set) by appending \glsacrpluralsuffix to the short form. These commands may be changed by the associated language files, but they can’t be added to the usual caption hooks as there’s no guarantee when they’ll be expanded (as discussed earlier). A different approach is used by glossaries-extra, which has category attributes to determine whether or not to append a suffix when forming the default value of shortplural.

Since \newacronym uses \newglossaryentry, you can use commands like \gls and \glsreset as with any other glossary entry.

Since \newacronym sets type=\acronymtype, if you want to load a file containing acronym definitions using \loadglsentries[type]{filename}, the optional argument ⟨type⟩ will not have an effect unless you explicitly set the type as type=\glsdefaulttype in the optional argument to \newacronym. See §4.6 Loading Entries From a File.

Example 21 (Defining an Abbreviation)

The following defines the abbreviation IDN:

\newacronym{idn}{IDN}{identification number}

\gls{idn} will produce “identification number (IDN)” on first use and “IDN” on subsequent uses. If you want to use one of the small caps acronym styles, described in §13.1 Changing the Abbreviation Style, you need to use lower case characters for the shortened form:

\newacronym{idn}{idn}{identification number}

Now \gls{idn} will produce “identification number (idn)” on first use and “idn” on subsequent uses.

____________________________

Avoid nested definitions.

Recall from the warning in §4 Defining Glossary Entries that you should avoid using the \gls-like and \glstext-like commands within the value of keys like text and first due to complications arising from nested links. The same applies to abbreviations defined using \newacronym.

For example, suppose you have defined:

\newacronym{ssi}{SSI}{server side includes}  
\newacronym{html}{HTML}{hypertext markup language}

you may be tempted to do:

\newacronym{shtml}{S\gls{html}}{\gls{ssi} enabled \gls{html}}

Don’t! This will break the case-changing commands, such as \Gls, it will cause inconsistencies on first use, and, if hyperlinks are enabled, will cause nested hyperlinks. It will also confuse the commands used by the entry formatting (such as \glslabel).

Instead, consider doing:

\newacronym  
 [description={\gls{ssi} enabled \gls{html}}]  
 {shtml}{SHTML}{SSI enabled HTML}

or

\newacronym  
 [description={\gls{ssi} enabled \gls{html}}]  
 {shtml}{SHTML}  
 {server side includes enabled hypertext markup language}

Similarly for the \glstext-like commands.

Other approaches are available with glossaries-extra. See the section “Nested Links” in the glossaries-extra user manual.

The commands described below are similar to the \glstext-like commands in that they don’t modify the first use flag. However, their display is governed by \defentryfmt with \glscustomtext set as appropriate. All caveats that apply to the \glstext-like commands also apply to the following commands. (Including the warning immediately above this box.)

The optional arguments are the same as those for the \glstext-like commands, and there are similar star and plus variants that switch off or on the hyperlinks. As with the \glstext-like commands, the link text is placed in the argument of \glstextformat.


\acrshort[options]{label}[insert]

This sets the link text to the short form (within the argument of \acronymfont) for the entry given by ⟨label⟩. The short form is as supplied by the short key, which \newacronym implicitly sets.

There are also analogous upper case variants:


\Acrshort[options]{label}[insert]


\ACRshort[options]{label}[insert]

There are also plural versions:


\acrshortpl[options]{label}[insert]


\Acrshortpl[options]{label}[insert]


\ACRshortpl[options]{label}[insert]

The short plural form is as supplied by the shortplural key, which \newacronym implicitly sets.


\acrlong[options]{label}[insert]

This sets the link text to the long form for the entry given by ⟨label⟩. The long form is as supplied by the long key, which \newacronym implicitly sets.

There are also analogous upper case variants:


\Acrlong[options]{label}[insert]


\ACRlong[options]{label}[insert]

Again there are also plural versions:


\acrlongpl[options]{label}[insert]


\Acrlongpl[options]{label}[insert]


\ACRlongpl[options]{label}[insert]

The long plural form is as supplied by the longplural key, which \newacronym implicitly sets.

The commands below display the full form of the acronym, but note that this isn’t necessarily the same as the form used on first use. These full-form commands are shortcuts that use the above commands, rather than creating the link text from the complete full form. These full-form commands have star and plus variants and optional arguments that are passed to the above commands.


\acrfull[options]{label}[insert]

This is a shortcut for


\acrfullfmt{options}{label}{insert}

which by default does

\acrfullformat
 {\acrlong[options]{label}{insert}}
 {\acrshort[options]{label}}
where


\acrfullformat{long}{short}

by default does ⟨long⟩ (⟨short⟩). This command is now deprecated for new acronym styles but is used by the default for backward compatibility if \setacronymstyle (§13.1 Changing the Abbreviation Style) hasn’t been used. (For further details of these format commands see the documented code, glossaries-code.pdf.)

There are also analogous upper case variants:


\Acrfull[options]{label}[insert]


\ACRfull[options]{label}[insert]

and plural versions:


\acrfullpl[options]{label}[insert]


\Acrfullpl[options]{label}[insert]


\ACRfullpl[options]{label}[insert]

If you find the above commands too cumbersome to write, you can use the shortcuts package option to activate the shorter command names listed in table 13.1.


Table 13.1: Synonyms provided by the package option shortcuts
Shortcut Command Equivalent Command
\acs \acrshort
\Acs \Acrshort
\acsp \acrshortpl
\Acsp \Acrshortpl
\acl \acrlong
\Acl \Acrlong
\aclp \acrlongpl
\Aclp \Acrlongpl
\acf \acrfull
\Acf \Acrfull
\acfp \acrfullpl
\Acfp \Acrfullpl
\ac \gls
\Ac \Gls
\acp \glspl
\Acp \Glspl

It is also possible to access the long and short forms without adding information to the glossary using commands analogous to \glsentrytext (described in §9 Using Glossary Terms Without Links).

The commands that convert the first letter to upper case come with the same caveats as those for analogous commands like \Glsentrytext (non-expandable, can’t be used in PDF bookmarks, care needs to be taken if the first letter is an accented character etc). See §9 Using Glossary Terms Without Links.

The long form can be accessed using:


\glsentrylong{label}

or, with the first letter converted to upper case:


\Glsentrylong{label}

Plural forms:


\glsentrylongpl{label}


\Glsentrylongpl{label}

Similarly, to access the short form:


\glsentryshort{label}

or, with the first letter converted to upper case:


\Glsentryshort{label}

Plural forms:


\glsentryshortpl{label}


\Glsentryshortpl{label}

And the full form can be obtained using:


\glsentryfull{label}


\Glsentryfull{label}


\glsentryfullpl{label}


\Glsentryfullpl{label}

These again use \acrfullformat by default, but the new styles described in the section below use different formatting commands.

Top

13.1 Changing the Abbreviation Style

It may be that the default style doesn’t suit your requirements in which case you can switch to another style using


\setacronymstyle{style name}

where ⟨style name⟩ is the name of the required style.

You must use \setacronymstyle before you define the acronyms with \newacronym.

For example:

\usepackage[acronym]{glossaries}  
\makeglossaries  
\setacronymstyle{long-sc-short}  
\newacronym{html}{html}{hypertext markup language}  
\newacronym{xml}{xml}{extensible markup language}

Unpredictable results can occur if you try to use multiple styles.

If you need multiple abbreviation styles, then try using the glossaries-extra package, which has better abbreviation management.

Unlike the default behaviour of \newacronym, the styles used via \setacronymstyle don’t use the first or text keys, but instead they use \defglsentryfmt to set a custom format that uses the long and short keys (or their plural equivalents). This means that these styles cope better with plurals that aren’t formed by simply appending the singular form with the letter “s”. In fact, most of the predefined styles use \glsgenacfmt and modify the definitions of commands like \genacrfullformat.

Note that when you use \setacronymstyle the name key is set to


\acronymentry{label}

and the sort key is set to


\acronymsort{short}{long}

These commands are redefined by the acronym styles. However, you can redefine them again after the style has been set but before you use \newacronym. Protected expansion is performed on \acronymsort when the entry is defined.

Top

13.1.1 Predefined Acronym Styles

The glossaries package provides a number of predefined styles. These styles apply


\firstacronymfont{text}

to the short form on first use and


\acronymfont{text}

on subsequent use. The styles modify the definition of \acronymfont as required, but \firstacronymfont is only set once by the package when it’s loaded. By default \firstacronymfont{text} is the same as \acronymfont{text}. If you want the short form displayed differently on first use, you can redefine \firstacronymfont independently of the acronym style.

The predefined styles that contain sc in their name (for example long-sc-short) redefine \acronymfont to use \textsc, which means that the short form needs to be specified in lower case.

Some fonts don’t support bold small caps, so you may need to redefine \glsnamefont (see §10 Displaying a glossary) to switch to medium weight if you are using a glossary style that displays entry names in bold and you have chosen an acronym style that uses \textsc.

The predefined styles that contain sm in their name (for example long-sm-short) redefine \acronymfont to use \textsmaller.

Note that the glossaries package doesn’t define or load any package that defines \textsmaller. If you use one of the acronym styles that set \acronymfont to \textsmaller you must explicitly load the relsize package or otherwise define \textsmaller.

The remaining predefined styles redefine \acronymfont{text} to simply do its argument ⟨text⟩.

In most cases, the predefined styles adjust \acrfull and \glsentryfull (and their plural and upper case variants) to reflect the style. The only exceptions to this are the dua and footnote styles (and their variants).

The following styles are supplied by the glossaries package:

Example 22 (Adapting a Predefined Acronym Style)

Suppose I want to use the footnote-sc-desc style, but I want the name key set to the short form followed by the long form in parentheses and the sort key set to the short form. Then I need to specify the footnote-sc-desc style:

\setacronymstyle{footnote-sc-desc}

and then redefine \acronymsort and \acronymentry:

\renewcommand*{\acronymsort}[2]{#1}% sort by short form  
\renewcommand*{\acronymentry}[1]{%  
  \acronymfont{\glsentryshort{#1}}\space (\glsentrylong{#1})}%

(I’ve used \space for extra clarity, but you can just use an actual space instead.)

Note that the default Computer Modern fonts don’t support bold small caps, so another font is required. For example:

\usepackage[T1]{fontenc}

The alternative is to redefine \acronymfont so that it always switches to medium weight to ensure the small caps setting is used. For example:

\renewcommand*{\acronymfont}[1]{\textmd{\scshape #1}}

The sample file sampleFnAcrDesc.tex illustrates this example.

____________________________

Top

13.1.2 Defining A Custom Acronym Style

You may find that the predefined acronyms styles that come with the glossaries package don’t suit your requirements. In this case you can define your own style using:


\newacronymstyle{style name}{display}{definitions}

where ⟨style name⟩ is the name of the new style (avoid active characters). The second argument, ⟨display⟩, is equivalent to the mandatory argument of \defglsentryfmt. You can simply use \glsgenacfmt or you can customize the display using commands like \ifglsused, \glsifplural and \glscapscase. (See §6.3 Changing the format of the link text for further details.) If the style is likely to be used with a mixed glossary (that is entries in that glossary are defined both with \newacronym and \newglossaryentry) then you can test if the entry is an acronym and use \glsgenacfmt if it is or \glsgenentryfmt if it isn’t. For example, the long-short style sets ⟨display⟩ as

\ifglshaslong{\glslabel}{\glsgenacfmt}{\glsgenentryfmt}%

(You can use \ifglshasshort instead of \ifglshaslong to test if the entry is an acronym if you prefer.)

The third argument, ⟨definitions⟩, can be used to redefine the commands that affect the display style, such as \acronymfont or, if ⟨display⟩ uses \glsgenacfmt, \genacrfullformat and its variants.

Note that \setacronymstyle redefines \glsentryfull and \acrfullfmt to use \genacrfullformat (and similarly for the plural and upper case variants). If this isn’t appropriate for the style (as in the case of styles like footnote and dua) \newacronymstyle should redefine these commands within ⟨definitions⟩.

Within \newacronymstyle’s ⟨definitions⟩ argument you can also redefine


\GenericAcronymFields

This is a list of additional fields to be set in \newacronym. You can use the following token registers to access the entry label, long form and short form: \glslabeltok, \glslongtok and \glsshorttok. As with all TeX registers, you can access their values by preceding the register with \the. For example, the long-short style does: