The Not So Short

Introduction to L

X 2

Or L

X 2

in 157 minutes

by Tobias Oetiker

Hubert Partl, Irene Hyna and Elisabeth Schlegl

Version 5.05, July 18, 2015

This document is free; you can redistribute it and/or modify it under the terms

of the GNU General Public License as published by the Free Software Foundation;

either version 2 of the License, or (at your option) any later version.

This document is distributed in the hope that it will be useful, but without

any warranty; without even the implied warranty of merchantability or ﬁtness for a

particular purpose. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with

this document; if not, write to the Free Software Foundation, Inc., 675 Mass Ave,

Cambridge, MA 02139, USA.

Thank you!

Much of the material used in this introduction comes from an Austrian

introduction to L

X 2.09 written in German by:

Hubert Partl <partl@mail.boku.ac.at>

Zentraler Informatikdienst der Universität für Bodenkultur Wien

Irene Hyna <Irene.Hyna@bmwf.ac.at>

Bundesministerium für Wissenschaft und Forschung Wien

Elisabeth Schlegl <noemail>

in Graz

If you are interested in the German document, you can ﬁnd a version

updated for L

X 2

by Jörg Knappen at CTAN://info/lshort/german

iv Thank you!

The following individuals helped with corrections, suggestions and material

to improve this paper. They put in a big eﬀort to help me get this document

into its present shape. I would like to sincerely thank all of them. Naturally,

all the mistakes you’ll ﬁnd in this book are mine. If you ever ﬁnd a word

that is spelled correctly, it must have been one of the people below dropping

me a line.

Eric Abrahamsen, Lenimar Nunes de Andrade, Eilinger August, Rosemary Bailey,

Barbara Beeton, Marc Bevand, Connor Blakey, Salvatore Bonaccorso, Pietro Braione,

Friedemann Brauer, Markus Brühwiler, Jan Busa, David Carlisle, Neil Carter,

Carl Cerecke, Mike Chapman, Pierre Chardaire, Xingyou Chen, Christopher Chin,

Diego Clavadetscher, Wim van Dam, Benjamin Deschwanden Jan Dittberner,

Michael John Downes, Matthias Dreier, David Dureisseix, Hans Ehrbar, Elliot, Rock-

rush Engch, William Faulk, Robin Fairbairns, Johan Falk, Jörg Fischer, Frank Fis-

chli, Daniel Flipo, Frank, Mic Milic Frederickx, David Frey, Erik Frisk, Hans Fugal,

Robert Funnell, Greg Gamble, Andy Goth, Cyril Goutte, Kasper B. Graversen,

Arlo Griﬃths, Alexandre Guimond, Neil Hammond, Christoph Hamburger, Ras-

mus Borup Hansen, Joseph Hilferty, Daniel Hirsbrunner, Martien Hulsen, Björn

Hvittfeldt, Morten Høgholm, Werner Icking, Eric Jacoboni, Jakob, Alan Jeﬀrey, Mar-

tin Jenkins, Byron Jones, David Jones, Johannes-Maria Kaltenbach, Nils Kanning,

Andrzej Kawalec, Christian Kern, Alain Kessi, Axel Kielhorn, Sander de Kievit,

Kjetil Kjernsmo, Tobias Klauser, Jörg Knappen, Michael Koundouros, Matt Kraai,

Tobias Krewer, Flori Lambrechts, Mike Lee, Maik Lehradt, Rémi Letot, Axel Liljen-

crantz, Jasper Loy, Johan Lundberg, Martin Maechler, Alexander Mai, Claus Malten,

Kevin Van Maren, Pablo Markin, I. J. Vera Marún, Hendrik Maryns, Chris McCor-

mack, Aleksandar S. Milosevic, Henrik Mitsch, Stefan M. Moser, Philipp Nagele,

Richard Nagy, Manuel Oetiker, Urs Oswald, Hubert Partl, Marcelo Pasin, Mar-

tin Pﬁster, Lan Thuy Pham, Breno Pietracci, Demerson Andre Polli, Maksym Polyakov,

Nikos Pothitos, John Reﬂing, Mike Ressler, Brian Ripley, Kurt Rosenfeld, Bernd Rosen-

lecher, Chris Rowley, Young U. Ryu, Risto Saarelma, András Salamon, José Car-

los Santos, Christopher Sawtell, Gilles Schintgen, Craig Schlenter, Hanspeter Schmid,

Baron Schwartz, Jordi Serra i Solanich, Miles Spielberg, Susan Stewart, Matthieu Stigler,

Geoﬀrey Swindale, Laszlo Szathmary, Boris Tobotras, Josef Tkadlec, Scott Veirs,

Didier Verna, Carl-Gustav Werner, Fabian Wernli, Matthew Widmann, David Wood-

house, Chris York, Rick Zaccone, Fritz Zaucker, and Mikhail Zotov.

Preface

X [

] is a typesetting system that is very suitable for producing scientiﬁc

and mathematical documents of high typographical quality. It is also suitable

for producing all sorts of other documents, from simple letters to complete

books. L

X uses T

X [2] as its formatting engine.

This short introduction describes

X 2

and should be suﬃcient for

most applications of L

X. Refer to [

] for a complete description of the

X system.

This introduction is split into 6 chapters:

Chapter 1

tells you about the basic structure of

X 2

documents. You

will also learn a bit about the history of L

X. After reading this

chapter, you should have a rough understanding how L

X works.

Chapter 2

goes into the details of typesetting your documents. It explains

most of the essential L

X commands and environments. After reading

this chapter, you will be able to write your ﬁrst documents.

Chapter 3

explains how to typeset formulae with L

X. Many examples

demonstrate how to use one of L

X’s main strengths. At the end

of the chapter are tables listing all mathematical symbols available in

Chapter 4

explains indexes, bibliography generation and inclusion of EPS

graphics. It introduces creation of PDF documents with pdfL

X and

presents some handy extension packages.

Chapter 5

shows how to use L

X for creating graphics. Instead of drawing

a picture with some graphics program, saving it to a ﬁle and then

including it into L

X, you describe the picture and have L

X draw

it for you.

Chapter 6

contains some potentially dangerous information about how to

alter the standard document layout produced by L

X. It will tell you

how to change things such that the beautiful output of L

X turns

ugly or stunning, depending on your abilities.

vi Preface

It is important to read the chapters in order—the book is not that big, after

all. Be sure to carefully read the examples, because a lot of the information

is in the examples placed throughout the book.

X is available for most computers, from the PC and Mac to large UNIX

and VMS systems. On many university computer clusters you will ﬁnd that

a L

X installation is available, ready to use. Information on how to access

the local L

X installation should be provided in the Local Guide [

]. If you

have problems getting started, ask the person who gave you this booklet.

The scope of this document is not to tell you how to install and set up a

X system, but to teach you how to write your documents so that they

can be processed by L

If you need to get hold of any L

X related material, have a look at one of

the Comprehensive T

X Archive Network (CTAN) sites. The homepage is

at http://www.ctan.org.

You will ﬁnd other references to CTAN throughout the book, especially

pointers to software and documents you might want to download. Instead of

writing down complete urls, I just wrote

CTAN:

followed by whatever location

within the CTAN tree you should go to.

If you want to run L

X on your own computer, take a look at what is

available from CTAN://systems.

If you have ideas for something to be added, removed or altered in this

document, please let me know. I am especially interested in feedback from

X novices about which bits of this intro are easy to understand and

which could be explained better.

Tobias Oetiker <tobi@oetiker.ch>

OETIKER+PARTNER AG

Aarweg 15

4600 Olten

Switzerland

The current version of this document is available on

CTAN://info/lshort

Contents

Thank you! iii

Preface v

1 Things You Need to Know 1

1.1 The Name of the Game . . . . . . . . . . . . . . . . . . . . . 1

1.1.1 T

X . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1.1.2 L

X . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2 Basics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1.2.1 Author, Book Designer, and Typesetter . . . . . . . . 2

1.2.2 Layout Design . . . . . . . . . . . . . . . . . . . . . . 2

1.2.3 Advantages and Disadvantages . . . . . . . . . . . . . 3

1.3 L

X Input Files . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.3.1 Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

1.3.2 Special Characters . . . . . . . . . . . . . . . . . . . . 5

1.3.3 L

X Commands . . . . . . . . . . . . . . . . . . . . 5

1.3.4 Comments . . . . . . . . . . . . . . . . . . . . . . . . . 6

1.4 Input File Structure . . . . . . . . . . . . . . . . . . . . . . . 7

1.5 A Typical Command Line Session . . . . . . . . . . . . . . . 7

1.6 The Layout of the Document . . . . . . . . . . . . . . . . . . 9

1.6.1 Document Classes . . . . . . . . . . . . . . . . . . . . 9

1.6.2 Packages . . . . . . . . . . . . . . . . . . . . . . . . . 10

1.6.3 Page Styles . . . . . . . . . . . . . . . . . . . . . . . . 13

1.7 Files You Might Encounter . . . . . . . . . . . . . . . . . . . 13

1.8 Big Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2 Typesetting Text 17

2.1 The Structure of Text and Language . . . . . . . . . . . . . . 17

2.2 Line Breaking and Page Breaking . . . . . . . . . . . . . . . . 19

2.2.1 Justiﬁed Paragraphs . . . . . . . . . . . . . . . . . . . 19

2.2.2 Hyphenation . . . . . . . . . . . . . . . . . . . . . . . 20

2.3 Ready-Made Strings . . . . . . . . . . . . . . . . . . . . . . . 21

2.4 Special Characters and Symbols . . . . . . . . . . . . . . . . . 21

viii CONTENTS

2.4.1 Quotation Marks . . . . . . . . . . . . . . . . . . . . . 21

2.4.2 Dashes and Hyphens . . . . . . . . . . . . . . . . . . . 22

2.4.3 Tilde (∼) . . . . . . . . . . . . . . . . . . . . . . . . . 22

2.4.4 Slash (/) . . . . . . . . . . . . . . . . . . . . . . . . . 22

2.4.5 Degree Symbol (◦) . . . . . . . . . . . . . . . . . . . . 22

2.4.6 The Euro Currency Symbol (e) . . . . . . . . . . . . . 23

2.4.7 Ellipsis (. . . ) . . . . . . . . . . . . . . . . . . . . . . . 24

2.4.8 Ligatures . . . . . . . . . . . . . . . . . . . . . . . . . 24

2.4.9 Accents and Special Characters . . . . . . . . . . . . . 24

2.5 International Language Support . . . . . . . . . . . . . . . . . 25

2.5.1 Support for Portuguese . . . . . . . . . . . . . . . . . 27

2.5.2 Support for French . . . . . . . . . . . . . . . . . . . . 28

2.5.3 Support for German . . . . . . . . . . . . . . . . . . . 29

2.5.4 Support for Korean . . . . . . . . . . . . . . . . . . . . 30

2.5.5 Writing in Greek . . . . . . . . . . . . . . . . . . . . . 31

2.5.6 Support for Cyrillic . . . . . . . . . . . . . . . . . . . 32

2.5.7 Support for Mongolian . . . . . . . . . . . . . . . . . . 33

2.5.8 The Unicode option . . . . . . . . . . . . . . . . . . . 34

2.6 The Space Between Words . . . . . . . . . . . . . . . . . . . . 38

2.7 Titles, Chapters, and Sections . . . . . . . . . . . . . . . . . . 39

2.8 Cross References . . . . . . . . . . . . . . . . . . . . . . . . . 41

2.9 Footnotes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41

2.10 Emphasized Words . . . . . . . . . . . . . . . . . . . . . . . . 42

2.11 Environments . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

2.11.1 Itemize, Enumerate, and Description . . . . . . . . . . 42

2.11.2 Flushleft, Flushright, and Center . . . . . . . . . . . . 43

2.11.3 Quote, Quotation, and Verse . . . . . . . . . . . . . . 44

2.11.4 Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . 44

2.11.5 Printing Verbatim . . . . . . . . . . . . . . . . . . . . 45

2.11.6 Tabular . . . . . . . . . . . . . . . . . . . . . . . . . . 45

2.12 Floating Bodies . . . . . . . . . . . . . . . . . . . . . . . . . . 48

2.13 Protecting Fragile Commands . . . . . . . . . . . . . . . . . . 51

3 Typesetting Mathematical Formulae 53

3.1 The A

S-L

X bundle . . . . . . . . . . . . . . . . . . . . . 53

3.2 Single Equations . . . . . . . . . . . . . . . . . . . . . . . . . 53

3.2.1 Math Mode . . . . . . . . . . . . . . . . . . . . . . . . 55

3.3 Building Blocks of a Mathematical Formula . . . . . . . . . . 56

3.4 Single Equations that are Too Long: multline . . . . . . . . . 61

3.5 Multiple Equations . . . . . . . . . . . . . . . . . . . . . . . . 62

3.5.1 Problems with Traditional Commands . . . . . . . . . 62

3.5.2 IEEEeqnarray Environment . . . . . . . . . . . . . . . 64

3.5.3 Common Usage . . . . . . . . . . . . . . . . . . . . . . 64

3.6 Arrays and Matrices . . . . . . . . . . . . . . . . . . . . . . . 67

CONTENTS ix

3.7 Spacing in Math Mode . . . . . . . . . . . . . . . . . . . . . . 68

3.7.1 Phantoms . . . . . . . . . . . . . . . . . . . . . . . . . 69

3.8 Fiddling with the Math Fonts . . . . . . . . . . . . . . . . . . 70

3.8.1 Bold Symbols . . . . . . . . . . . . . . . . . . . . . . . 70

3.9 Theorems, Lemmas, . . . . . . . . . . . . . . . . . . . . . . . . 71

3.9.1 Proofs and End-of-Proof Symbol . . . . . . . . . . . . 72

3.10 List of Mathematical Symbols . . . . . . . . . . . . . . . . . . 75

4 Specialities 83

4.1 Including Encapsulated PostScript . . . . . . . . . . . . . . 83

4.2 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

4.3 Indexing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

4.4 Fancy Headers . . . . . . . . . . . . . . . . . . . . . . . . . . 87

4.5 The Verbatim Package . . . . . . . . . . . . . . . . . . . . . . 89

4.6 Installing Extra Packages . . . . . . . . . . . . . . . . . . . . 89

4.7 Working with pdfL

X . . . . . . . . . . . . . . . . . . . . . 90

4.7.1 PDF Documents for the Web . . . . . . . . . . . . . . 91

4.7.2 The Fonts . . . . . . . . . . . . . . . . . . . . . . . . . 92

4.7.3 Using Graphics . . . . . . . . . . . . . . . . . . . . . . 93

4.7.4 Hypertext Links . . . . . . . . . . . . . . . . . . . . . 93

4.7.5 Problems with Links . . . . . . . . . . . . . . . . . . . 96

4.7.6 Problems with Bookmarks . . . . . . . . . . . . . . . . 96

4.7.7 Source Compatibility Between L

X and pdfL

X . . 97

4.8 Working with X

X . . . . . . . . . . . . . . . . . . . . . . 98

4.8.1 The Fonts . . . . . . . . . . . . . . . . . . . . . . . . . 98

4.8.2 Compatibility Between X

X and pdfL

X . . . . 100

4.9 Creating Presentations . . . . . . . . . . . . . . . . . . . . . . 100

5 Producing Mathematical Graphics 105

5.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

5.2 The picture Environment . . . . . . . . . . . . . . . . . . . . 106

5.2.1 Basic Commands . . . . . . . . . . . . . . . . . . . . . 106

5.2.2 Line Segments . . . . . . . . . . . . . . . . . . . . . . 107

5.2.3 Arrows . . . . . . . . . . . . . . . . . . . . . . . . . . 108

5.2.4 Circles . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

5.2.5 Text and Formulas . . . . . . . . . . . . . . . . . . . . 110

5.2.6 \multiput and \linethickness . . . . . . . . . . . . 110

5.2.7 Ovals . . . . . . . . . . . . . . . . . . . . . . . . . . . 111

5.2.8 Multiple Use of Predeﬁned Picture Boxes . . . . . . . 112

5.2.9 Quadratic Bézier Curves . . . . . . . . . . . . . . . . . 113

5.2.10 Catenary . . . . . . . . . . . . . . . . . . . . . . . . . 114

5.2.11 Rapidity in the Special Theory of Relativity . . . . . . 115

5.3 The PGF and TikZ Graphics Packages . . . . . . . . . . . . . 115

x CONTENTS

6 Customising L

X 119

6.1 New Commands, Environments and Packages . . . . . . . . . 119

6.1.1 New Commands . . . . . . . . . . . . . . . . . . . . . 120

6.1.2 New Environments . . . . . . . . . . . . . . . . . . . . 121

6.1.3 Extra Space . . . . . . . . . . . . . . . . . . . . . . . . 121

6.1.4 Commandline L

X . . . . . . . . . . . . . . . . . . . 122

6.1.5 Your Own Package . . . . . . . . . . . . . . . . . . . . 123

6.2 Fonts and Sizes . . . . . . . . . . . . . . . . . . . . . . . . . . 123

6.2.1 Font Changing Commands . . . . . . . . . . . . . . . 123

6.2.2 Danger, Will Robinson, Danger . . . . . . . . . . . . . 126

6.2.3 Advice . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

6.3 Spacing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127

6.3.1 Line Spacing . . . . . . . . . . . . . . . . . . . . . . . 127

6.3.2 Paragraph Formatting . . . . . . . . . . . . . . . . . . 128

6.3.3 Horizontal Space . . . . . . . . . . . . . . . . . . . . . 129

6.3.4 Vertical Space . . . . . . . . . . . . . . . . . . . . . . 130

6.4 Page Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

6.5 More Fun With Lengths . . . . . . . . . . . . . . . . . . . . . 133

6.6 Boxes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

6.7 Rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135

A Installing L

X 137

A.1 What to Install . . . . . . . . . . . . . . . . . . . . . . . . . . 137

A.2 Cross Platform Editor . . . . . . . . . . . . . . . . . . . . . . 137

A.3 T

X on Mac OS X . . . . . . . . . . . . . . . . . . . . . . . . 138

A.3.1 T

X Distribution . . . . . . . . . . . . . . . . . . . . . 138

A.3.2 OSX T

X Editor . . . . . . . . . . . . . . . . . . . . . 138

A.3.3 Treat yourself to PDFView . . . . . . . . . . . . . . . 138

A.4 T

X on Windows . . . . . . . . . . . . . . . . . . . . . . . . . 138

A.4.1 Getting T

X . . . . . . . . . . . . . . . . . . . . . . . 138

A.4.2 A L

X editor . . . . . . . . . . . . . . . . . . . . . . 139

A.4.3 Document Preview . . . . . . . . . . . . . . . . . . . . 139

A.4.4 Working with graphics . . . . . . . . . . . . . . . . . . 139

A.5 T

X on Linux . . . . . . . . . . . . . . . . . . . . . . . . . . . 139

Bibliography 141

Index 144

List of Figures

1.1 A Minimal L

X File. . . . . . . . . . . . . . . . . . . . . . . 7

1.2 Example of a Realistic Journal Article. . . . . . . . . . . . . . 8

4.1 Example fancyhdr Setup. . . . . . . . . . . . . . . . . . . . . . 88

4.2 Sample code for the beamer class . . . . . . . . . . . . . . . . 101

6.1 Example Package. . . . . . . . . . . . . . . . . . . . . . . . . 123

6.2 Layout parameters for this book. . . . . . . . . . . . . . . . . 132

List of Tables

1.1 Document Classes. . . . . . . . . . . . . . . . . . . . . . . . . 10

1.2 Document Class Options. . . . . . . . . . . . . . . . . . . . . 11

1.3 Some of the Packages Distributed with L

X. . . . . . . . . . 12

1.4 The Predeﬁned Page Styles of L

X. . . . . . . . . . . . . . . 13

2.1 A bag full of Euro symbols . . . . . . . . . . . . . . . . . . . 23

2.2 Accents and Special Characters. . . . . . . . . . . . . . . . . . 25

2.3 Preamble for Portuguese documents. . . . . . . . . . . . . . . 28

2.4 Special commands for French. . . . . . . . . . . . . . . . . . . 29

2.5 German Special Characters. . . . . . . . . . . . . . . . . . . . 30

2.6 Preamble for Greek documents. . . . . . . . . . . . . . . . . . 31

2.7 Greek Special Characters. . . . . . . . . . . . . . . . . . . . . 32

2.8 Bulgarian, Russian, and Ukrainian . . . . . . . . . . . . . . . 33

2.9 Float Placing Permissions. . . . . . . . . . . . . . . . . . . . . 49

3.1 Math Mode Accents. . . . . . . . . . . . . . . . . . . . . . . . 75

3.2 Greek Letters. . . . . . . . . . . . . . . . . . . . . . . . . . . . 75

3.3 Binary Relations. . . . . . . . . . . . . . . . . . . . . . . . . . 76

3.4 Binary Operators. . . . . . . . . . . . . . . . . . . . . . . . . 76

3.5 BIG Operators. . . . . . . . . . . . . . . . . . . . . . . . . . . 77

3.6 Arrows. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

3.7 Arrows as Accents. . . . . . . . . . . . . . . . . . . . . . . . . 77

3.8 Delimiters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

3.9 Large Delimiters. . . . . . . . . . . . . . . . . . . . . . . . . . 78

3.10 Miscellaneous Symbols. . . . . . . . . . . . . . . . . . . . . . 78

3.11 Non-Mathematical Symbols. . . . . . . . . . . . . . . . . . . . 78

3.12 A

S Delimiters. . . . . . . . . . . . . . . . . . . . . . . . . . 79

3.13 A

S Greek and Hebrew. . . . . . . . . . . . . . . . . . . . . 79

3.14 Math Alphabets. . . . . . . . . . . . . . . . . . . . . . . . . . 79

3.15 A

S Binary Operators. . . . . . . . . . . . . . . . . . . . . . 79

3.16 A

S Binary Relations. . . . . . . . . . . . . . . . . . . . . . 80

3.17 A

S Arrows. . . . . . . . . . . . . . . . . . . . . . . . . . . . 81

3.18 A

S Negated Binary Relations and Arrows. . . . . . . . . . 82

xiv LIST OF TABLES

3.19 A

S Miscellaneous. . . . . . . . . . . . . . . . . . . . . . . . 82

4.1 Key Names for graphicx Package. . . . . . . . . . . . . . . . . 84

4.2 Index Key Syntax Examples. . . . . . . . . . . . . . . . . . . 87

6.1 Fonts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

6.2 Font Sizes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

6.3 Absolute Point Sizes in Standard Classes. . . . . . . . . . . . 125

6.4 Math Fonts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

6.5 T

X Units. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

Chapter 1

Things You Need to Know

The ﬁrst part of this chapter presents a short overview of the philosophy and

history of

X 2

. The second part focuses on the basic structures of a L

document. After reading this chapter, you should have a rough knowledge of

how L

X works, which you will need to understand the rest of this book.

1.1 The Name of the Game

1.1.1 T

X is a computer program created by Donald E. Knuth [

]. It is aimed at

typesetting text and mathematical formulae. Knuth started writing the T

typesetting engine in 1977 to explore the potential of the digital printing

equipment that was beginning to inﬁltrate the publishing industry at that

time, especially in the hope that he could reverse the trend of deteriorating

typographical quality that he saw aﬀecting his own books and articles. T

as we use it today was released in 1982, with some slight enhancements added

in 1989 to better support 8-bit characters and multiple languages. T

X is

renowned for being extremely stable, for running on many diﬀerent kinds of

computers, and for being virtually bug free. The version number of T

X is

converging to π and is now at 3.141592653.

X is pronounced “Tech,” with a “ch” as in the German word “Ach”

in the Scottish “Loch.” The “ch” originates from the Greek alphabet where

X is the letter “ch” or “chi”. T

X is also the ﬁrst syllable of the Greek word

τεχνική (technique). In an ASCII environment, T

X becomes TeX.

In german there are actually two pronounciations for “ch” and one might assume that

the soft “ch” sound from “Pech” would be a more appropriate. Asked about this, Knuth

wrote in the German Wikipedia: I do not get angry when people pronounce T

X in their

favorite way . . . and in Germany many use a soft ch because the X follows the vowel e, not

the harder ch that follows the vowe l a. In Russia, ‘tex’ is a very common word, pronounced

‘tyekh’. But I believe the most proper pronunciation is heard in Greece, where you have the

harsher ch of ach and Loch.

2 Things You Need to Know

1.1.2 L

X enables authors to typeset and print their work at the highest typo-

graphical quality, using a predeﬁned, professional layout. L

X was originally

written by Leslie Lamport [

]. It uses the T

X formatter as its typesetting

engine. These days L

X is maintained by Frank Mittelbach.

X is pronounced “Lay-tech” or “Lah-tech.” If you refer to L

X in

an ASCII environment, you type

LaTeX

X 2

is pronounced “Lay-tech

two e” and typed LaTeX2e.

1.2 Basics

1.2.1 Author, Book Designer, and Typesetter

To publish something, authors give their typed manuscript to a publishing

company. One of their book designers then decides the layout of the document

(column width, fonts, space before and after headings, . . . ). The book designer

writes his instructions into the manuscript and then gives it to a typesetter,

who typesets the book according to these instructions.

A human book designer tries to ﬁnd out what the author had in mind

while writing the manuscript. He decides on chapter headings, citations,

examples, formulae, etc. based on his professional knowledge and from the

contents of the manuscript.

In a L

X environment, L

X takes the role of the book designer and

uses T

X as its typesetter. But L

X is “only” a program and therefore

needs more guidance. The author has to provide additional information to

describe the logical structure of his work. This information is written into

the text as “L

X commands.”

This is quite diﬀerent from the WYSIWYG

approach that most modern

word processors, such as MS Word or LibreOﬃce, take. With these applica-

tions, authors specify the document layout interactively while typing text

into the computer. They can see on the screen how the ﬁnal work will look

when it is printed.

When using L

X it is not normally possible to see the ﬁnal output while

typing the text, but the ﬁnal output can be previewed on the screen after

processing the ﬁle with L

X. Then corrections can be made before actually

sending the document to the printer.

1.2.2 Layout Design

Typographical design is a craft. Unskilled authors often commit serious

formatting errors by assuming that book design is mostly a question of

aesthetics—“If a document looks good artistically, it is well designed.” But

What you see is what you get.

1.2 Basics 3

as a document has to be read and not hung up in a picture gallery, the

readability and understandability is much more important than the beautiful

look of it. Examples:

•

The font size and the numbering of headings have to be chosen to make

the structure of chapters and sections clear to the reader.

•

The line length has to be short enough not to strain the eyes of the

reader, while long enough to ﬁll the page beautifully.

With WYSIWYG systems, authors often generate aesthetically pleasing

documents with very little or inconsistent structure. L

X prevents such

formatting errors by forcing the author to declare the logical structure of his

document. L

X then chooses the most suitable layout.

1.2.3 Advantages and Disadvantages

When people from the WYSIWYG world meet people who use L

X, they

often discuss “the advantages of L

X over a normal word processor” or the

opposite. The best thing to do when such a discussion starts is to keep a low

proﬁle, since such discussions often get out of hand. But sometimes there is

no escaping . . .

So here is some ammunition. The main advantages of L

X over normal

word processors are the following:

•

Professionally crafted layouts are available, which make a document

really look as if “printed.”

•

The typesetting of mathematical formulae is supported in a convenient

way.

•

Users only need to learn a few easy-to-understand commands that

specify the logical structure of a document. They almost never need to

tinker with the actual layout of the document.

•

Even complex structures such as footnotes, references, table of contents,

and bibliographies can be generated easily.

•

Free add-on packages exist for many typographical tasks not directly

supported by basic L

X. For example, packages are available to

include PostScript graphics or to typeset bibliographies conforming

to exact standards. Many of these add-on packages are described in

The L

X Companion [3].

•

X encourages authors to write well-structured texts, because this

is how L

X works—by specifying structure.

4 Things You Need to Know

•

X, the formatting engine of

X 2

, is highly portable and free.

Therefore the system runs on almost any hardware platform available.

X also has some disadvantages, and I guess it’s a bit diﬃcult for me to

ﬁnd any sensible ones, though I am sure other people can tell you hundreds

;-)

• L

X does not work well for people who have sold their souls . . .

•

Although some parameters can be adjusted within a predeﬁned docu-

ment layout, the design of a whole new layout is diﬃcult and takes a

lot of time.

• It is very hard to write unstructured and disorganized documents.

•

Your hamster might, despite some encouraging ﬁrst steps, never be

able to fully grasp the concept of Logical Markup.

1.3 L

X Input Files

The input for L

X is a plain text ﬁle. On Unix/Linux text ﬁles are pretty

common. On windows, one would use Notepad to create a text ﬁle. It

contains the text of the document, as well as the commands that tell L

how to typeset the text. If you are working with a L

X IDE, it will contain

a program for creating L

X input ﬁles in text format.

1.3.1 Spaces

“Whitespace” characters, such as blank or tab, are treated uniformly as

“space” by L

X. Several consecutive whitespace characters are treated as

one “space.” Whitespace at the start of a line is generally ignored, and a

single line break is treated as “whitespace.”

An empty line between two lines of text deﬁnes the end of a paragraph.

Several empty lines are treated the same as one empty line. The text below

is an example. On the left hand side is the text from the input ﬁle, and on

the right hand side is the formatted output.

It does not matter whether you

enter one or several spaces

after a word.

An empty line starts a new

paragraph.

It does not matter whether you enter one

or several spaces after a word.

An empty line starts a new paragraph.

Rumour says that this is one of the key elements that will be addressed in the upcoming

X3 system.

1.3 L

X Input Files 5

1.3.2 Special Characters

The following symbols are reserved characters that either have a special

meaning under L

X or are not available in all the fonts. If you enter them

directly in your text, they will normally not print, but rather coerce L

to do things you did not intend.

# $ % ^ & _ { } ~ \

As you will see, these characters can be used in your documents all the

same by using a preﬁx backslash:

\# \$ \% \^{} \& \_ \{ \} \~{}

\textbackslash

# $ % ˆ & _ { } ˜ \

The other symbols and many more can be printed with special commands

in mathematical formulae or as accents. The backslash character \ can not

be entered by adding another backslash in front of it (

); this sequence is

used for line breaking. Use the \textbackslash command instead.

1.3.3 L

X Commands

X commands are case sensitive, and take one of the following two formats:

•

They start with a backslash

and then have a name consisting of

letters only. Command names are terminated by a space, a number or

any other ‘non-letter.’

• They consist of a backslash and exactly one non-letter.

•

Many commands exist in a ‘starred variant’ where a star is appended

to the command name.

X ignores whitespace after commands. If you want to get a space after

a command, you have to put either an empty parameter

{}

and a blank or a

special spacing command after the command name. The empty parameter

{}

stops L

X from eating up all the white space after the command name.

New \TeX users may miss whitespaces

after a command. % renders wrong

Experienced \TeX{} users are

\TeX perts, and know how to use

whitespaces. % renders correct

New T

Xusers may miss whitespaces af-

ter a command. Experienced T

X users

are T

Xperts, and know how to use white-

spaces.

Some commands require a parameter, which has to be given between

curly braces

{ }

after the command name. Some commands take optional

parameters, which are inserted after the command name in square brack-

ets [ ].

6 Things You Need to Know

\command[optional parameter]{parameter}

The next examples use some L

X commands. Don’t worry about them;

they will be explained later.

You can \textsl{lean} on me!

You can lean on me!

Please, start a new line

right here!\newline

Thank you!

Please, start a new line right here!

Thank you!

1.3.4 Comments

When L

X encounters a

character while processing an input ﬁle, it ignores

the rest of the present line, the line break, and all whitespace at the beginning

of the next line.

This can be used to write notes into the input ﬁle, which will not show

up in the printed version.

This is an % stupid

% Better: instructive <----

example: Supercal%

ifragilist%

icexpialidocious

This is an example: Supercalifragilisticex-

pialidocious

The

character can also be used to split long input lines where no

whitespace or line breaks are allowed.

For longer comments you could use the

comment

environment provided by

the verbatim package. Add the line

\usepackage{verbatim}

to the preamble

of your document as explained below to use this command.

This is another

\begin{comment}

rather stupid,

but helpful

\end{comment}

example for embedding

comments in your document.

This is another example for embedding

comments in your document.

Note that this won’t work inside complex environments, like math for

example.

1.4 Input File Structure 7

1.4 Input File Structure

When

X 2

processes an input ﬁle, it expects it to follow a certain

structure. Thus every input ﬁle must start with the command

\documentclass{...}

This speciﬁes what sort of document you intend to write. After that, add

commands to inﬂuence the style of the whole document, or load packages

that add new features to the L

X system. To load such a package you use

the command

\usepackage{...}

When all the setup work is done,

you start the body of the text with

the command

\begin{document}

Now you enter the text mixed with some useful L

X commands. At

the end of the document you add the

\end{document}

command, which tells L

X to call it a day. Anything that follows this

command will be ignored by L

Figure 1.1 shows the contents of a minimal

X 2

ﬁle. A slightly more

complicated input ﬁle is given in Figure 1.2.

1.5 A Typical Command Line Session

I bet you must be dying to try out the neat small L

X input ﬁle shown

on page 7. Here is some help: L

X itself comes without a GUI or fancy

buttons to press. It is just a program that crunches away at your input ﬁle.

Some L

X installations feature a graphical front-end where there is a L

button to start compiling your input ﬁle. On other systems there might be

The area between \documentclass and \begin{document} is called the preamble.

\documentclass{article}

\begin{document}

Small is beautiful.

\end{document}

Figure 1.1: A Minimal L

X File.

8 Things You Need to Know

some typing involved, so here is how to coax L

X into compiling your input

ﬁle on a text based system. Please note: this description assumes that a

working L

X installation already sits on your computer.

Edit/Create your L

X input ﬁle. This ﬁle must be plain ASCII text.

On Unix all the editors will create just that. On Windows you might

want to make sure that you save the ﬁle in ASCII or Plain Text format.

When picking a name for your ﬁle, make sure it bears the extension

.tex.

Open a shell or cmd window, cd to the directory where your input ﬁle

is located and run L

X on your input ﬁle. If successful you will end

up with a

.dvi

ﬁle. It may be necessary to run L

X several times to

get the table of contents and all internal references right. When your

input ﬁle has a bug L

X will tell you about it and stop processing

your input ﬁle. Type ctrl-D to get back to the command line.

latex foo.tex

Now you may view the DVI ﬁle. There are several ways to do that.

This is the case with most well groomed Unix Systems, and . . . Real Men use Unix,

so . . . ;-)

\documentclass[a4paper,11pt]{article}

% define the title

\author{H.~Partl}

\title{Minimalism}

\begin{document}

% generates the title

\maketitle

% insert the table of contents

\tableofcontents

\section{Some Interesting Words}

Well, and here begins my lovely article.

\section{Good Bye World}

\ldots{} and here it ends.

\end{document}

Figure 1.2: Example of a Realistic Journal Article. Note that all the

commands you see in this example will be explained later in the introduction.

1.6 The Layout of the Document 9

Look at the ﬁle on screen with

xdvi foo.dvi &

This only works on Unix with X11. If you are on Windows you might

want to try yap (yet another previewer).

Convert the dvi ﬁle to PostScript for printing or viewing with

GhostScript.

dvips -Pcmz foo.dvi -o foo.ps

If you are lucky your L

X system even comes with the

dvipdf

tool,

which allows you to convert your .dvi ﬁles straight into pdf.

dvipdf foo.dvi

1.6 The Layout of the Document

1.6.1 Document Classes

The ﬁrst information L

X needs to know when processing an input ﬁle is

the type of document the author wants to create. This is speciﬁed with the

\documentclass command.

\documentclass[options]{class}

Here class speciﬁes the type of document to be created. Table 1.1 lists the

document classes explained in this introduction. The

X 2

distribution

provides additional classes for other documents, including letters and slides.

The options parameter customises the behaviour of the document class. The

options have to be separated by commas. The most common options for the

standard document classes are listed in Table 1.2.

Example: An input ﬁle for a L

X document could start with the line

\documentclass[11pt,twoside,a4paper]{article}

which instructs L

X to typeset the document as an article with a base

font size of eleven points, and to produce a layout suitable for double sided

printing on A4 paper.

10 Things You Need to Know

1.6.2 Packages

While writing your document, you will probably ﬁnd that there are some

areas where basic L

X cannot solve your problem. If you want to include

graphics, coloured text or source code from a ﬁle into your document, you

need to enhance the capabilities of L

X. Such enhancements are called

packages. Packages are activated with the

\usepackage[options]{package}

command, where package is the name of the package and options is a list

of keywords that trigger special features in the package. The

\usepackage

command goes into the preamble of the document. See section 1.4 for details.

Some packages come with the

X 2

base distribution (See Table 1.3).

Others are provided separately. You may ﬁnd more information on the

packages installed at your site in your Local Guide [

]. The prime source for

information about L

X packages is The L

X Companion [

]. It contains

descriptions on hundreds of packages, along with information of how to write

your own extensions to L

X 2

Modern T

X distributions come with a large number of packages prein-

stalled. If you are working on a Unix system, use the command

texdoc

for

accessing package documentation.

Table 1.1: Document Classes.

article

for articles in scientiﬁc journals, presentations, short reports, pro-

gram documentation, invitations, . . .

proc a class for proceedings based on the article class.

minimal

is as small as it can get. It only sets a page size and a base font. It

is mainly used for debugging purposes.

report

for longer reports containing several chapters, small books, PhD

theses, . . .

book for real books

slides

for slides. The class uses big sans serif letters. You might want to

consider using the Beamer class instead.

1.6 The Layout of the Document 11

Table 1.2: Document Class Options.

10pt, 11pt, 12pt Sets the size of the main font in the document. If no

option is speciﬁed, 10pt is assumed.

a4paper, letterpaper, . . . Deﬁnes the paper size. The default size is

letterpaper. Besides that, a5paper, b5paper, executivepaper,

and legalpaper can be speciﬁed.

fleqn Typesets displayed formulae left-aligned instead of centred.

leqno Places the numbering of formulae on the left hand side instead of

the right.

titlepage, notitlepage

Speciﬁes whether a new page should be started

after the document title or not. The article class does not start a

new page by default, while report and book do.

onecolumn, twocolumn Instructs L

X to typeset the document in one

column or two columns.

twoside, oneside Speciﬁes whether double or single sided output

should be generated. The classes

article

and

report

are single sided

and the book class is double sided by default. Note that this option

concerns the style of the document only. The option

twoside

does not

tell the printer you use that it should actually make a two-sided

printout.

landscape Changes the layout of the document to print in landscape

mode.

openright, openany Makes chapters begin either only on right hand

pages or on the next page available. This does not work with the

article class, as it does not know about chapters. The report class

by default starts chapters on the next page available and the book

class starts them on right hand pages.

12 Things You Need to Know

Table 1.3: Some of the Packages Distributed with L

doc Allows the documentation of L

X programs.

Described in doc.dtx

and in The L

X Companion [3].

exscale Provides scaled versions of the math extension font.

Described in ltexscale.dtx.

fontenc Speciﬁes which font encoding L

X should use.

Described in ltoutenc.dtx.

ifthen Provides commands of the form

‘if. . . then do. . . otherwise do. . . .’

Described in ifthen.dtx and The L

X Companion [3].

latexsym

To access the L

X symbol font, you should use the

latexsym

package. Described in

latexsym.dtx

and in The L

X Companion [

makeidx

Provides commands for producing indexes. Described in section 4.3

and in The L

X Companion [3].

syntonly Processes a document without typesetting it.

inputenc

Allows the speciﬁcation of an input encoding such as ASCII, ISO

Latin-1, ISO Latin-2, 437/850 IBM code pages, Apple Macintosh, Next,

ANSI-Windows or user-deﬁned one. Described in inputenc.dtx.

This ﬁle should be installed on your system, and you should be able to get a

dvi

ﬁle

by typing

latex doc.dtx

in any directory where you have write permission. The same is

true for all the other ﬁles mentioned in this table.

1.7 Files You Might Encounter 13

1.6.3 Page Styles

X supports three predeﬁned header/footer combinations—so-called page

styles. The style parameter of the

\pagestyle{style}

command deﬁnes which one to use. Table 1.4 lists the predeﬁned page styles.

Table 1.4: The Predeﬁned Page Styles of L

plain

prints the page numbers on the bottom of the page, in the middle of

the footer. This is the default page style.

headings

prints the current chapter heading and the page number in the

header on each page, while the footer remains empty. (This is the style

used in this document)

empty sets both the header and the footer to be empty.

It is possible to change the page style of the current page with the

command

\thispagestyle{style}

A description how to create your own headers and footers can be found

in The L

X Companion [3] and in section 4.4 on page 87.

1.7 Files You Might Encounter

When you work with L

X you will soon ﬁnd yourself in a maze of ﬁles

with various extensions and probably no clue. The following list explains the

various ﬁle types you might encounter when working with T

X. Please note

that this table does not claim to be a complete list of extensions, but if you

ﬁnd one missing that you think is important, please drop me a line.

.tex L

X or T

X input ﬁle. Can be compiled with latex.

.sty

X Macro package. Load this into your L

X document using the

\usepackage command.

.dtx

Documented T

X. This is the main distribution format for L

X style

ﬁles. If you process a .dtx ﬁle you get documented macro code of the

X package contained in the .dtx ﬁle.

14 Things You Need to Know

.ins

The installer for the ﬁles contained in the matching .dtx ﬁle. If you

download a L

X package from the net, you will normally get a .dtx

and a .ins ﬁle. Run L

X on the .ins ﬁle to unpack the .dtx ﬁle.

.cls

Class ﬁles deﬁne what your document looks like. They are selected

with the \documentclass command.

.fd Font description ﬁle telling L

X about new fonts.

The following ﬁles are generated when you run L

X on your input ﬁle:

.dvi

Device Independent File. This is the main result of a L

X compile

run. Look at its content with a DVI previewer program or send it to a

printer with dvips or a similar application.

.log

Gives a detailed account of what happened during the last compiler

run.

.toc

Stores all your section headers. It gets read in for the next compiler

run and is used to produce the table of contents.

.lof This is like .toc but for the list of ﬁgures.

.lot And again the same for the list of tables.

.aux

Another ﬁle that transports information from one compiler run to the

next. Among other things, the .aux ﬁle is used to store information

associated with cross-references.

.idx

If your document contains an index. L

X stores all the words that

go into the index in this ﬁle. Process this ﬁle with

makeindex

. Refer

to section 4.3 on page 86 for more information on indexing.

.ind

The processed .idx ﬁle, ready for inclusion into your document on the

next compile cycle.

.ilg Logﬁle telling what makeindex did.

1.8 Big Projects

When working on big documents, you might want to split the input ﬁle into

several parts. L

X has two commands that help you to do that.

\include{ﬁlename}

Use this command in the document body to insert the contents of another ﬁle

named ﬁlename.tex. Note that L

X will start a new page before processing

the material input from ﬁlename.tex.

1.8 Big Projects 15

The second command can be used in the preamble. It allows you to

instruct L

X to only input some of the \included ﬁles.

\includeonly{ﬁlename,ﬁlename,. . . }

After this command is executed in the preamble of the document, only

\include

commands for the ﬁlenames that are listed in the argument of the

\includeonly command will be executed.

The

\include

command starts typesetting the included text on a new

page. This is helpful when you use

\includeonly

, because the page breaks

will not move, even when some include ﬁles are omitted. Sometimes this

might not be desirable. In this case, use the

\input{ﬁlename}

command. It simply includes the ﬁle speciﬁed. No ﬂashy suits, no strings

attached.

To make L

X quickly check your document use the syntonly package.

This makes L

X skim through your document only checking for proper

syntax and usage of the commands, but doesn’t produce any (DVI) output.

As L

X runs faster in this mode you may save yourself valuable time. Usage

is very simple:

\usepackage{syntonly}

\syntaxonly

When you want to produce pages, just comment out the second line (by

adding a percent sign).

Chapter 2

Typesetting Text

After reading the previous chapter, you should know about the basic stuﬀ of

which a

X 2

document is made. In this chapter I will ﬁll in the remaining

structure you will need to know in order to produce real world material.

2.1 The Structure of Text and Language

By Hanspeter Schmid < hanspi@schmid-werren.ch>

The main point of writing a text (some modern DAAC

literature excluded),

is to convey ideas, information, or knowledge to the reader. The reader will

understand the text better if these ideas are well-structured, and will see and

feel this structure much better if the typographical form reﬂects the logical

and semantical structure of the content.

X is diﬀerent from other typesetting systems in that you just have

to tell it the logical and semantical structure of a text. It then derives the

typographical form of the text according to the “rules” given in the document

class ﬁle and in various style ﬁles.

The most important text unit in L

X (and in typography) is the

paragraph. We call it “text unit” because a paragraph is the typographical

form that should reﬂect one coherent thought, or one idea. You will learn in

the following sections how to force line breaks with e.g.

, and paragraph

breaks with e.g. leaving an empty line in the source code. Therefore, if a new

thought begins, a new paragraph should begin, and if not, only line breaks

should be used. If in doubt about paragraph breaks, think about your text

as a conveyor of ideas and thoughts. If you have a paragraph break, but

the old thought continues, it should be removed. If some totally new line of

thought occurs in the same paragraph, then it should be broken.

Most people completely underestimate the importance of well-placed

paragraph breaks. Many people do not even know what the meaning of

Diﬀerent At All Cost, a translation of the Swiss German UVA (Um’s Verrecken Anders).

18 Typesetting Text

a paragraph break is, or, especially in L

X, introduce paragraph breaks

without knowing it. The latter mistake is especially easy to make if equations

are used in the text. Look at the following examples, and ﬁgure out why

sometimes empty lines (paragraph breaks) are used before and after the

equation, and sometimes not. (If you don’t yet understand all commands

well enough to understand these examples, please read this and the following

chapter, and then read this section again.)

% Example 1

\ldots when Einstein introduced his formula

\begin{equation}

e = m \cdot c^2 \; ,

\end{equation}

which is at the same time the most widely known

and the least well understood physical formula.

% Example 2

\ldots from which follows Kirchhoff’s current law:

\begin{equation}

\sum_{k=1}^{n} I_k = 0 \; .

\end{equation}

Kirchhoff’s voltage law can be derived \ldots

% Example 3

\ldots which has several advantages.

\begin{equation}

I_D = I_F - I_R

\end{equation}

is the core of a very different transistor model. \ldots

The next smaller text unit is a sentence. In English texts, there is a

larger space after a period that ends a sentence than after one that ends an

abbreviation. L

X tries to ﬁgure out which one you wanted to have. If

X gets it wrong, you must tell it what you want. This is explained later

in this chapter.

The structuring of text even extends to parts of sentences. Most languages

have very complicated punctuation rules, but in many languages (including

German and English), you will get almost every comma right if you remember

what it represents: a short stop in the ﬂow of language. If you are not sure

about where to put a comma, read the sentence aloud and take a short

2.2 Line Breaking and Page Breaking 19

breath at every comma. If this feels awkward at some place, delete that

comma; if you feel the urge to breathe (or make a short stop) at some other

place, insert a comma.

Finally, the paragraphs of a text should also be structured logically at a

higher level, by putting them into chapters, sections, subsections, and so on.

However, the typographical eﬀect of writing e.g.

\section{The Structure

of Text and Language}

is so obvious that it is almost self-evident how

these high-level structures should be used.

2.2 Line Breaking and Page Breaking

2.2.1 Justiﬁed Paragraphs

Books are often typeset with each line having the same length. L

inserts the necessary line breaks and spaces between words by optimizing

the contents of a whole paragraph. If necessary, it also hyphenates words

that would not ﬁt comfortably on a line. How the paragraphs are typeset

depends on the document class. Normally the ﬁrst line of a paragraph is

indented, and there is no additional space between two paragraphs. Refer to

section 6.3.2 for more information.

In special cases it might be necessary to order L

X to break a line:

\\ or \newline

starts a new line without starting a new paragraph.

\\*

additionally prohibits a page break after the forced line break.

\newpage

starts a new page.

\linebreak[n], \nolinebreak[n], \pagebreak[n], \nopagebreak[n]

suggest places where a break may (or may not) happen. They enable the

author to inﬂuence their actions with the optional argument n, which can

be set to a number between zero and four. By setting n to a value below

4, you leave L

X the option of ignoring your command if the result would

look very bad. Do not confuse these “break” commands with the “new”

commands. Even when you give a “break” command, L

X still tries to even

out the right border of the line and the total length of the page, as described

in the next section; this can lead to unpleasant gaps in your text. If you

20 Typesetting Text

really want to start a “new line” or a “new page”, then use the corresponding

command. Guess their names!

X always tries to produce the best line breaks possible. If it cannot

ﬁnd a way to break the lines in a manner that meets its high standards, it

lets one line stick out on the right of the paragraph. L

X then complains

(“overfull hbox”) while processing the input ﬁle. This happens most often

when L

X cannot ﬁnd a suitable place to hyphenate a word.

Instruct

X to lower its standards a little by giving the

\sloppy

command. It

prevents such over-long lines by increasing the inter-word spacing—even if

the ﬁnal output is not optimal. In this case a warning (“underfull hbox”) is

given to the user. In most such cases the result doesn’t look very good. The

command \fussy brings L

X back to its default behaviour.

2.2.2 Hyphenation

X hyphenates words whenever necessary. If the hyphenation algorithm

does not ﬁnd the correct hyphenation points, remedy the situation by using

the following commands to tell T

X about the exception.

The command

\hyphenation{word list}

causes the words listed in the argument to be hyphenated only at the points

marked by “

”. The argument of the command should only contain words

built from normal letters, or rather signs that are considered to be normal

letters by L

X. The hyphenation hints are stored for the language that

is active when the hyphenation command occurs. This means that if you

place a hyphenation command into the preamble of your document it will

inﬂuence the English language hyphenation. If you place the command

after the

\begin{document}

and you are using some package for national

language support like babel, then the hyphenation hints will be active in the

language activated through babel.

The example below will allow “hyphenation” to be hyphenated as well as

“Hyphenation”, and it prevents “FORTRAN”, “Fortran” and “fortran” from

being hyphenated at all. No special characters or symbols are allowed in the

argument.

Example:

\hyphenation{FORTRAN Hy-phen-a-tion}

Although L

X gives you a warning when that happens (

Overfull \hbox

) and displays

the oﬀending line, such lines are not always easy to ﬁnd. If you use the option

draft

the

\documentclass

command, these lines will be marked with a thick black line on the

right margin.

2.3 Ready-Made Strings 21

The command

inserts a discretionary hyphen into a word. This

also becomes the only point hyphenation is allowed in this word. This

command is especially useful for words containing special characters (e.g.

accented characters), because L

X does not automatically hyphenate words

containing special characters.

I think this is: su\-per\-cal\-%

i\-frag\-i\-lis\-tic\-ex\-pi\-%

al\-i\-do\-cious

I think this is: supercalifragilisticexpiali-

docious

Several words can be kept together on one line with the command

\mbox{text}

It causes its argument to be kept together under all circumstances.

My phone number will change soon.

It will be \mbox{0116 291 2319}.

The parameter

\mbox{\emph{filename}} should

contain the name of the file.

My phone number will change soon. It will

be 0116 291 2319.

The parameter

ﬁlename

should contain the

name of the ﬁle.

\fbox

is similar to

\mbox

, but in addition there will be a visible box

drawn around the content.

2.3 Ready-Made Strings

In some of the examples on the previous pages, you have seen some very

simple L

X commands for typesetting special text strings:

Command Example Description

\today July 18, 2015 Current date

\TeX T

X Your favorite typesetter

\LaTeX L

X The Name of the Game

\LaTeXe L

X 2

The current incarnation

2.4 Special Characters and Symbols

2.4.1 Quotation Marks

You should not use the

for quotation marks as you would on a typewriter.

In publishing there are special opening and closing quotation marks. In L

use two

(grave accent) for opening quotation marks and two

(vertical

quote) for closing quotation marks. For single quotes you use just one of

each.

22 Typesetting Text

‘‘Please press the ‘x’ key.’’

“Please press the ‘x’ key.”

Yes I know the rendering is not ideal, it’s really a back-tick or grave

accent (

) for opening quotes and vertical quote (

) for closing, despite what

the font chosen might suggest.

2.4.2 Dashes and Hyphens

X knows four kinds of dashes. Access three of them with diﬀerent number

of consecutive dashes. The fourth sign is actually not a dash at all—it is the

mathematical minus sign:

daughter-in-law, X-rated\\

pages 13--67\\

yes---or no? \\

$0$, $1$ and $-1$

daughter-in-law, X-rated

pages 13–67

yes—or no?

0, 1 and −1

The names for these dashes are: ‘-’ hyphen, ‘–’ en-dash, ‘—’ em-dash

and ‘−’ minus sign.

2.4.3 Tilde (∼)

A character often seen in web addresses is the tilde. To generate this in

X use

\~{}

but the result (˜) is not really what you want. Try this

instead:

http://www.rich.edu/\~{}bush \\

http://www.clever.edu/$\sim$demo

http://www.rich.edu/˜bush

http://www.clever.edu/∼demo

2.4.4 Slash (/)

In order to typeset a slash between two words, one can simply type e.g.

read/write

, but this makes L

X treat the two words as one. Hyphenation

is disabled for these two words, so there may be ‘overfull’ errors. To over-

come this, use

\slash

. For example type ‘

read\slash write

’ which allows

hyphenation. But normal ‘/’ character may be still used for ratios or units,

e.g. 5 MB/s.

2.4.5 Degree Symbol (◦)

Printing the degree symbol in pure L

2.4 Special Characters and Symbols 23

It’s $-30\,^{\circ}\mathrm{C}$.

I will soon start to

super-conduct.

It’s

−

◦

C. I will soon start to super-

conduct.

The textcomp package makes the degree symbol also available as

\textdegree

or in combination with the C by using the \textcelsius.

30 \textcelsius{} is

86 \textdegree{}F.

30 ℃ is 86 °F.

2.4.6 The Euro Currency Symbol (e)

When writing about money these days, you need the Euro symbol. Many

current fonts contain a Euro symbol. After loading the textcomp package in

the preamble of your document

\usepackage{textcomp}

use the command

\texteuro

to access it.

If your font does not provide its own Euro symbol or if you do not like

the font’s Euro symbol, you have two more choices:

First the eurosym package. It provides the oﬃcial Euro symbol:

\usepackage[oﬃcial]{eurosym}

If you prefer a Euro symbol that matches your font, use the option

gen

in place of the official option.

Table 2.1: A bag full of Euro symbols

LM+textcomp \texteuro € € €

eurosym \euro e e e

[gen]eurosym \euro AC AC AC

24 Typesetting Text

2.4.7 Ellipsis (. . . )

On a typewriter, a comma or a period takes the same amount of space as

any other letter. In book printing, these characters occupy only a little space

and are set very close to the preceding letter. Therefore, entering ‘ellipsis’

by just typing three dots would produce the wrong result. Instead, there is

a special command for these dots. It is called

\ldots (low dots)

Not like this ... but like this:\\

New York, Tokyo, Budapest, \ldots

Not like this ... but like this:

New York, Tokyo, Budapest, . . .

2.4.8 Ligatures

Some letter combinations are typeset not just by setting the diﬀerent letters

one after the other, but by actually using special symbols.

ﬀ ﬁ ﬂ ﬃ. . . instead of ff fi fl ffi . . .

These so-called ligatures can be prohibited by inserting an

\mbox{}

between

the two letters in question. This might be necessary with words built from

two words.

\Large Not shelfful\\

but shelf\mbox{}ful

Not shelﬀul

but shelfful

2.4.9 Accents and Special Characters

X supports the use of accents and special characters from many languages.

Table 2.2 shows all sorts of accents being applied to the letter o. Naturally

other letters work too.

To place an accent on top of an i or a j, its dots have to be removed.

This is accomplished by typing \i and \j.

H\^otel, na\"\i ve, \’el\‘eve,\\

sm\o rrebr\o d, !‘Se\~norita!,\\

Sch\"onbrunner Schlo\ss{}

Stra\ss e

Hôtel, naïve, élève,

smørrebrød, ¡Señorita!,

Schönbrunner Schloß Straße

2.5 International Language Support 25

2.5 International Language Support

When you write documents in languages other than English, there are three

areas where L

X has to be conﬁgured appropriately:

All automatically generated text strings

have to be adapted to the

new language. For many languages, these changes can be accomplished

by using the babel package by Johannes Braams.

X needs to know the hyphenation rules for the new language.

Getting hyphenation rules into L

X is a bit more tricky. It means

rebuilding the format ﬁle with diﬀerent hyphenation patterns enabled.

Your Local Guide [5] should give more information on this.

Language speciﬁc typographic rules. In French for example, there is a

mandatory space before each colon character (:).

If your system is already conﬁgured appropriately, activate the babel

package by adding the command

\usepackage[language]{babel}

after the

\documentclass

command. A list of the languages built into your

X system will be displayed every time the compiler is started. Babel will

automatically activate the appropriate hyphenation rules for the language

you choose. If your L

X format does not support hyphenation in the

language of your choice, babel will still work but will disable hyphenation,

which has quite a negative eﬀect on the appearance of the typeset document.

Table of Contents, List of Figures, . . .

Table 2.2: Accents and Special Characters.

ò \‘o ó \’o ô \^o õ \~o

¯o \=o ˙o \.o ö \"o ç \c c

˘o \u o ˇo \v o ő \H o ¸o \c o

\d o o

\b o oo \t oo

œ \oe Œ \OE æ \ae Æ \AE

å \aa Å \AA

ø \o Ø \O ł \l Ł \L

ı \i  \j ¡ !‘ ¿ ?‘

26 Typesetting Text

Babel also speciﬁes new commands for some languages, which simplify

the input of special characters. The German language, for example, contains

a lot of umlauts (äöü). With babel loaded, enter an ö by typing

instead

of \"o.

If you call babel with multiple languages

\usepackage[languageA,languageB]{babel}

then the last language in the option list will be active (i.e. languageB). Use

the command

\selectlanguage{languageA}

to change the active language.

Most modern computer systems allow you to input letters of national

alphabets directly from the keyboard. In order to handle a variety of input

encodings used for diﬀerent groups of languages and/or on diﬀerent computer

platforms L

X employs the inputenc package:

\usepackage[encoding]{inputenc}

When using this package, you should consider that other people might

not be able to display your input ﬁles on their computer, because they use a

diﬀerent encoding. For example, the German umlaut ä on OS/2 is encoded

as 132, on Unix systems using ISO-LATIN 1 it is encoded as 228, while

in Cyrillic encoding cp1251 for Windows this letter does not exist at all;

therefore you should use this feature with care. The following encodings may

come in handy, depending on the type of system you are working on

Operating encodings

system western Latin Cyrillic

Mac applemac macukr

Unix latin1 koi8-ru

Windows ansinew cp1251

DOS, OS/2 cp850 cp866nav

\usepackage[utf8]{inputenc}

will enable you to create L

X input ﬁles in

utf8

, a multi-byte encoding in

which each character can be encoded in as little as one byte and as many as

four bytes.

To learn more about supported input encodings for Latin-based and Cyrillic-based

languages, read the documentation for

inputenc.dtx

and

cyinpenc.dtx

respectively.

Section 4.6 tells how to produce package documentation.

2.5 International Language Support 27

Since the turn of the Century most Operating Systems are based on

Unicode (Windows XP, MacOS X). Therefore it is recommended to use

utf8

for any new project. The

utf8

encoding used by inputenc only deﬁnes the

characters that are actually provided by the fonts used. If you need more

(non-latin) characters have a look at X

X in section 4.8 a Unicode based

X-engine.

Font encoding is a diﬀerent matter. It deﬁnes at which position inside a

X-font each letter is stored. Multiple input encodings could be mapped

into one font encoding, which reduces the number of required font sets. Font

encodings are handled through fontenc package:

\usepackage[encoding]{fontenc}

where encoding is font encoding. It is possible to load several encodings

simultaneously.

The default L

X font encoding is

OT1

, the encoding of the original

Computer Modern T

X font. It contains only the 128 characters of the 7-bit

ASCII character set. When accented characters are required, T

X creates

them by combining a normal character with an accent. While the resulting

output looks perfect, this approach stops the automatic hyphenation from

working inside words containing accented characters. Besides, some Latin

letters could not be created by combining a normal character with an accent,

to say nothing about letters of non-Latin alphabets, such as Greek or Cyrillic.

To overcome these shortcomings, several 8-bit CM-like font sets were

created. Extended Cork (EC) fonts in

encoding contains letters and

punctuation characters for most of the European languages using Latin

script. The LH font set contains letters necessary to typeset documents

in languages using Cyrillic script. Because of the large number of Cyrillic

glyphs, they are arranged into four font encodings—

T2A

T2B

T2C

, and

The CB bundle contains fonts in

LGR

encoding for the composition of Greek

text.

Improve/enable hyphenation in non-English documents by using these

fonts. Another advantage of using new CM-like fonts is that they provide

fonts of CM families in all weights, shapes, and optically scaled font sizes.

2.5.1 Support for Portuguese

By Demerson Andre Polli <polli@linux.ime.usp.br>

To enable hyphenation and change all automatic text to Portuguese, use the

Find a list of languages supported by each of these encodings in [11].

28 Typesetting Text

Table 2.3: Preamble for Portuguese documents.

\usepackage[portuguese]{babel}

\usepackage[latin1]{inputenc}

\usepackage[T1]{fontenc}

command:

\usepackage[portuguese]{babel}

Or if you are in Brazil, substitute brazilian as the language.

As there are a lot of accents in Portuguese you might want to use

\usepackage[latin1]{inputenc}

to be able to input them correctly as well as

\usepackage[T1]{fontenc}

to get the hyphenation right.

See table 2.3 for the preamble you need to write in the Portuguese

language. Note that the example is for the latin1 input encoding. Modern

systems might be using utf8 instead.

2.5.2 Support for French

By Daniel Flipo <daniel.flipo@univ-lille1.fr>

Some hints for those creating French documents with L

X: load French

language support with the following command:

\usepackage[francais]{babel}

This enables French hyphenation, if you have conﬁgured your L

system accordingly. It also changes all automatic text into French:

\chapter

prints Chapitre,

\today

prints the current date in French and so on. A set

of new commands also becomes available, which allows you to write French

input ﬁles more easily. Check out table 2.4 for inspiration.

You will also notice that the layout of lists changes when switching to the

French language. For more information on what the

francais

option of babel

does and how to customize its behaviour, run L

X on ﬁle

frenchb.dtx

and read the produced ﬁle frenchb.dvi.

2.5 International Language Support 29

Table 2.4: Special commands for French.

\og guillemets \fg{} « guillemets »

M\up{me}, D\up{r} M

, D

1\ier{}, 1\iere{}, 1\ieres{} 1

, 1

res

2\ieme{} 4\iemes{} 2

\No 1, \no 2 N

1, n

20~\degres C, 45\degres 20 °C, 45°

\bsc{M. Durand} M. Durand

\nombre{1234,56789} 1 234,567 89

Recent versions of frenchb rely on numprint to implement the

\nombre

command.

2.5.3 Support for German

Some hints for those creating German documents with L

X: load German

language support with the following command:

\usepackage[german]{babel}

This enables German hyphenation, if you have conﬁgured your L

system accordingly. It also changes all automatic text into German. Eg.

“Chapter” becomes “Kapitel.” A set of new commands also becomes available,

which allows you to write German input ﬁles more quickly even when you

don’t use the inputenc package. Check out table 2.5 for inspiration. With

inputenc, all this becomes moot, but your text also is locked in a particular

encoding world.

In German books you often ﬁnd French quotation marks («guillemets»).

German typesetters, however, use them diﬀerently. A quote in a German

book would look like »this«. In the German speaking part of Switzerland,

typesetters use «guillemets» the same way the French do.

A major problem arises from the use of commands like

\flq

: If you use

the OT1 font (which is the default font) the guillemets will look like the math

symbol “



”, which turns a typesetter’s stomach. T1 encoded fonts, on the

other hand, do contain the required symbols. So if you are using this type of

quote, make sure you use the T1 encoding. (\usepackage[T1]{fontenc})

30 Typesetting Text

Table 2.5: German Special Characters.

"a ä "s ß

"‘ „ "’ “

"< or \flqq « "> or \frqq »

\flq ‹ \frq ›

\dq "

2.5.4 Support for Korean

To process Hangul

characters or prepare a document written in Korean

using L

X, put the following code in the preamble of the document.

\usepackage{kotex}

A document containing the declaration above will have to be processed

by pdfL

X, X

X, or LuaL

X. Make sure that the input ﬁle written in

Hangul is encoded in Unicode UTF-8. The package called ko.T

is under

continuous development by the Korean T

X Users Group

and the Korean

X Society.

Many people use this package to create Korean documents

for their everyday needs. ko.T

X has been available on CTAN since 2014. It

is included T

X Live, MiKT

X and other modern T

X distributions. So in

all likelyhood you can start working right away without installing any extra

packages.

ko.T

X does not use the

babel

package. Many functions related to

Korean can be activated using the options and conﬁguration commands

provided by the kotex package. If you want to compose a real world Korean

document, you are advised to consult the package documentation (These

documents are written in Korean).

With ko.T

X, you also get oblivoir, a memoir-based document class,

tailored for Korean document preparation. So your Korean document would

Written by Karnes Kim

karnes@ktug.org

and Kihwang Lee

leekh@ktug.org

behalf of the Korean T

X Users Group and the Korean T

X Society.

Hangul is the name of the Korean writing system. Refer to

http://en.wikipedia.

org/wiki/Hangul for more information.

Reads “Korean T

X”. ko.T

X is the name of a collection of packages including

cjk-ko

kotex-utf, xetexko, and luatexko.

http://ktug.org

http://ktug.kr

2.5 International Language Support 31

Table 2.6: Preamble for Greek documents.

\usepackage[english,greek]{babel}

\usepackage[iso-8859-7]{inputenc}

start like this:

\documentclass{oblivoir}

To generate an index for a Korean document, execute

komkindex

instead

makeindex

. It is a version of the

makeindex

utility modiﬁed for Korean

processing. For lexicographical sorting of the Korean index items, you can

use index style kotex.ist provided by ko.T

X as follows:

komkindex -s kotex foo.idx

You can also use

xindy

for index generation as the Korean module for

xindy is included in T

X Live.

There is another Korean/Hangul typesetting package called CJK. As

the name of the package suggests, it has facilities for typesetting Chinese,

Japanese, and Korean characters. It supports multiple encodings of the CJK

characters. The following is a simple example of typesetting UTF-8 encoded

Hangul using CJK package. It is useful when you submit a manuscript

to some academic journals that allow typesetting author names in native

languages.

\usepackage{CJK}

\begin{CJK}{UTF8}{}

\CJKfamily{nanummj}

...

\end{CJK}

2.5.5 Writing in Greek

By Nikolaos Pothitos <pothitos@di.uoa.gr>

See table 2.6 for the preamble you need to write in the Greek language. This

preamble enables hyphenation and changes all automatic text to Greek.

If you select the

utf8x

option for the package

inputenc

, L

X will understand Greek

and polytonic Greek Unicode characters.

32 Typesetting Text

A set of new commands also becomes available, which allows you to

write Greek input ﬁles more easily. In order to temporarily switch to

English and vice versa, one can use the commands

\textlatin{

english text

}

and

\textgreek{

greek text

}

that both take one argument which is then

typeset using the requested font encoding. Otherwise use the command

\selectlanguage{...}

described in a previous section. Check out table 2.7

for some Greek punctuation characters. Use \euro for the Euro symbol.

Table 2.7: Greek Special Characters.

; · ? ;

(( « )) »

‘‘ ‘ ’’ ’

2.5.6 Support for Cyrillic

By Maksym Polyakov <polyama@myrealbox.com>

Version 3.7h of babel includes support for the

T2*

encodings and for typeset-

ting Bulgarian, Russian and Ukrainian texts using Cyrillic letters.

Support for Cyrillic is based on standard L

X mechanisms plus the

fontenc and inputenc packages. But, if you are going to use Cyrillics in math

mode, you need to load mathtext package before fontenc:

\usepackage{mathtext}

\usepackage[T1,T2A]{fontenc}

\usepackage[koi8-ru]{inputenc}

\usepackage[english,bulgarian,russian,ukranian]{babel}

Generally, babel will authomatically choose the default font encoding, for

the above three languages this is

T2A

. However, documents are not restricted

to a single font encoding. For multi-lingual documents using Cyrillic and

Latin-based languages it makes sense to include Latin font encoding explicitly.

babel will take care of switching to the appropriate font encoding when a

diﬀerent language is selected within the document.

In addition to enabling hyphenations, translating automatically gener-

ated text strings, and activating some language speciﬁc typographic rules

(like

\frenchspacing

), babel provides some commands allowing typesetting

according to the standards of Bulgarian, Russian, or Ukrainian languages.

For all three languages, language speciﬁc punctuation is provided: The

Cyrillic dash for the text (it is little narrower than Latin dash and surrounded

If you use A

S-L

X packages, load them before fontenc and babel as well.

2.5 International Language Support 33

by tiny spaces), a dash for direct speech, quotes, and commands to facilitate

hyphenation, see Table 2.8.

Table 2.8: The extra deﬁnitions made by Bulgarian, Russian, and Ukrainian

options of babel

"| disable ligature at this position.

an explicit hyphen sign, allowing hyphenation in the rest of the word.

"--- Cyrillic emdash in plain text.

"--~ Cyrillic emdash in compound names (surnames).

"--* Cyrillic emdash for denoting direct speech.

, but producing no hyphen sign (for compound words with

hyphen, e.g. x-""y or some other signs as “disable/enable”).

"~ for a compound word mark without a breakpoint.

for a compound word mark with a breakpoint, allowing hyphenation

in the composing words.

", thinspace for initials with a breakpoint in following surname.

"‘ for German left double quotes (looks like ,,).

"’ for German right double quotes (looks like “).

"< for French left double quotes (looks like <<).

"> for French right double quotes (looks like >>).

The Russian and Ukrainian options of babel deﬁne the commands

\Asbuk

and

\asbuk

, which act like

\Alph

and

\alph

, but produce capital and

small letters of Russian or Ukrainian alphabets (whichever is the active

language of the document). The Bulgarian option of babel provides the

commands

\enumBul

and

\enumLat

(

\enumEng

), which make

\Alph

and

\alph

produce letters of either Bulgarian or Latin (English) alphabets. The

default behaviour of

\Alph

and

\alph

for the Bulgarian language option is

to produce letters from the Bulgarian alphabet.

2.5.7 Support for Mongolian

To use L

X for typesetting Mongolian you have a choice between two

packages: Multilingual Babel and MonT

X by Oliver Corﬀ.

MonT

X includes support for both Cyrillic and traditional Mongolian

Script. In order to access the commands of MonT

X, add:

\usepackage[language,encoding]{mls}

to the preamble. Choose the language option xalx to generate captions

and dates in Modern Mongolian. To write a complete document in the

the commands for turning counters into a, b, c, . . .

34 Typesetting Text

traditional Mongolian script you have to choose bicig for the language option.

The document language option bicig enables the “Simpliﬁed Transliteration”

input method.

Enable and disable Latin Transliteration Mode with

\SetDocumentEncodingLMC

and

\SetDocumentEncodingNeutral

More information about MonT

X is available from

CTAN://language/

mongolian/montex/doc.

Mongolian Cyrillic script is supported by babel. Activate Mongolian

language support with the following commands:

\usepackage[T2A]{fontenc}

\usepackage[mn]{inputenc}

\usepackage[mongolian]{babel}

where

is the

cp1251

input encoding. For a more modern approach invoke

utf8 instead.

2.5.8 The Unicode option

By Axel Kielhorn <A.Kielhorn@web.de>

Unicode is the way to go if you want to include several languages in one

document, especially when these languages are not using the latin script.

There are two T

X-engines that are capable of processing Unicode input:

was developed for MacOS X but is now available for all architectures.

It was ﬁrst included into TexLive 2007.

LuaT

is the successor of pdfT

X. It was ﬁrst included into TexLive 2008.

The following describes X

X as distributed with TexLive 2010.

Quickstart

To convert an existing L

X ﬁle to X

X the following needs to be done:

1. Save the ﬁle as UTF-8

2.5 International Language Support 35

2. Remove

\usepackage{inputenc}

\usepackage{fontenc}

\usepackage{textcomp}

from the preamble.

3. Change

\usepackage[languageA]{babel}

\usepackage{polyglossia}

\setdefaultlanguage[babelshorthands]{languageA}

4. Add

\usepackage[Ligatures=TeX]{fontspec}

to the preamble.

The package polyglossia[

] is a replacement for babel. It takes care of

the hyphenation patterns and automatically generated text strings. The

option

babelshorthands

enables babel compatible shorthands for german

and catalan.

The package fontspec[

] handles font loading for X

X and

LuaT

The default font is Latin Modern Roman. It is a little known fact that some

X command are ligatures deﬁned in the Computer Modern fonts. If you

want to use them with a non-T

X font, you have to fake them. The option

Ligatures=TeX deﬁnes the following ligatures:

-- –

--- —

’’ ”

‘‘ “

!‘ ¡

?‘ ¿

,, „

<< «

>> »

36 Typesetting Text

It’s all Γρκ to me

So far there has been no advantage to using a Unicode T

X engine. This

changes when we leave the Latin script and move to a more interesting

language like Greek or Russian. With a Unicode based system, you can

simply

enter the characters in your editor and T

X will understand them.

Writing in diﬀerent languages is easy, just specify the languages in the

preamble:

\setdefaultlanguage{english}

\setotherlanguage[babelshorthands]{german}

To write a paragraph in German, you can use the German environment:

English text.

\begin{german}

Deutscher Text.

\end{german}

More English text.

If you just need a word in a foreign language you can use the

\text

language

command:

Englisch text. \textgerman{Gesundheit} is actually a German word.

This may look unnecessary since the only advantage is a correct hyphen-

ation, but when the second language is a little bit more exotic it will be

worth the eﬀort.

Sometimes the font used in the main document does not contain glyphs

that are required in the second language

. The solution is to deﬁne a font

that will be used for that language. Whenever a new language is activated,

polyglossia will ﬁrst check whether a font has been deﬁned for that language.

\newfontfamily\russianfont[Script=Cyrillic,(...)]{(font)}

Now you can write

\textrussian{Pravda} is a russian newspaper.

Since this document is written in Latin1-encoding, I cannot show the

actual Cyrillic letters.

The package xgreek[

] oﬀers support for writing either ancient or modern

(monotonic or polytonic) greek.

For small values of simple.

Latin Modern does not contain Cyrillic letters

2.5 International Language Support 37

Right to Left (RTL) languages.

Some languages are written left to right, others are written right to left(RTL).

polyglossia needs the bidi[

] package

in order to support RTL languages.

The bidi package should be the last package you load, even after hyperref

which is usually the last package. (Since polyglossia loads bidi this means

that polyglossia should be the last package loaded.)

The package xepersian[

] oﬀers support for the Persian language. It

supplies Persian L

X-commands that allows you to enter commands like

\section

in Persian, which makes this really attractive to native speakers.

xepersian is the only package that supports kashida with X

X. A package

for Syriac which uses a similar algorithm is under development.

The IranNastaliq font provided by the SCICT

is available at their

website http://www.scict.ir/Portal/Home/Default.aspx.

The arabxetex[

] package supports several languages with an Arabic

script:

• arab (Arabic)

• persian

• urdu

• sindhi

• pashto

• ottoman (turk)

• kurdish

• kashmiri

• malay (jawi)

• uighur

It oﬀers a font mapping that enables X

X to process input using the

ArabT

X ASCII transcription.

Fonts that support several Arabic laguages are oﬀered by the IRMUG

at http://wiki.irmug.org/index.php/X_Series_2.

There is no package available for Hebrew because none is needed. The

Hebrew support in polyglossia should be suﬃcient. But you do need a

suitable font with real Unicode Hebrew. SBL Hebrew is free for non-

commercial use and available at

http://www.sbl-site.org/educational/

bidi does not support LuaT

Supreme Council of Information and Communication Technology

Iranian Mac User Group

38 Typesetting Text

biblicalfonts.aspx

. Another font available under the Open Font License

is Ezra SIL, available at

http://www.sil.org/computing/catalog/show_

software.asp?id=76.

Remember to select the correct script:

\newfontfamily\hebrewfont[Script=Hebrew]{SBL Hebrew}

\newfontfamily\hebrewfont[Script=Hebrew]{Ezra SIL}

Chinese, Japanese and Korean (CJK)

The package xeCJK[

] takes care of font selection and punctuation for these

languages.

2.6 The Space Between Words

To get a straight right margin in the output, L

X inserts varying amounts

of space between the words. It inserts slightly more space at the end of a

sentence, as this makes the text more readable. L

X assumes that sentences

end with periods, question marks or exclamation marks. If a period follows

an uppercase letter, this is not taken as a sentence ending, since periods after

uppercase letters normally occur in abbreviations.

Any exception from these assumptions has to be speciﬁed by the author.

A backslash in front of a space generates a space that will not be enlarged. A

tilde ‘

’ character generates a space that cannot be enlarged and additionally

prohibits a line break. The command

in front of a period speciﬁes that

this period terminates a sentence even when it follows an uppercase letter.

Mr.~Smith was happy to see her\\

cf.~Fig.~5\\

I like BASIC\@. What about you?

Mr. Smith was happy to see her

cf. Fig. 5

I like BASIC. What about you?

The additional space after periods can be disabled with the command

\frenchspacing

which tells L

X not to insert more space after a period than after an

ordinary character. This is very common in non-English languages, except

bibliographies. If you use

\frenchspacing

, the command

is not necessary.

2.7 Titles, Chapters, and Sections 39

2.7 Titles, Chapters, and Sections

To help the reader ﬁnd his or her way through your work, you should divide

it into chapters, sections, and subsections. L

X supports this with special

commands that take the section title as their argument. It is up to you to

use them in the correct order.

The following sectioning commands are available for the

article

class:

\section{...}

\subsection{...}

\subsubsection{...}

\paragraph{...}

\subparagraph{...}

If you want to split your document into parts without inﬂuencing the

section or chapter numbering use

\part{...}

When you work with the

report

book

class, an additional top-level

sectioning command becomes available

\chapter{...}

As the

article

class does not know about chapters, it is quite easy

to add articles as chapters to a book. The spacing between sections, the

numbering and the font size of the titles will be set automatically by L

Two of the sectioning commands are a bit special:

•

The

\part

command does not inﬂuence the numbering sequence of

chapters.

•

The

\appendix

command does not take an argument. It just changes

the chapter numbering to letters.

X creates a table of contents by taking the section headings and page

numbers from the last compile cycle of the document. The command

\tableofcontents

expands to a table of contents at the place it is issued. A new document

has to be compiled (“L

Xed”) twice to get a correct table of contents.

Sometimes it might be necessary to compile the document a third time.

X will tell you when this is necessary.

For the article style it changes the section numbering.

40 Typesetting Text

All sectioning commands listed above also exist as “starred” versions. A

“starred” version of a command is built by adding a star

after the command

name. This generates section headings that do not show up in the table

of contents and are not numbered. The command

\section{Help}

, for

example, would become \section*{Help}.

Normally the section headings show up in the table of contents exactly

as they are entered in the text. Sometimes this is not possible, because the

heading is too long to ﬁt into the table of contents. The entry for the table

of contents can then be speciﬁed as an optional argument in front of the

actual heading.

\chapter[Title for the table of contents]{A long

and especially boring title, shown in the text}

The title of the whole document is generated by issuing a

\maketitle

command. The contents of the title have to be deﬁned by the commands

\title{...}, \author{...} and optionally \date{...}

before calling

\maketitle

. In the argument to

\author

, you can supply

several names separated by \and commands.

An example of some of the commands mentioned above can be found in

Figure 1.2 on page 8.

Apart from the sectioning commands explained above,

X 2

intro-

duced three additional commands for use with the

book

class. They are

useful for dividing your publication. The commands alter chapter headings

and page numbering to work as you would expect in a book:

\frontmatter

should be the very ﬁrst command after the start of the

document body (

\begin{document}

). It will switch page numbering

to Roman numerals and sections will be non-enumerated as if you were

using the starred sectioning commands (eg

\chapter*{Preface}

) but

the sections will still show up in the table of contents.

\mainmatter

comes right before the ﬁrst chapter of the book. It turns on

Arabic page numbering and restarts the page counter.

\appendix

marks the start of additional material in your book. After this

command chapters will be numbered with letters.

\backmatter

should be inserted before the very last items in your book,

such as the bibliography and the index. In the standard document

classes, this has no visual eﬀect.

2.8 Cross References 41

2.8 Cross References

In books, reports and articles, there are often cross-references to ﬁgures,

tables and special segments of text. L

X provides the following commands

for cross referencing

\label{marker}, \ref{marker} and \pageref{marker}

where marker is an identiﬁer chosen by the user. L

X replaces

\ref

the number of the section, subsection, ﬁgure, table, or theorem after which

the corresponding

\label

command was issued.

\pageref

prints the page

number of the page where the

\label

command occurred.

As with section

titles and page numbers for the table of contents, the numbers from the

previous compile cycle are used.

A reference to this subsection

\label{sec:this} looks like:

‘‘see section~\ref{sec:this} on

page~\pageref{sec:this}.’’

A reference to this subsection looks like:

“see section 2.8 on page 41.”

2.9 Footnotes

With the command

\footnote{footnote text}

a footnote is printed at the foot of the current page. Footnotes should always

be put

after the word or sentence they refer to. Footnotes referring to a

sentence or part of it should therefore be put after the comma or period.

Footnotes\footnote{This is

a footnote.} are often used

by people using \LaTeX.

Footnotes

are often used by people using

This is a footnote.

Note that these commands are not aware of what they refer to.

\label

just saves the

last automatically generated number.

“put” is one of the most common English words.

Note that footnotes distract the reader from the main body of your document. After

all, everybody reads the footnotes—we are a curious species, so why not just integrate

everything you want to say into the body of the document?

A guidepost doesn’t necessarily go where it’s pointing to :-).

42 Typesetting Text

2.10 Emphasized Words

If a text is typed using a typewriter, important words are

emphasized by

underlining them.

\underline{text}

In printed books, however, words are emphasized by typesetting them in

an italic font. As an author you shouldn’t care either way. The important

bit is, to tell L

X that a particular bit of text is important and should be

emphasized. Hence the command

\emph{text}

to emphasize text. What the command actually does with its argument

depends on the context:

\emph{If you use

emphasizing inside a piece

of emphasized text, then

\LaTeX{} uses the

\emph{normal} font for

emphasizing.}

If you use emphasizing inside a piece of

emphasized text, then L

X uses the nor-

mal font for emphasizing.

If you want control over font and font size, section 6.2 on page 123 might

provide some inspiration.

2.11 Environments

\begin{environment} text \end{environment}

Where environment is the name of the environment. Environments can be

nested within each other as long as the correct nesting order is maintained.

\begin{aaa}...\begin{bbb}...\end{bbb}...\end{aaa}

In the following sections all important environments are explained.

2.11.1 Itemize, Enumerate, and Description

The

itemize

environment is suitable for simple lists, the

enumerate

environ-

ment for enumerated lists, and the

description

environment for descriptions.

2.11 Environments 43

\flushleft

\begin{enumerate}

\item You can nest the list

environments to your taste:

\begin{itemize}

\item But it might start to

look silly.

\item[-] With a dash.

\end{itemize}

\item Therefore remember:

\begin{description}

\item[Stupid] things will not

become smart because they are

in a list.

\item[Smart] things, though,

can be presented beautifully

in a list.

\end{description}

\end{enumerate}

1. You can nest the list environments

to your taste:

• But it might start to look

silly.

- With a dash.

2. Therefore remember:

Stupid things will not become

smart because they are in a

list.

Smart things, though, can be

presented beautifully in a list.

2.11.2 Flushleft, Flushright, and Center

The environments

flushleft

and

flushright

generate paragraphs that

are either left- or right-aligned. The

center

environment generates centred

text. If you do not issue \\ to specify line breaks, L

X will automatically

determine line breaks.

\begin{flushleft}

This text is\\ left-aligned.

\LaTeX{} is not trying to make

each line the same length.

\end{flushleft}

This text is

left-aligned. L

X is not trying to make

each line the same length.

\begin{flushright}

This text is right-\\aligned.

\LaTeX{} is not trying to make

each line the same length.

\end{flushright}

This text is right-

aligned. L

X is not trying to make each

line the same length.

\begin{center}

At the centre\\of the earth

\end{center}

At the centre

of the earth

44 Typesetting Text

2.11.3 Quote, Quotation, and Verse

The

quote

environment is useful for quotes, important phrases and examples.

A typographical rule of thumb

for the line length is:

\begin{quote}

On average, no line should

be longer than 66 characters.

\end{quote}

This is why \LaTeX{} pages have

such large borders by default

and also why multicolumn print

is used in newspapers.

A typographical rule of thumb for the line

length is:

On average, no line should be

longer than 66 characters.

This is why L

X pages have such large

borders by default and also why multicol-

umn print is used in newspapers.

There are two similar environments: the

quotation

and the

verse

envi-

ronments. The

quotation

environment is useful for longer quotes going over

several paragraphs, because it indents the ﬁrst line of each paragraph. The

verse

environment is useful for poems where the line breaks are important.

The lines are separated by issuing a

at the end of a line and an empty

line after each verse.

I know only one English poem by

heart. It is about Humpty Dumpty.

\begin{flushleft}

\begin{verse}

Humpty Dumpty sat on a wall:\\

Humpty Dumpty had a great fall.\\

All the King’s horses and all

the King’s men\\

Couldn’t put Humpty together

again.

\end{verse}

\end{flushleft}

I know only one English poem by heart. It

is about Humpty Dumpty.

Humpty Dumpty sat on a

wall:

Humpty Dumpty had a great

fall.

All the King’s horses and all

the King’s men

Couldn’t put Humpty

together again.

2.11.4 Abstract

In scientiﬁc publications it is customary to start with an abstract which gives

the reader a quick overview of what to expect. L

X provides the

abstract

environment for this purpose. Normally

abstract

is used in documents

typeset with the article document class.

\begin{abstract}

The abstract abstract.

\end{abstract}

The abstract abstract.

2.11 Environments 45

2.11.5 Printing Verbatim

Text that is enclosed between

\begin{verbatim}

and

\end{verbatim}

will

be directly printed, as if typed on a typewriter, with all line breaks and

spaces, without any L

X command being executed.

Within a paragraph, similar behavior can be accessed with

\verb+text+

The

is just an example of a delimiter character. Use any character except

letters,

or space. Many L

X examples in this booklet are typeset with

this command.

The \verb|\ldots| command \ldots

\begin{verbatim}

10 PRINT "HELLO WORLD ";

20 GOTO 10

\end{verbatim}

The \ldots command . . .

10 PRINT "HELLO WORLD ";

20 GOTO 10

\begin{verbatim*}

the starred version of

the verbatim

environment emphasizes

the spaces in the text

\end{verbatim*}

the␣starred␣version␣of

the␣␣␣␣␣␣verbatim

environment␣emphasizes

the␣spaces␣␣␣in␣the␣text

The \verb command can be used in a similar fashion with a star:

\verb*|like this :-) |

like␣␣␣this␣:-)␣

The

verbatim

environment and the

\verb

command may not be used

within parameters of other commands.

2.11.6 Tabular

The

tabular

environment can be used to typeset beautiful tables with

optional horizontal and vertical lines. L

X determines the width of the

columns automatically.

The table spec argument of the

\begin{tabular}[pos]{table spec}

command deﬁnes the format of the table. Use an

for a column of left-

aligned text, r for right-aligned text, and c for centred text; p{width }

46 Typesetting Text

for a column containing justiﬁed text with line breaks, and

for a vertical

line.

If the text in a column is too wide for the page, L

X won’t automatically

wrap it. Using

p{width }

you can deﬁne a special type of column which

will wrap-around the text as in a normal paragraph.

The pos argument speciﬁes the vertical position of the table relative to

the baseline of the surrounding text. Use one of the letters

and

specify table alignment at the top, bottom or centre.

Within a

tabular

environment,

jumps to the next column,

starts

a new line and

\hline

inserts a horizontal line. Add partial lines by using

\cline{i-j}

, where

and

are the column numbers the line should extend

over.

\begin{tabular}{|r|l|}

\hline

7C0 & hexadecimal \\

3700 & octal \\ \cline{2-2}

11111000000 & binary \\

\hline \hline

1984 & decimal \\

\hline

\end{tabular}

7C0 hexadecimal

3700 octal

11111000000 binary

1984 decimal

\begin{tabular}{|p{4.7cm}|}

\hline

Welcome to Boxy’s paragraph.

We sincerely hope you’ll

all enjoy the show.\\

\hline

\end{tabular}

Welcome to Boxy’s paragraph.

We sincerely hope you’ll all en-

joy the show.

The column separator can be speciﬁed with the

@{...}

construct. This

command kills the inter-column space and replaces it with whatever is

between the curly braces. One common use for this command is explained

below in the decimal alignment problem. Another possible application is to

suppress leading space in a table with @{} .

\begin{tabular}{@{} l @{}}

\hline

no leading space\\

\hline

\end{tabular}

no leading space

2.11 Environments 47

\begin{tabular}{l}

\hline

leading space left and right\\

\hline

\end{tabular}

leading space left and right

Since there is no built-in way to align numeric columns to a decimal

point,

we can “cheat” and do it by using two columns: a right-aligned integer

and a left-aligned fraction. The

@{.}

command in the

\begin{tabular}

line replaces the normal inter-column spacing with just a “.”, giving the

appearance of a single, decimal-point-justiﬁed column. Don’t forget to

replace the decimal point in your numbers with a column separator (

A column label can be placed above our numeric “column” by using the

\multicolumn command.

\begin{tabular}{c r @{.} l}

Pi expression &

\multicolumn{2}{c}{Value} \\

\hline

$\pi$ & 3&1416 \\

$\pi^{\pi}$ & 36&46 \\

$(\pi^{\pi})^{\pi}$ & 80662&7 \\

\end{tabular}

Pi expression Value

π 3.1416

36.46

(π

)

80662.7

\begin{tabular}{|c|c|}

\hline

\multicolumn{2}{|c|}{Ene} \\

\hline

Mene & Muh! \\

\hline

\end{tabular}

Ene

Mene Muh!

Material typeset with the tabular environment always stays together on

one page. If you want to typeset long tables, you might want to use the

longtable environments.

Sometimes the default L

X tables do feel a bit cramped. So you

may want to give them a bit more breathing space by setting a higher

\arraystretch and \tabcolsep value.

If the ‘tools’ bundle is installed on your system, have a look at the dcolumn package.

48 Typesetting Text

\begin{tabular}{|l|}

\hline

These lines\\\hline

are tight\\\hline

\end{tabular}

{\renewcommand{\arraystretch}{1.5}

\renewcommand{\tabcolsep}{0.2cm}

\begin{tabular}{|l|}

\hline

less cramped\\\hline

table layout\\\hline

\end{tabular}}

These lines

are tight

less cramped

table layout

If you just want to grow the height of a single row in your table add an

invisible vertical bar

. Use a zero width \rule to implement this trick.

\begin{tabular}{|c|}

\hline

\rule{1pt}{4ex}Pitprop \ldots\\

\hline

\rule{0pt}{4ex}Strut\\

\hline

\end{tabular}

Pitprop . . .

Strut

The

and

in the example above are T

X units. Read more on units

in table 6.5 on page 130.

A number of extra commands, enhancing the tabular environment are

available in the booktabs package. It makes the creation of professional

looking tables with proper spacing quite a bit simpler.

2.12 Floating Bodies

Today most publications contain a lot of ﬁgures and tables. These elements

need special treatment, because they cannot be broken across pages. One

method would be to start a new page every time a ﬁgure or a table is too

large to ﬁt on the present page. This approach would leave pages partially

empty, which looks very bad.

The solution to this problem is to ‘ﬂoat’ any ﬁgure or table that does not

ﬁt on the current page to a later page, while ﬁlling the current page with

body text. L

X oﬀers two environments for ﬂoating bodies; one for tables

and one for ﬁgures. To take full advantage of these two environments it is

important to understand approximately how L

X handles ﬂoats internally.

Otherwise ﬂoats may become a major source of frustration, because L

never puts them where you want them to be.

In professional typesetting, this is called a strut.

2.12 Floating Bodies 49

Let’s ﬁrst have a look at the commands L

X supplies for ﬂoats:

Any material enclosed in a

figure

table

environment will be treated

as ﬂoating matter. Both ﬂoat environments support an optional parameter

\begin{figure}[placement speciﬁer] or \begin{table}[. . . ]

called the placement speciﬁer. This parameter is used to tell L

X about the

locations to which the ﬂoat is allowed to be moved. A placement speciﬁer is

constructed by building a string of ﬂoat-placing permissions. See Table 2.9.

For example, a table could be started with the following line

\begin{table}[!hbp]

The placement speciﬁer

[!hbp]

allows L

X to place the table right here

(

) or at the bottom (

) of some page or on a special ﬂoats page (

), and all

this even if it does not look that good (

). If no placement speciﬁer is given,

the standard classes assume [tbp].

X will place every ﬂoat it encounters according to the placement

speciﬁer supplied by the author. If a ﬂoat cannot be placed on the current

page it is deferred either to the ﬁgures queue or the tables queue.

When

a new page is started, L

X ﬁrst checks if it is possible to ﬁll a special

‘ﬂoat’ page with ﬂoats from the queues. If this is not possible, the ﬁrst ﬂoat

on each queue is treated as if it had just occurred in the text: L

X tries

again to place it according to its respective placement speciﬁers (except ‘h,’

which is no longer possible). Any new ﬂoats occurring in the text get placed

into the appropriate queues. L

X strictly maintains the original order of

appearance for each type of ﬂoat. That’s why a ﬁgure that cannot be placed

pushes all further ﬁgures to the end of the document. Therefore:

These are FIFO—‘ﬁrst in ﬁrst out’—queues!

Table 2.9: Float Placing Permissions.

Spec Permission to place the ﬂoat . . .

here at the very place in the text where it oc-

curred. This is useful mainly for small ﬂoats.

t at the top of a page

b at the bottom of a page

p on a special page containing only ﬂoats.

without considering most of the internal parame-

ters

, which could otherwhise stop this ﬂoat from

being placed.

Such as the maximum number of ﬂoats allowed on one page.

50 Typesetting Text

If L

X is not placing the ﬂoats as you expected, it is often only

one ﬂoat jamming one of the two ﬂoat queues.

While it is possible to give L

X single-location placement speciﬁers,

this causes problems. If the ﬂoat does not ﬁt in the location speciﬁed it

becomes stuck, blocking subsequent ﬂoats. In particular, you should never,

ever use the [h] option—it is so bad that in more recent versions of L

X, it

is automatically replaced by [ht].

Having explained the diﬃcult bit, there are some more things to mention

about the table and figure environments. Use the

\caption{caption text}

command to deﬁne a caption for the ﬂoat. A running number and the string

“Figure” or “Table” will be added by L

The two commands

\listoffigures and \listoftables

operate analogously to the

\tableofcontents

command, printing a list of

ﬁgures or tables, respectively. These lists will display the whole caption, so if

you tend to use long captions you must have a shorter version of the caption

for the lists. This is accomplished by entering the short version in brackets

after the \caption command.

\caption[Short]{LLLLLoooooonnnnnggggg}

Use

\label

and

\ref

to create a reference to a ﬂoat within your text.

Note that the

\label

command must come after the

\caption

command

since you want it to reference the number of the caption.

The following example draws a square and inserts it into the document.

You could use this if you wanted to reserve space for images you are going

to paste into the ﬁnished document.

Figure~\ref{white} is an example of Pop-Art.

\begin{figure}[!hbtp]

\makebox[\textwidth]{\framebox[5cm]{\rule{0pt}{5cm}}}

\caption{Five by Five in Centimetres.\label{white}}

\end{figure}

In the example above, L

X will try really hard (

) to place the ﬁgure right

here (

If this is not possible, it tries to place the ﬁgure at the bottom (

)

of the page. Failing to place the ﬁgure on the current page, it determines

assuming the ﬁgure queue is empty.

2.13 Protecting Fragile Commands 51

whether it is possible to create a ﬂoat page containing this ﬁgure and maybe

some tables from the tables queue. If there is not enough material for a

special ﬂoat page, L

X starts a new page, and once more treats the ﬁgure

as if it had just occurred in the text.

Under certain circumstances it might be necessary to use the

\clearpage or even the \cleardoublepage

command. It orders L

X to immediately place all ﬂoats remaining in the

queues and then start a new page.

\cleardoublepage

even goes to a new

right-hand page.

You will learn how to include PostScript drawings in your

X 2

documents later in this introduction.

2.13 Protecting Fragile Commands

Text given as arguments of commands like

\caption

\section

may show

up more than once in the document (e.g. in the table of contents as well as

in the body of the document). Some commands will break when used in the

argument of

\section

-like commands. Compilation of your document will

fail. These commands are called fragile commands—for example,

\footnote

\phantom

. These fragile commands need protection (don’t we all?).

Protect them by putting the

\protect

command in front of them. Now they

will work properly even when used in moving arguments.

\protect

only aﬀects the next command, not even to its arguments. In

most cases a superﬂuous \protect won’t hurt.

\section{I am considerate

\protect\footnote{and protect my footnotes}}

Chapter 3

Typesetting Mathematical

Formulae

Now you are ready! In this chapter, we will attack the main strength of T

mathematical typesetting. But be warned, this chapter only scratches the surface.

While the things explained here are suﬃcient for many people, don’t despair

if you can’t ﬁnd a solution to your mathematical typesetting needs here. It is

highly likely that your problem is addressed in A

S-L

3.1 The A

S-L

X bundle

If you want to typeset (advanced) mathematics, you should use

-L

The

-L

X bundle is a collection of packages and classes for mathemat-

ical typesetting. We will mostly deal with the amsmath package which is a

part of the bundle.

-L

X is produced by The American Mathematical

Society and it is used extensively for mathematical typesetting. L

X itself

does provide some basic features and environments for mathematics, but they

are limited (or maybe it’s the other way around:

-L

X is unlimited!)

and in some cases inconsistent.

-L

X is a part of the required distribution and is provided with all

recent L

X distributions.

In this chapter, we assume amsmath is loaded

in the preamble; \usepackage{amsmath}.

3.2 Single Equations

A mathematical formula can be typeset in-line within a paragraph (text style),

or the paragraph can be broken and the formula typeset separately (display

style). Mathematical equations within a paragraph are entered between

and $:

If yours is missing it, go to CTAN://pkg/amslatex.

54 Typesetting Mathematical Formulae

Add $a$ squared and $b$ squared

to get $c$ squared. Or, using

a more mathematical approach:

$a^2 + b^2 = c^2$

Add

squared and

squared to get

squared. Or, using a more mathematical

approach: a

+ b

= c

\TeX{} is pronounced as

$\tau\epsilon\chi$\\[5pt]

100~m$^{3}$ of water\\[5pt]

This comes from my $\heartsuit$

X is pronounced as τχ

100 m

of water

This comes from my ♥

If you want your larger equations to be set apart from the rest of the

paragraph, it is preferable to display them rather than to break the para-

graph apart. To do this, you enclose them between

\begin{equation}

and

\end{equation}

You can then

\label

an equation number and refer to it

somewhere else in the text by using the

\eqref

command. If you want to

name the equation something speciﬁc, you \tag it instead.

Add $a$ squared and $b$ squared

to get $c$ squared. Or, using

a more mathematical approach

\begin{equation}

a^2 + b^2 = c^2

\end{equation}

Einstein says

\begin{equation}

E = mc^2 \label{clever}

\end{equation}

He didn’t say

\begin{equation}

1 + 1 = 3 \tag{dumb}

\end{equation}

This is a reference to

\eqref{clever}.

Add

squared and

squared to get

squared. Or, using a more mathematical

approach

+ b

= c

(3.1)

Einstein says

E = mc

(3.2)

He didn’t say

1 + 1 = 3 (dumb)

This is a reference to (3.2).

If you don’t want L

X to number the equations, use the starred version of

equation

using an asterisk,

equation*

, or even easier, enclose the equation

in \[ and \]:

This is an amsmath command. If you don’t have access to the package for some obscure

reason, you can use L

X’s own displaymath environment instead.

This is again from amsmath. Standard L

X’s has only the

equation

environment

without the star.

3.2 Single Equations 55

Add $a$ squared and $b$ squared

to get $c$ squared. Or, using

a more mathematical approach

\begin{equation*}

a^2 + b^2 = c^2

\end{equation*}

or you can type less for the

same effect:

\[ a^2 + b^2 = c^2 \]

Add

squared and

squared to get

squared. Or, using a more mathematical

approach

+ b

= c

or you can type less for the same eﬀect:

+ b

= c

While

is short and sweet, it does not allow switching between numbered

and not numbered style as easily as equation and equation*.

Note the diﬀerence in typesetting style between text style and display

style equations:

This is text style:

$\lim_{n \to \infty}

\sum_{k=1}^n \frac{1}{k^2}

= \frac{\pi^2}{6}$.

And this is display style:

\begin{equation}

\lim_{n \to \infty}

\sum_{k=1}^n \frac{1}{k^2}

= \frac{\pi^2}{6}

\end{equation}

This is text style: lim

n→∞

k=1

And this is display style:

lim

n→∞

k=1

(3.3)

In text style, enclose tall or deep math expressions or sub expressions

\smash

. This makes L

X ignore the height of these expressions. This

keeps the line spacing even.

A $d_{e_{e_p}}$ mathematical

expression followed by a

$h^{i^{g^h}}$ expression. As

opposed to a smashed

\smash{$d_{e_{e_p}}$} expression

followed by a

\smash{$h^{i^{g^h}}$} expression.

mathematical expression followed

by a

expression. As opposed to a

smashed

expression followed by a

expression.

3.2.1 Math Mode

There are also diﬀerences between math mode and text mode. For example,

in math mode:

Most spaces and line breaks do not have any signiﬁcance, as all spaces

are either derived logically from the mathematical expressions, or have

to be speciﬁed with special commands such as

\quad

\qquad

(we’ll get back to that later, see section 3.7).

56 Typesetting Mathematical Formulae

2. Empty lines are not allowed. Only one paragraph per formula.

Each letter is considered to be the name of a variable and will be

typeset as such. If you want to typeset normal text within a formula

(normal upright font and normal spacing) then you have to enter the

text using the

\text{...}

command (see also section 3.8 on page 70).

$\forall x \in \mathbf{R}:

\qquad x^{2} \geq 0$

∀x ∈ R : x

≥ 0

$x^{2} \geq 0\qquad

\text{for all }x\in\mathbf{R}$

≥ 0 for all x ∈ R

Mathematicians can be very fussy about which symbols are used: it would

be conventional here to use the ‘blackboard bold’ font, which is obtained

using \mathbb from the package amssymb.

The last example becomes

$x^{2} \geq 0\qquad

\text{for all } x

\in \mathbb{R}$

≥ 0 for all x ∈ R

See Table 3.14 on page 79 and Table 6.4 on page 125 for more math fonts.

3.3 Building Blocks of a Mathematical Formula

In this section, we describe the most important commands used in mathe-

matical typesetting. Most of the commands in this section will not require

amsmath (if they do, it will be stated clearly), but load it anyway.

Lowercase Greek letters

are entered as

\alpha

\beta

\gamma

, . . . ,

uppercase letters are entered as \Gamma, \Delta, . . .

Take a look at Table 3.2 on page 75 for a list of Greek letters.

$\lambda,\xi,\pi,\theta,

\mu,\Phi,\Omega,\Delta$

λ, ξ, π, θ, µ, Φ, Ω, ∆

Exponents, Superscripts and Subscripts

can be speciﬁed using the

and the

characters. Most math mode commands act only on the next

character, so if you want a command to aﬀect several characters, you have

to group them together using curly braces: {...}.

amssymb is not a part of the

-L

X bundle, but it is perhaps still a part of your

X distribution. Check your distribution or go to

CTAN:/fonts/amsfonts/latex/

obtain it.

There is no uppercase Alpha, Beta etc. deﬁned in

X 2

because it looks the same

as a normal roman A, B. . .

3.3 Building Blocks of a Mathematical Formula 57

Table 3.3 on page 76 lists a lot of binary relations like ⊆ and ⊥.

$p^3_{ij} \qquad

m_\text{Knuth}\qquad

\sum_{k=1}^3 k \\[5pt]

a^x+y \neq a^{x+y}\qquad

e^{x^2} \neq {e^x}^2$

Knuth

k=1

+ y 6= a

x+y

6= e

The

square root

is entered as

\sqrt

; the

root is generated with

\sqrt[n]

. The size of the root sign is determined automatically by L

If just the sign is needed, use \surd.

See various kinds of arrows like → and  on Table 3.6 on page 77.

$\sqrt{x} \Leftrightarrow x^{1/2}

\quad \sqrt[3]{2}

\quad \sqrt{x^{2} + \sqrt{y}}

\quad \surd[x^2 + y^2]$

√

x ⇔ x

1/2

√

[

]

While the

dot

sign to indicate the multiplication operation is normally

left out, it is sometimes written to help the eye in grouping a formula. Use

\cdot

to typeset a single centered dot.

\cdots

is three centered

dots

while

\ldots

sets the dots low (on the baseline). Besides that, there are

\vdots

for vertical and

\ddots

for diagonal dots. There are more examples in

section 3.6.

$\Psi = v_1 \cdot v_2

\cdot \ldots \qquad

n! = 1 \cdot 2

\cdots (n-1) \cdot n$

Ψ =

·v

·. . . n

! = 1

···

(

n −

·n

The commands

\overline

and

\underline

create

horizontal lines

directly over or under an expression:

$0.\overline{3} =

\underline{\underline{1/3}}$

0.3 = 1/3

The commands

\overbrace

and

\underbrace

create long

horizontal

braces over or under an expression:

$\underbrace{\overbrace{a+b+c}^6

\cdot \overbrace{d+e+f}^7}

_\text{meaning of life} = 42$

z }| {

a + b + c ·

z }| {

d + e + f

| {z }

meaning of life

= 42

To add mathematical accents such as

small arrows

tilde

signs to

variables, the commands given in Table 3.1 on page 75 might be useful. Wide

58 Typesetting Mathematical Formulae

hats and tildes covering several characters are generated with

\widetilde

and

\widehat

. Notice the diﬀerence between

\hat

and

\widehat

and the

placement of

\bar

for a variable with subscript. The apostrophe mark

’

gives a prime:

$f(x) = x^2 \qquad f’(x)

= 2x \qquad f’’(x) = 2\\[5pt]

\hat{XY} \quad \widehat{XY}

\quad \bar{x_0} \quad \bar{x}_0$

f(x) = x

(x) = 2x f

(x) = 2

XY ¯x

¯x

Vectors

are often speciﬁed by adding small arrow symbols on the tops

of variables. This is done with the

\vec

command. The two commands

\overrightarrow

and

\overleftarrow

are useful to denote the vector from

A to B:

$\vec{a} \qquad

\vec{AB} \qquad

\overrightarrow{AB}$

a



−−→

Names of functions are often typeset in an upright font, and not in italics

as variables are, so L

X supplies the following commands to typeset the

most common function names:

\arccos \cos \csc \exp \ker \limsup

\arcsin \cosh \deg \gcd \lg \ln

\arctan \cot \det \hom \lim \log

\arg \coth \dim \inf \liminf \max

\sinh \sup \tan \tanh \min \Pr

\sec \sin

\begin{equation*}

\lim_{x \rightarrow 0}

\frac{\sin x}{x}=1

\end{equation*}

lim

x→0

sin x

= 1

For functions missing from the list, use the

\DeclareMathOperator

command. There is even a starred version for functions with limits. This

command works only in the preamble so the commented lines in the example

below must be put into the preamble.

%\DeclareMathOperator{\argh}{argh}

%\DeclareMathOperator*{\nut}{Nut}

\begin{equation*}

3\argh = 2\nut_{x=1}

\end{equation*}

3 argh = 2 Nut

x=1

For the modulo function, there are two commands:

\bmod

for the binary

operator “a mod b” and \pmod for expressions such as “x ≡ a (mod b):”

3.3 Building Blocks of a Mathematical Formula 59

$a\bmod b \\

x\equiv a \pmod{b}$

a mod b

x ≡ a (mod b)

A built-up

fraction

is typeset with the

\frac{...}{...}

command. In

in-line equations, the fraction is shrunk to ﬁt the line. This style is obtainable

in display style with

\tfrac

. The reverse, i.e. display style fraction in text,

is made with

\dfrac

. Often the slashed form 1

2 is preferable, because it

looks better for small amounts of ‘fraction material:’

In display style:

\begin{equation*}

3/8 \qquad \frac{3}{8}

\qquad \tfrac{3}{8}

\end{equation*}

In display style:

3/8

In text style:

$1\frac{1}{2}$~hours \qquad

$1\dfrac{1}{2}$~hours

In text style: 1

hours 1

hours

Here the \partial command for partial derivatives is used:

\begin{equation*}

\sqrt{\frac{x^2}{k+1}}\qquad

x^\frac{2}{k+1}\qquad

\frac{\partial^2f}

{\partial x^2}

\end{equation*}

k + 1

k+1

∂

∂x

To typeset binomial coeﬃcients or similar structures, use the command

\binom from amsmath:

Pascal’s rule is

\begin{equation*}

\binom{n}{k} =\binom{n-1}{k}

+ \binom{n-1}{k-1}

\end{equation*}

Pascal’s rule is







n − 1





n − 1

k − 1



For binary relations it may be useful to stack symbols over each other.

\stackrel{#1}{#2}

puts the symbol given in

in superscript-like size over

#2 which is set in its usual position.

\begin{equation*}

f_n(x) \stackrel{*}{\approx} 1

\end{equation*}

(x)

∗

≈ 1

60 Typesetting Mathematical Formulae

The

integral operator

is generated with

\int

, the

sum op erator

with

\sum

, and the

product operator

with

\prod

. The upper and lower limits

are speciﬁed with ^ and _ like subscripts and superscripts:

\begin{equation*}

\sum_{i=1}^n \qquad

\int_0^{\frac{\pi}{2}} \qquad

\prod_\epsilon

\end{equation*}

i=1



To get more control over the placement of indices in complex expressions,

amsmath provides the \substack command:

\begin{equation*}

\sum^n_{\substack{0<i<n \\

j\subseteq i}}

P(i,j) = Q(i,j)

\end{equation*}

0<i<n

j⊆i

P (i, j) = Q(i, j)

X provides all sorts of symbols for

bracketing

and other

delim-

iters

(e.g. [

h k l

). Round and square brackets can be entered with the

corresponding keys and curly braces with

, but all other delimiters are

generated with special commands (e.g. \updownarrow).

\begin{equation*}

{a,b,c} \neq \{a,b,c\}

\end{equation*}

a, b, c 6= {a, b, c}

If you put

\left

in front of an opening delimiter and

\right

in front of

a closing delimiter, L

X will automatically determine the correct size of

the delimiter. Note that you must close every

\left

with a corresponding

\right

. If you don’t want anything on the right, use the invisible “

\right.

”:

\begin{equation*}

1 + \left(\frac{1}{1-x^{2}}

\right)^3 \qquad

\left. \ddagger \frac{~}{~}\right)

\end{equation*}

1 +



1 − x



‡



In some cases it is necessary to specify the correct size of a mathematical

delimiter by hand, which can be done using the commands

\big

\Big

\bigg

and \Bigg as preﬁxes to most delimiter commands:

$\Big((x+1)(x-1)\Big)^{2}$\\

$\big( \Big( \bigg( \Bigg( \quad

\big\} \Big\} \bigg\} \Bigg\} \quad

\big\| \Big\| \bigg\| \Bigg\| \quad

\big\Downarrow \Big\Downarrow

\bigg\Downarrow \Bigg\Downarrow$



(x + 1)(x − 1)













)





3.4 Single Equations that are Too Long: multline 61

For a list of all delimiters available, see Table 3.8 on page 78.

3.4 Single Equations that are Too Long: multline

If an equation is too long, we have to wrap it somehow. Unfortunately,

wrapped equations are usually less easy to read than not wrapped ones. To

improve the readability, there are certain rules on how to do the wrapping:

In general one should always wrap an equation

before

an equality sign

or an operator.

A wrap before an equality sign is preferable to a wrap before any

operator.

A wrap before a plus- or minus-operator is preferable to a wrap before

a multiplication-operator.

4. Any other type of wrap should be avoided if at all possible.

The easiest way to achieve such a wrapping is the use of the

multline

environment:

\begin{multline}

a + b + c + d + e + f

+ g + h + i

= j + k + l + m + n

\end{multline}

a + b + c + d + e + f + g + h + i

= j + k + l + m + n (3.4)

The diﬀerence from the

equation

environment is that an arbitrary line-break

(or also multiple line-breaks) can be introduced. This is done by putting a

on those places where the equation needs to be wrapped. Similarly to

equation*

there also exists a

multline*

version for preventing an equation

number.

Often the

IEEEeqnarray

environment (see section 3.5) will yield better

results. Consider the following situation:

\begin{equation}

a = b + c + d + e + f

+ g + h + i + j

+ k + l + m + n + o + p

\label{eq:equation_too_long}

\end{equation}

a = b+c+d+e+f+g+h+i+j+k+l+m+n+o+p

(3.5)

Here it is actually the RHS that is too long to ﬁt on one line. The

multline

environment creates the following output:

The multline-environment is from amsmath.

62 Typesetting Mathematical Formulae

\begin{multline}

a = b + c + d + e + f

+ g + h + i + j \\

+ k + l + m + n + o + p

\end{multline}

a = b + c + d + e + f + g + h + i + j

+ k + l + m + n + o + p (3.6)

This is better than

(3.5)

, but it has the disadvantage that the equality

sign loses its natural greater importance with respect to the plus operator in

front of

. The better solution is provided by the

IEEEeqnarray

environment

that will be discussed in detail in Section 3.5.

3.5 Multiple Equations

In the most general situation we have a sequence of several equalities that

do not ﬁt onto one line. Here we need to work with vertical alignment in

order to keep the array of equations in a nice and readable structure.

Before we oﬀer our suggestions on how to do this, we start with a few

bad examples that show the biggest drawbacks of some common solutions.

3.5.1 Problems with Traditional Commands

To group multiple equations the align environment

could be used:

\begin{align}

a & = b + c \\

& = d + e

\end{align}

a = b + c (3.7)

= d + e (3.8)

this approach fails once a single line is too long:

\begin{align}

a & = b + c \\

& = d + e + f + g + h + i

+ j + k + l \nonumber \\

& + m + n + o \\

& = p + q + r + s

\end{align}

a = b + c (3.9)

= d + e + f + g + h + i + j + k + l

+ m + n + o (3.10)

= p + q + r + s (3.11)

Here +

should be below

and not below the equality sign. Of course, one

could add some space (

\hspace{...}

), but this will never yield a precise

arrangement (and is bad style. . . ).

A better solution is oﬀered by the eqnarray environment:

The

align

-environment can also be used to group several blocks of equations beside

each other. Another excellent use case for the

IEEEeqnarray

environment. Try an argument

like {rCl+rCl}.

3.5 Multiple Equations 63

\begin{eqnarray}

a & = & b + c \\

& = & d + e + f + g + h + i

+ j + k + l \nonumber \\

&& +\: m + n + o \\

& = & p + q + r + s

\end{eqnarray}

a = b + c (3.12)

= d + e + f + g + h + i + j + k + l

+ m + n + o (3.13)

= p + q + r + s (3.14)

This is still not optimal. The spaces around the equality signs are too

big. Particularly, they are

not

the same as in the

multline

and

equation

environments:

\begin{eqnarray}

a & = & a = a

\end{eqnarray}

a = a = a (3.15)

. . . and the expression sometimes overlaps with the equation number even

though there would be enough room on the left:

\begin{eqnarray}

a & = & b + c

& = & d + e + f + g + h^2

+ i^2 + j

\label{eq:faultyeqnarray}

\end{eqnarray}

a = b + c (3.16)

= d + e + f + g + h

+ i

+ j(3.17)

While the environment oﬀers a command

\lefteqn

that can be used when

the LHS is too long:

\begin{eqnarray}

\lefteqn{a + b + c + d

+ e + f + g + h}\nonumber\\

& = & i + j + k + l + m

& = & n + o + p + q + r + s

\end{eqnarray}

a + b + c + d + e + f + g + h

= i + j + k + l + m (3.18)

= n + o + p + q + r + s (3.19)

This is not optimal either as the RHS is too short and the array is not

properly centered:

\begin{eqnarray}

\lefteqn{a + b + c + d

+ e + f + g + h}

\nonumber \\

& = & i + j

\end{eqnarray}

a + b + c + d + e + f + g + h

= i + j (3.20)

64 Typesetting Mathematical Formulae

Having badmouthed the competition suﬃciently, I can now steer you gently

towards the glorious . . .

3.5.2 IEEEeqnarray Environment

The

IEEEeqnarray

environment is a very powerful command with many

options. Here, we will only introduce its basic functionalities. For more

information please refer to the manual.

First of all, in order to be able to use the

IEEEeqnarray

environment

one needs to load the package

IEEEtrantools. Include the following line in

the header of your document:

\usepackage[retainorgcmds]{IEEEtrantools}

The strength of

IEEEeqnarray

is the ability to specify the number of

columns in the equation array. Usually, this speciﬁcation will be

{rCl}

i.e., three columns, the ﬁrst column right-justiﬁed, the middle one centered

with a little more space around it (therefore we specify capital

instead of

lower-case c) and the third column left-justiﬁed:

\begin{IEEEeqnarray}{rCl}

a & = & b + c

& = & d + e + f + g + h

+ i + j + k \nonumber\\

&& \negmedspace {} + l + m + n + o

& = & p + q + r + s

\end{IEEEeqnarray}

a = b + c (3.21)

= d + e + f + g + h + i + j + k

+ l + m + n + o (3.22)

= p + q + r + s (3.23)

Any number of columns can be speciﬁed:

{c}

will give only one column

with all entries centered, or

{rCll}

would add a fourth, left-justiﬁed column

to use for comments. Moreover, beside

for math mode

entries there are also

for left, centered, and right text mode entries.

Additional space can be added with

and

in increasing order.

Note the spaces around the equality signs in contrast to the space produced

by the eqnarray environment.

3.5.3 Common Usage

In the following we will describe how we use

IEEEeqnarray

to solve the most

common problems.

If a line overlaps with the equation number as in (3.17), the command

The oﬃcial manual is called

CTAN://macros/latex/contrib/IEEEtran/IEEEtran_

HOWTO.pdf. The part about IEEEeqnarray can be found in Appendix F.

The IEEEtrantools package may not be included in your setup, it can be found on

CTAN.

For more spacing types refer to Section 3.9.1.

3.5 Multiple Equations 65

\IEEEeqnarraynumspace

can be used: it has to be added in the corresponding line and makes sure

that the whole equation array is shifted by the size of the equation numbers

(the shift depends on the size of the number!): instead of

\begin{IEEEeqnarray}{rCl}

a & = & b + c

& = & d + e + f + g + h

+ i + j + k

& = & l + m + n

\end{IEEEeqnarray}

a = b + c (3.24)

= d + e + f + g + h + i + j + k(3.25)

= l + m + n (3.26)

we get

\begin{IEEEeqnarray}{rCl}

a & = & b + c

& = & d + e + f + g + h

+ i + j + k

\IEEEeqnarraynumspace\\

& = & l + m + n.

\end{IEEEeqnarray}

a = b + c (3.27)

= d + e + f + g + h + i + j + k (3.28)

= l + m + n. (3.29)

If the LHS is too long, as a replacement for the faulty

\lefteqn

command,

IEEEeqnarray

oﬀers the

\IEEEeqnarraymulticol

command which works

in all situations:

\begin{IEEEeqnarray}{rCl}

\IEEEeqnarraymulticol{3}{l}{

a + b + c + d + e + f

+ g + h

}\nonumber\\ \quad

& = & i + j

& = & k + l + m

\end{IEEEeqnarray}

a + b + c + d + e + f + g + h

= i + j (3.30)

= k + l + m (3.31)

The usage is identical to the

\multicolumns

command in the

tabular

environment. The ﬁrst argument

{3}

speciﬁes that three columns shall be

combined into one which will be left-justiﬁed {l}.

Note that by inserting

\quad

commands one can easily adapt the depth

of the equation signs,

e.g.,

I think that one quad is the distance that looks good for most cases.

66 Typesetting Mathematical Formulae

\begin{IEEEeqnarray}{rCl}

\IEEEeqnarraymulticol{3}{l}{

a + b + c + d + e + f

+ g + h

}\nonumber\\ \qquad\qquad

& = & i + j

& = & k + l + m

\end{IEEEeqnarray}

a + b + c + d + e + f + g + h

= i + j (3.32)

= k + l + m (3.33)

If an equation is split into two or more lines, L

X interprets the ﬁrst

+ or

−

as a sign instead of operator. Therefore, it is necessary to add an

additional space \: between the operator and the term: instead of

\begin{IEEEeqnarray}{rCl}

a & = & b + c

& = & d + e + f + g + h

+ i + j + k \nonumber\\

&& + l + m + n + o

& = & p + q + r + s

\end{IEEEeqnarray}

a = b + c (3.34)

= d + e + f + g + h + i + j + k

+l + m + n + o (3.35)

= p + q + r + s (3.36)

we should write

\begin{IEEEeqnarray}{rCl}

a & = & b + c

& = & d + e + f + g + h

+ i + j + k \nonumber\\

&& \negmedspace {} + l + m + n + o

& = & p + q + r + s

\end{IEEEeqnarray}

a = b + c (3.37)

= d + e + f + g + h + i + j + k

+ l + m + n + o (3.38)

= p + q + r + s (3.39)

Note the space diﬀerence between + and

! The construction

{} + l

forces

the

-sign to be a binary operator rather than just a sign, and the unwanted

ensuing space between

{}

and

is compensated by a negative medium space

\negmedspace.

If a particular line should not have an equation number, the number

can be suppressed using

\nonumber

(or

\IEEEnonumber

). If on such a line

a label

\label{eq:...}

is deﬁned, then this label is passed on to the next

equation number that is not suppressed. Place the labels right before the

line-break

or the next to the equation it belongs to. Apart from improving

the readability of the source code this prevents a compilation error when a

\IEEEmulticol command follows the label-deﬁnition.

3.6 Arrays and Matrices 67

There also exists a *-version where all equation numbers are suppressed.

In this case an equation number can be made to appear using the command

\IEEEyesnumber:

\begin{IEEEeqnarray*}{rCl}

a & = & b + c \\

& = & d + e \IEEEyesnumber\\

& = & f + g

\end{IEEEeqnarray*}

a = b + c

= d + e (3.40)

= f + g

Sub-numbers are also easily possible using \IEEEyessubnumber:

\begin{IEEEeqnarray}{rCl}

a & = & b + c

\IEEEyessubnumber\\

& = & d + e

\nonumber\\

& = & f + g

\IEEEyessubnumber

\end{IEEEeqnarray}

a = b + c (3.40a)

= d + e

= f + g (3.40b)

3.6 Arrays and Matrices

To typeset

arrays

, use the

array

environment. It works in a similar way to

the tabular environment. The \\ command is used to break the lines:

\begin{equation*}

\mathbf{X} = \left(

\begin{array}{ccc}

x_1 & x_2 & \ldots \\

x_3 & x_4 & \ldots \\

\vdots & \vdots & \ddots

\end{array} \right)

\end{equation*}

X =







. . .







The

array

environment can also be used to typeset piecewise functions

by using a “.” as an invisible \right delimiter:

\begin{equation*}

|x| = \left\{

\begin{array}{rl}

-x & \text{if } x < 0,\\

0 & \text{if } x = 0,\\

x & \text{if } x > 0.

\end{array} \right.

\end{equation*}

|x| =







−x if x < 0,

0 if x = 0,

x if x > 0.

68 Typesetting Mathematical Formulae

The

cases

environment from amsmath simpliﬁes the syntax, so it is worth

a look:

\begin{equation*}

|x| =

\begin{cases}

-x & \text{if } x < 0,\\

0 & \text{if } x = 0,\\

x & \text{if } x > 0.

\end{cases}

\end{equation*}

|x| =











−x if x < 0,

0 if x = 0,

x if x > 0.

Matrices can be typeset by

array

, but amsmath provides a better solution

using the diﬀerent

matrix

environments. There are six versions with diﬀerent

delimiters:

matrix

(none),

pmatrix

bmatrix

Bmatrix {

vmatrix |

and

Vmatrix k

. You don’t have to specify the number of columns as with

array

The maximum number is 10, but it is customisable (though it is not very

often you need 10 columns!):

\begin{equation*}

\begin{matrix}

1 & 2 \\

3 & 4

\end{matrix} \qquad

\begin{bmatrix}

p_{11} & p_{12} & \ldots

& p_{1n} \\

p_{21} & p_{22} & \ldots

& p_{2n} \\

\vdots & \vdots & \ddots

& \vdots \\

p_{m1} & p_{m2} & \ldots

& p_{mn}

\end{bmatrix}

\end{equation*}

1 2

3 4







. . . p







3.7 Spacing in Math Mode

If the spacing within formulae chosen by L

X is not satisfactory, it can be

adjusted by inserting special spacing commands:

for

quad

( ),

for

quad

( ) and

for

quad

( ). The escaped space character

\␣

generates

a medium sized space comparable to the interword spacing and

\quad

( )

and

\qquad

( ) produce large spaces. The size of a

\quad

corresponds

to the width of the character ‘M’ of the current font.

produces a negative

space of −

quad (− ).

3.7 Spacing in Math Mode 69

\begin{equation*}

\int_1^2 \ln x \mathrm{d}x

\qquad

\int_1^2 \ln x \,\mathrm{d}x

\end{equation*}

ln xdx

ln x dx

Note that ‘d’ in the diﬀerential is conventionally set in roman. In the

next example, we deﬁne a new command

\ud

(upright d) which produces “ d”

(notice the spacing before the

), so we don’t have to write it every time.

The \newcommand is placed in the preamble.

\newcommand{\ud}{\,\mathrm{d}}

\begin{equation*}

\int_a^b f(x)\ud x

\end{equation*}

f(x) dx

If you want to typeset multiple integrals, you’ll discover that the spacing

between the integrals is too wide. You can correct it using

, but amsmath

provides an easier way for ﬁne-tuning the spacing, namely the

\iint

\iiint

\iiiint, and \idotsint commands.

\newcommand{\ud}{\,\mathrm{d}}

\begin{IEEEeqnarray*}{c}

\int\int f(x)g(y)

\ud x \ud y \\

\int\!\!\!\int

f(x)g(y) \ud x \ud y \\

\iint f(x)g(y) \ud x \ud y

\end{IEEEeqnarray*}

Z Z

f(x)g(y) dx dy

See the electronic document

testmath.tex

(distributed with

-L

or Chapter 8 of The L

X Companion [3] for further details.

3.7.1 Phantoms

When vertically aligning text using

and

X is sometimes just a little too

helpful. Using the

\phantom

command you can reserve space for characters

that do not show up in the ﬁnal output. The easiest way to understand this

is to look at an example:

\begin{equation*}

{}^{14}_{6}\text{C}

\qquad \text{versus} \qquad

{}^{14}_{\phantom{1}6}\text{C}

\end{equation*}

C versus

70 Typesetting Mathematical Formulae

If you want to typeset a lot of isotopes as in the example, the mhchem

package is very useful for typesetting isotopes and chemical formulae too.

3.8 Fiddling with the Math Fonts

Diﬀerent math fonts are listed on Table 3.14 on page 79.

$\Re \qquad

\mathcal{R} \qquad

\mathfrak{R} \qquad

\mathbb{R} \qquad $

< R R R

The last two require amssymb or amsfonts.

Sometimes you need to tell L

X the correct font size. In math mode,

this is set with the following four commands:

\displaystyle (123), \textstyle (123), \scriptstyle (123) and

\scriptscriptstyle (123).

is placed in a fraction, it’ll be typeset in text style unless you tell

X otherwise:

\begin{equation*}

P = \frac{\displaystyle{

\sum_{i=1}^n (x_i- x)

(y_i- y)}}

{\displaystyle{\left[

\sum_{i=1}^n(x_i-x)^2

\sum_{i=1}^n(y_i- y)^2

\right]^{1/2}}}

\end{equation*}

P =

i=1

− x)(y

− y)

i=1

− x)

i=1

− y)

1/2

Changing styles generally aﬀects the way big operators and limits are

displayed.

3.8.1 Bold Symbols

It is quite diﬃcult to get bold symbols in L

X; this is probably intentional

as amateur typesetters tend to overuse them. The font change command

\mathbf

gives bold letters, but these are roman (upright) whereas math-

ematical symbols are normally italic, and furthermore it doesn’t work on

lower case Greek letters. There is a

\boldmath

command, but this can only

be used outside math mode. It works for symbols too, though:

$\mu, M \qquad

\mathbf{\mu}, \mathbf{M}$

\qquad \boldmath{$\mu, M$}

µ, M µ, M µ, M

3.9 Theorems, Lemmas, . . . 71

The package amsbsy (included by amsmath) as well as the package bm

from the

tools

bundle make this much easier as they include a

\boldsymbol

command:

$\mu, M \qquad

\boldsymbol{\mu}, \boldsymbol{M}$

µ, M µ, M

3.9 Theorems, Lemmas, . . .

When writing mathematical documents, you probably need a way to typeset

“Lemmas”, “Deﬁnitions”, “Axioms” and similar structures.

\newtheorem{name}[counter]{text}[section]

The name argument is a short keyword used to identify the “theorem”.

With the text argument you deﬁne the actual name of the “theorem”, which

will be printed in the ﬁnal document.

The arguments in square brackets are optional. They are both used to

specify the numbering used on the “theorem”. Use the counter argument to

specify the name of a previously declared “theorem”. The new “theorem”

will then be numbered in the same sequence. The section argument allows

you to specify the sectional unit within which the “theorem” should get its

numbers.

After executing the

\newtheorem

command in the preamble of your

document, you can use the following command within the document.

\begin{name}[text]

This is my interesting theorem

\end{name}

The amsthm package (part of

-L

X) provides the

\theoremstyle{

style

}

command which lets you deﬁne what the theorem is all about by picking

from three predeﬁned styles:

definition

(fat title, roman body),

plain

(fat

title, italic body) or remark (italic title, roman body).

This should be enough theory. The following examples should remove

any remaining doubt, and make it clear that the

\newtheorem

environment

is way too complex to understand.

First deﬁne the theorems:

\theoremstyle{definition} \newtheorem{law}{Law}

\theoremstyle{plain} \newtheorem{jury}[law]{Jury}

\theoremstyle{remark} \newtheorem*{marg}{Margaret}

72 Typesetting Mathematical Formulae

\begin{law} \label{law:box}

Don’t hide in the witness box

\end{law}

\begin{jury}[The Twelve]

It could be you! So beware and

see law~\ref{law:box}.\end{jury}

\begin{jury}

You will disregard the last

statement.\end{jury}

\begin{marg}No, No, No\end{marg}

\begin{marg}Denis!\end{marg}

Law 1. Don’t hide in the witness box

Jury 2

(The Twelve)

It could be you! So

beware and see law 1.

Jury 3.

You will disregard the last state-

ment.

Margaret. No, No, No

Margaret. Denis!

The “Jury” theorem uses the same counter as the “Law” theorem, so it

gets a number that is in sequence with the other “Laws”. The argument in

square brackets is used to specify a title or something similar for the theorem.

\newtheorem{mur}{Murphy}[section]

\begin{mur} If there are two or

more ways to do something, and

one of those ways can result in

a catastrophe, then someone

will do it.\end{mur}

Murphy 3.9.1. If there are two or more

ways to do something, and one of those

ways can result in a catastrophe, then

someone will do it.

The “Murphy” theorem gets a number that is linked to the number of

the current section. You could also use another unit, for example chapter or

subsection.

If you want to customize your theorems down to the last dot, the ntheorem

package oﬀers a plethora of options.

3.9.1 Proofs and End-of-Proof Symbol

The amsthm package also provides the proof environment.

\begin{proof}

Trivial, use

\begin{equation*}

E=mc^2.

\end{equation*}

\end{proof}

Proof. Trivial, use

E = mc

With the command

\qedhere

you can move the ‘end of proof’ symbol

around for situations where it would end up alone on a line.

3.9 Theorems, Lemmas, . . . 73

\begin{proof}

Trivial, use

\begin{equation*}

E=mc^2. \qedhere

\end{equation*}

\end{proof}

Proof. Trivial, use

E = mc

Unfortunately, this correction does not work for IEEEeqnarray:

\begin{proof}

This is a proof that ends

with an equation array:

\begin{IEEEeqnarray*}{rCl}

a & = & b + c \\

& = & d + e. \qedhere

\end{IEEEeqnarray*}

\end{proof}

Proof.

This is a proof that ends with an

equation array:

a = b + c

= d + e.

The reason for this is the internal structure of

IEEEeqnarray

: it always

puts two invisible columns at both sides of the array that only contain a

stretchable space. By this

IEEEeqnarray

ensures that the equation array

is horizontally centered. The

\qedhere

command should actually be put

outside this stretchable space, but this does not happen as these columns

are invisible to the user.

There is a very simple remedy. Deﬁne the stretching explicitly!

\begin{proof}

This is a proof that ends

with an equation array:

\begin{IEEEeqnarray*}{+rCl+x*}

a & = & b + c \\

& = & d + e. & \qedhere

\end{IEEEeqnarray*}

\end{proof}

Proof.

This is a proof that ends with an

equation array:

a = b + c

= d + e.

Note that the

{+rCl+x*}

denotes stretchable spaces, one on the left

of the equations (which, if not speciﬁed, will be done automatically by

IEEEeqnarray

!) and one on the right of the equations. But now on the

right, after the stretching column, we add an empty column

. This column

will only be needed on the last line if the

\qedhere

command is put there.

Finally, we specify a

. This is a null-space that prevents

IEEEeqnarray

from adding another unwanted +-space!

In the case of equation numbering, there is a similar problem. Comparing

74 Typesetting Mathematical Formulae

\begin{proof}

This is a proof that ends

with a numbered equation:

\begin{equation}

a = b + c.

\end{equation}

\end{proof}

Proof.

This is a proof that ends with a

numbered equation:

a = b + c. (3.41)

with

\begin{proof}

This is a proof that ends

with a numbered equation:

\begin{equation}

a = b + c. \qedhere

\end{equation}

\end{proof}

Proof.

This is a proof that ends with a

numbered equation:

a = b + c. (3.42)

you notice that in the (correct) second version the

is much closer to the

equation than in the ﬁrst version.

Similarly, the correct way of putting the QED-symbol at the end of an

equation array is as follows:

\begin{proof}

This is a proof that ends

with an equation array:

\begin{IEEEeqnarray}{+rCl+x*}

a & = & b + c \\

& = & d + e. \\

&&& \qedhere\nonumber

\end{IEEEeqnarray}

\end{proof}

Proof.

This is a proof that ends with an

equation array:

a = b + c (3.43)

= d + e. (3.44)

which contrasts with

\begin{proof}

This is a proof that ends

with an equation array:

\begin{IEEEeqnarray}{rCl}

a & = & b + c \\

& = & d + e.

\end{IEEEeqnarray}

\end{proof}

Proof.

This is a proof that ends with an

equation array:

a = b + c (3.45)

= d + e. (3.46)

3.10 List of Mathematical Symbols 75

3.10 List of Mathematical Symbols

The following tables demonstrate all the symbols normally accessible from

math mode.

Note that some tables show symbols only accessible after loading the

amssymb package in the preamble of your document

. If the

pack-

age and fonts are not installed on your system, have a look at

CTAN:

pkg/amslatex

. An even more comprehensive list of symbols can be found

at CTAN:info/symbols/comprehensive.

Table 3.1: Math Mode Accents.

ˆa \hat{a} ˇa \check{a} ˜a \tilde{a}

`a \grave{a} ˙a \dot{a} ¨a \ddot{a}

¯a \bar{a} a \vec{a}



AAA \widehat{AAA}

´a \acute{a} ˘a \breve{a}



AAA \widetilde{AAA}

˚a \mathring{a}

Table 3.2: Greek Letters.

There is no uppercase of some of the letters like

\Alpha

\Beta

and so on,

because they look the same as normal roman letters: A, B. . .

α \alpha θ \theta o o υ \upsilon

β \beta ϑ \vartheta π \pi φ \phi

γ \gamma ι \iota  \varpi ϕ \varphi

δ \delta κ \kappa ρ \rho χ \chi

 \epsilon λ \lambda  \varrho ψ \psi

ε \varepsilon µ \mu σ \sigma ω \omega

ζ \zeta ν \nu ς \varsigma

η \eta ξ \xi τ \tau

Γ \Gamma Λ \Lambda Σ \Sigma Ψ \Psi

∆ \Delta Ξ \Xi Υ \Upsilon Ω \Omega

Θ \Theta Π \Pi Φ \Phi

The tables were derived from

symbols.tex

by David Carlisle and subsequently changed

extensively as suggested by Josef Tkadlec.

76 Typesetting Mathematical Formulae

Table 3.3: Binary Relations.

You can negate the following symbols by preﬁxing them with a

\not

com-

mand.

< < > > = =

≤ \leq or \le ≥ \geq or \ge ≡ \equiv

 \ll  \gg

= \doteq

≺ \prec  \succ ∼ \sim

 \preceq  \succeq ' \simeq

⊂ \subset ⊃ \supset ≈ \approx

⊆ \subseteq ⊇ \supseteq

∼

\cong

@ \sqsubset

A \sqsupset

1 \Join

v \sqsubseteq w \sqsupseteq  \bowtie

∈ \in 3 \ni , \owns ∝ \propto

` \vdash a \dashv |= \models

| \mid k \parallel ⊥ \perp

 \smile  \frown  \asymp

: : /∈ \notin 6= \neq or \ne

Use the latexsym package to access this symbol

Table 3.4: Binary Operators.

+ + − -

± \pm ∓ \mp  \triangleleft

· \cdot ÷ \div  \triangleright

× \times \ \setminus  \star

∪ \cup ∩ \cap ∗ \ast

t \sqcup u \sqcap ◦ \circ

∨ \vee , \lor ∧ \wedge , \land • \bullet

⊕ \oplus  \ominus  \diamond

 \odot  \oslash ] \uplus

⊗ \otimes  \bigcirc q \amalg

4 \bigtriangleup 5 \bigtriangledown † \dagger

 \lhd

 \rhd

‡ \ddagger

 \unlhd

 \unrhd

o \wr

3.10 List of Mathematical Symbols 77

Table 3.5: BIG Operators.

\sum

\bigcup

\bigvee

\prod

\bigcap

\bigwedge

\coprod

\bigsqcup

\biguplus

\int

\oint

\bigodot

\bigoplus

\bigotimes

Table 3.6: Arrows.

← \leftarrow or \gets ←− \longleftarrow

→ \rightarrow or \to −→ \longrightarrow

↔ \leftrightarrow ←→ \longleftrightarrow

⇐ \Leftarrow ⇐= \Longleftarrow

⇒ \Rightarrow =⇒ \Longrightarrow

⇔ \Leftrightarrow ⇐⇒ \Longleftrightarrow

7→ \mapsto 7−→ \longmapsto

← \hookleftarrow → \hookrightarrow

 \leftharpoonup  \rightharpoonup

 \leftharpoondown  \rightharpoondown

 \rightleftharpoons ⇐⇒ \iff (bigger spaces)

↑ \uparrow ↓ \downarrow

l \updownarrow ⇑ \Uparrow

⇓ \Downarrow m \Updownarrow

% \nearrow & \searrow

. \swarrow - \nwarrow

; \leadsto

Use the latexsym package to access this symbol

Table 3.7: Arrows as Accents.

−−→

AB \overrightarrow{AB} AB

−−→

\underrightarrow{AB}

←−−

AB \overleftarrow{AB} AB

←−−

\underleftarrow{AB}

←→

AB \overleftrightarrow{AB} AB

←→

\underleftrightarrow{AB}

78 Typesetting Mathematical Formulae

Table 3.8: Delimiters.

( ( ) ) ↑ \uparrow

[ [ or \lbrack ] ] or \rbrack ↓ \downarrow

{ \{ or \lbrace } \} or \rbrace l \updownarrow

h \langle i \rangle ⇑ \Uparrow

| | or \vert k \| or \Vert ⇓ \Downarrow

/ / \ \backslash m \Updownarrow

b \lfloor c \rfloor

e \rceil d \lceil

Table 3.9: Large Delimiters.





\lgroup





\rgroup





\lmoustache



\arrowvert

\Arrowvert



\bracevert





\rmoustache

Table 3.10: Miscellaneous Symbols.

. . . \dots ··· \cdots

. \vdots

\ddots

 \hbar ı \imath  \jmath  \ell

< \Re = \Im ℵ \aleph ℘ \wp

∀ \forall ∃ \exists  \mho

∂ \partial

’ 0 \prime ∅ \emptyset ∞ \infty

∇ \nabla 4 \triangle 2 \Box

3 \Diamond

⊥ \bot > \top ∠ \angle

√

\surd

♦ \diamondsuit ♥ \heartsuit ♣ \clubsuit ♠ \spadesuit

¬ \neg or \lnot  \flat  \natural  \sharp

Use the latexsym package to access this symbol

Table 3.11: Non-Mathematical Symbols.

These symbols can also be used in text mode.

‡ \ddag ¶ \P £ \pounds % \%

3.10 List of Mathematical Symbols 79

Table 3.12: A

S Delimiters.

p \ulcorner q \urcorner x \llcorner y \lrcorner

| \lvert | \rvert k \lVert k \rVert

Table 3.13: A

S Greek and Hebrew.

 \digamma κ \varkappa  \beth ג \gimel  \daleth

Table 3.14: Math Alphabets.

See Table 6.4 on 125 for other math fonts.

Example Command Required package

ABCDEabcde1234 \mathrm{ABCDE abcde 1234}

ABCDEabcde1234 \mathit{ABCDE abcde 1234}

ABCDEabcde \mathnormal{ABCDE abcde 1234}

ABCDE \mathcal{ABCDE abcde 1234}

A BC DE \mathscr{ABCDE abcde 1234} mathrsfs

ABCDEabcde1234 \mathfrak{ABCDE abcde 1234} amsfonts or amssymb

ABCDE \mathbb{ABCDE abcde 1234} amsfonts or amssymb

Table 3.15: A

S Binary Operators.

u \dotplus  \centerdot

 \ltimes  \rtimes  \divideontimes

d \doublecup e \doublecap  \smallsetminus

Y \veebar Z \barwedge [ \doublebarwedge

 \boxplus  \boxminus  \circleddash

 \boxtimes  \boxdot } \circledcirc

| \intercal ~ \circledast i \rightthreetimes

g \curlyvee f \curlywedge h \leftthreetimes

80 Typesetting Mathematical Formulae

Table 3.16: A

S Binary Relations.

 \lessdot  \gtrdot + \doteqdot

6 \leqslant > \geqslant : \risingdotseq

0 \eqslantless 1 \eqslantgtr ; \fallingdotseq

5 \leqq = \geqq P \eqcirc

≪ \lll or \llless ≫ \ggg $ \circeq

. \lesssim & \gtrsim , \triangleq

/ \lessapprox ' \gtrapprox l \bumpeq

≶ \lessgtr ≷ \gtrless m \Bumpeq

Q \lesseqgtr R \gtreqless ∼ \thicksim

S \lesseqqgtr T \gtreqqless ≈ \thickapprox

4 \preccurlyeq < \succcurlyeq  \approxeq

2 \curlyeqprec 3 \curlyeqsucc v \backsim

- \precsim % \succsim w \backsimeq

 \precapprox  \succapprox  \vDash

j \subseteqq k \supseteqq  \Vdash

 \shortparallel c \Supset  \Vvdash

J \blacktriangleleft A \sqsupset  \backepsilon

B \vartriangleright ∵ \because ∝ \varpropto

I \blacktriangleright b \Subset G \between

D \trianglerighteq a \smallfrown t \pitchfork

C \vartriangleleft  \shortmid ` \smallsmile

E \trianglelefteq ∴ \therefore @ \sqsubset

3.10 List of Mathematical Symbols 81

Table 3.17: A

S Arrows.

L99 \dashleftarrow 99K \dashrightarrow

⇔ \leftleftarrows ⇒ \rightrightarrows

 \leftrightarrows  \rightleftarrows

W \Lleftarrow V \Rrightarrow

 \twoheadleftarrow  \twoheadrightarrow

 \leftarrowtail  \rightarrowtail

 \leftrightharpoons  \rightleftharpoons

 \Lsh  \Rsh

" \looparrowleft # \looparrowright

 \curvearrowleft  \curvearrowright

 \circlearrowleft  \circlearrowright

( \multimap  \upuparrows

 \downdownarrows  \upharpoonleft

 \upharpoonright  \downharpoonright

\rightsquigarrow ! \leftrightsquigarrow

82 Typesetting Mathematical Formulae

Table 3.18: A

S Negated Binary Relations and Arrows.

≮ \nless ≯ \ngtr  \varsubsetneqq

 \lneq  \gneq  \varsupsetneqq

 \nleq  \ngeq  \nsubseteqq

 \nleqslant  \ngeqslant  \nsupseteqq

 \lneqq  \gneqq  \nmid

 \lvertneqq  \gvertneqq ∦ \nparallel

 \nleqq  \ngeqq  \nshortmid

 \lnsim  \gnsim  \nshortparallel

 \lnapprox  \gnapprox  \nsim

⊀ \nprec  \nsucc  \ncong

 \npreceq  \nsucceq  \nvdash

 \precneqq  \succneqq  \nvDash

 \precnsim  \succnsim  \nVdash

 \precnapprox  \succnapprox  \nVDash

 \subsetneq  \supsetneq  \ntriangleleft

 \varsubsetneq  \varsupsetneq  \ntriangleright

 \nsubseteq  \nsupseteq  \ntrianglelefteq

 \subsetneqq  \supsetneqq  \ntrianglerighteq

 \nleftarrow  \nrightarrow  \nleftrightarrow

 \nLeftarrow  \nRightarrow  \nLeftrightarrow

Table 3.19: A

S Miscellaneous.

 \hbar  \hslash k \Bbbk

 \square  \blacksquare s \circledS

M \vartriangle N \blacktriangle { \complement

O \triangledown H \blacktriangledown  \Game

♦ \lozenge  \blacklozenge F \bigstar

∠ \angle ] \measuredangle

 \diagup  \diagdown 8 \backprime

 \nexists  \Finv ∅ \varnothing

ð \eth ^ \sphericalangle  \mho

Chapter 4

Specialities

When putting together a large document, L

X will help with some special

features like index generation, bibliography management, and other things. A

much more complete description of specialities and enhancements possible with

X can be found in the L

X Manual [

] and The L

X Companion [

4.1 Including Encapsulated PostScript

X provides the basic facilities to work with ﬂoating bodies, such as

images or graphics, with the figure and table environments.

There are several ways to generate the actual graphics with basic L

or a L

X extension package, a few of them are described in chapter 5.

Please refer to The L

X Companion [

] and the L

X Manual [

] for more

information on that subject.

A much easier way to get graphics into a document is to generate them

with a specialised software package

and then include the ﬁnished graphics

in the document. Here again, L

X packages oﬀer many ways to do this,

but this introduction will only discuss the use of Encapsulated PostScript

(EPS) graphics, because it is quite easy to do and widely used. In order

to use pictures in the EPS format, you must have a PostScript printer

available for output.

A good set of commands for inclusion of graphics is provided in the

graphicx package by D. P. Carlisle. It is part of a whole family of packages

called the “graphics” bundle.

When working on a system with a PostScript printer available for

output and with the graphicx package installed, use the following step by

Such as XFig, Gnuplot, Gimp, Xara X . . .

Another possibility to output PostScript is the GhostScript program available

from

CTAN://support/ghostscript

. Windows and OS/2 users might want to look for

GSview.

CTAN://pkg/graphics

84 Specialities

step guide to include a picture into your document:

1. Export the picture from your graphics program in EPS format.

2. Load the graphicx package in the preamble of the input ﬁle with

\usepackage[driver]{graphicx}

where driver is the name of your “dvi to PostScript” converter

program. The most widely used program is called

dvips

. The name of

the driver is required, because there is no standard on how graphics

are included in T

X. Knowing the name of the driver, the graphicx

package can choose the correct method to insert information about the

graphics into the

.dvi

ﬁle, so that the printer understands it and can

correctly include the .eps ﬁle.

3. Use the command

\includegraphics[key=value, . . . ]{ﬁle}

to include ﬁle into your document. The optional parameter accepts a

comma separated list of keys and associated values. The keys can be

used to alter the width, height and rotation of the included graphic.

Table 4.1 lists the most important keys.

Table 4.1: Key Names for graphicx Package.

width scale graphic to the speciﬁed width

height scale graphic to the speciﬁed height

angle rotate graphic counterclockwise

scale scale graphic

If your software cannot export into EPS format, you can try to install a PostScript

printer driver (such as an Apple LaserWriter, for example) and then print to a ﬁle with

this driver. With some luck this ﬁle will be in EPS format. Note that an EPS must not

contain more than one page. Some printer drivers can be explicitly conﬁgured to produce

EPS format.

4.2 Bibliography 85

The following example code may help to clarify things:

\begin{figure}

\centering

\includegraphics[angle=90,

width=0.5\textwidth]{test}

\caption{This is a test.}

\end{figure}

It includes the graphic stored in the ﬁle

test.eps

. The graphic is ﬁrst

rotated by an angle of 90 degrees and then scaled to the ﬁnal width of 0.5

times the width of a standard paragraph. The aspect ratio is 1

0, because

no special height is speciﬁed. The width and height parameters can also be

speciﬁed in absolute dimensions. Refer to Table 6.5 on page 130 for more

information. If you want to know more about this topic, make sure to read

[9] and [13].

4.2 Bibliography

Produce a bibliography with the

thebibliography

environment. Each entry

starts with

\bibitem[label]{marker}

The marker is then used to cite the book, article or paper within the

document.

\cite{marker}

If you do not use the label option, the entries will get enumerated auto-

matically. The parameter after the

\begin{thebibliography}

command

deﬁnes how much space to reserve for the number of labels. In the example

below,

{99}

tells L

X to expect that none of the bibliography item numbers

will be wider than the number 99.

Partl~\cite{pa} has

proposed that \ldots

\begin{thebibliography}{99}

\bibitem{pa} H.~Partl:

\emph{German \TeX},

TUGboat Volume~9, Issue~1 (1988)

\end{thebibliography}

Partl [1] has proposed that . . .

Bibliography

[1]

H. Partl: German T

X, TUGboat

Volume 9, Issue 1 (1988)

86 Specialities

For larger projects, you might want to check out the BibT

X program.

BibT

X is included with most T

X distributions. It allows you to maintain

a bibliographic database and then extract the references relevant to things

you cited in your paper. The visual presentation of BibT

X-generated

bibliographies is based on a style-sheets concept that allows you to create

bibliographies following a wide range of established designs.

4.3 Indexing

A very useful feature of many books is their index. With L

X and the

support program

makeindex

an index can be generated quite easily. This

introduction will only explain the basic index generation commands. For a

more in-depth view, please refer to The L

X Companion [3].

To enable their indexing feature of L

X, the makeidx package must be

loaded in the preamble with

\usepackage{makeidx}

and the special indexing commands must be enabled by putting the

\makeindex

command in the preamble.

The content of the index is speciﬁed with

\index{key@formatted_entry}

commands, where formatted_entry will appear in the index and key will be

used for sorting. The formatted_entry is optional. If it is missing the key

will be used. You enter the index commands at the points in the text that

you want the ﬁnal index entries to point to. Table 4.2 explains the syntax

with several examples.

When the input ﬁle is processed with L

X, each

\index

command

writes an appropriate index entry, together with the current page number,

to a special ﬁle. The ﬁle has the same name as the L

X input ﬁle, but a

diﬀerent extension (

.idx

). This

.idx

ﬁle can then be processed with the

makeindex program:

makeindex ﬁlename

The

makeindex

program generates a sorted index with the same base ﬁle

name, but this time with the extension

.ind

. If now the L

X input ﬁle is

On systems not necessarily supporting ﬁlenames longer than 8 characters, the program

may be called makeidx.

4.4 Fancy Headers 87

Table 4.2: Index Key Syntax Examples.

Example Index Entry Comment

\index{hello} hello, 1 Plain entry

\index{hello!Peter} Peter, 3 Subentry under ‘hello’

\index{Sam@\textsl{Sam}} Sam, 2 Formatted entry

\index{Lin@\textbf{Lin}} Lin, 7 Formatted entry

\index{Kaese@K\"ase} Käse, 33 Formatted entry

\index{ecole@\’ecole} école, 4 Formatted entry

\index{Jenny|textbf} Jenny, 3 Formatted page number

\index{Joe|textit} Joe, 5 Formatted page number

processed again, this sorted index gets included into the document at the

point where L

X ﬁnds

\printindex

The showidx package that comes with

X 2

prints out all index entries

in the left margin of the text. This is quite useful for proofreading a document

and verifying the index.

Note that the

\index

command can aﬀect your layout if not used carefully.

My Word \index{Word}. As opposed

to Word\index{Word}. Note the

position of the full stop.

My Word . As opposed to Word. Note the

position of the full stop.

makeindex

has no clue about characters outside the ASCII range. To

get the sorting correct, use the

character as shown in the Käse and école

examples above.

4.4 Fancy Headers

The fancyhdr package,

written by Piet van Oostrum, provides a few simple

commands that allow you to customize the header and footer lines of your

document. Look at the top of this page, for an application of this package.

The tricky problem when customising headers and footers is to get things

like running section and chapter names in there. L

X accomplishes this

with a two-stage approach. In the header and footer deﬁnition, you use

the commands

\rightmark

and

\leftmark

to represent the current section

Available from CTAN://macros/latex/contrib/supported/fancyhdr.

88 Specialities

\documentclass{book}

\usepackage{fancyhdr}

\pagestyle{fancy}

% with this we ensure that the chapter and section

% headings are in lowercase.

\renewcommand{\chaptermark}[1]{%

\markboth{#1}{}}

\renewcommand{\sectionmark}[1]{%

\markright{\thesection\ #1}}

\fancyhf{} % delete current header and footer

\fancyhead[LE,RO]{\bfseries\thepage}

\fancyhead[LO]{\bfseries\rightmark}

\fancyhead[RE]{\bfseries\leftmark}

\renewcommand{\headrulewidth}{0.5pt}

\renewcommand{\footrulewidth}{0pt}

\addtolength{\headheight}{0.5pt} % space for the rule

\fancypagestyle{plain}{%

\fancyhead{} % get rid of headers on plain pages

\renewcommand{\headrulewidth}{0pt} % and the line

}

Figure 4.1: Example fancyhdr Setup.

4.5 The Verbatim Package 89

and chapter heading, respectively. The values of these two commands are

overwritten whenever a chapter or section command is processed.

For ultimate ﬂexibility, the

\chapter

command and its friends do not

redeﬁne

\rightmark

and

\leftmark

themselves. They call yet another

command (

\chaptermark

\sectionmark

, or

\subsectionmark

) that is re-

sponsible for redeﬁning \rightmark and \leftmark.

If you want to change the look of the chapter name in the header line,

you need only “renew” the \chaptermark command.

Figure 4.1 shows a possible setup for the fancyhdr package that makes

the headers look about the same as they look in this booklet. In any case, I

suggest you fetch the documentation for the package at the address mentioned

in the footnote.

4.5 The Verbatim Package

Earlier in this book, you got to know the

verbatim

environment. In this

section, you are going to learn about the verbatim package. The verbatim

package is basically a re-implementation of the

verbatim

environment that

works around some of the limitations of the original

verbatim

environment.

This by itself is not spectacular, but the implementation of the verbatim

package added new functionality, which is why I am mentioning the package

here. The verbatim package provides the

\verbatiminput{ﬁlename}

command, which allows you to include raw ASCII text into your document

as if it were inside a verbatim environment.

As the verbatim package is part of the ‘tools’ bundle, you should ﬁnd it

pre-installed on most systems. If you want to know more about this package,

make sure to read [10].

4.6 Installing Extra Packages

Most L

X installations come with a large set of pre-installed style packages,

but many more are available on the net. The main place to look for style

packages on the Internet is CTAN (http://www.ctan.org/).

Packages such as geometry, hyphenat, and many others are typically made

up of two ﬁles: a ﬁle with the extension

.ins

and another with the extension

.dtx

. There will often be a

readme.txt

with a brief description of the

package. You should of course read this ﬁle ﬁrst.

In any event, once you have copied the package ﬁles onto your machine,

you still have to process them in a way that (a) tells your T

X distribution

90 Specialities

about the new style package and (b) gives you the documentation. Here’s

how you do the ﬁrst part:

1. Run L

X on the .ins ﬁle. This will extract a .sty ﬁle.

Move the

.sty

ﬁle to a place where your distribution can ﬁnd it. Usually

this is in your

.../localtexmf /tex/latex

subdirectory (Windows or

OS/2 users should feel free to change the direction of the slashes).

Refresh your distribution’s ﬁle-name database. The command de-

pends on the L

Xdistribution you use: T

Xlive –

texhash

; web2c –

maktexlsr; MiKT

X – initexmf --update-fndb or use the GUI.

Now extract the documentation from the .dtx ﬁle:

Run L

X on the

.dtx

ﬁle. This will generate a

.dvi

ﬁle. Note that you

may have to run L

X several times before it gets the cross-references

right.

Check to see if L

X has produced a

.idx

ﬁle among the various ﬁles

you now have. If you do not see this ﬁle, then the documentation has

no index. Continue with step 5.

3. In order to generate the index, type the following:

makeindex -s gind.ist name

(where name stands for the main-ﬁle name without any extension).

4. Run L

X on the .dtx ﬁle once again.

Last but not least, make a

.ps

.pdf

ﬁle to increase your reading

pleasure.

Sometimes you will see that a

.glo

(glossary) ﬁle has been produced.

Run the following command between step 4 and 5:

makeindex -s gglo.ist -o name.gls name.glo

Be sure to run L

X on the

.dtx

one last time before moving on to step 5.

4.7 Working with pdfL

By Daniel Flipo <Daniel.Flipo@univ-lille1.fr>

PDF is a portable hypertext document format. Much as in a web page,

some words in the document are marked as hyperlinks. They link to other

places in the document or even to other documents. If you click on such a

hyperlink you get transported to the destination of the link. In the context

of L

X, this means that all occurrences of

\ref

and

\pageref

become

hyperlinks. Additionally, the table of contents, the index and all the other

similar structures become collections of hyperlinks.

4.7 Working with pdfL

X 91

Most web pages you ﬁnd today are written in HTML (HyperText Markup

Language). This format has two signiﬁcant disadvantages when writing

scientiﬁc documents:

Including mathematical formulae into HTML documents is not gener-

ally supported. While there is a standard for it, most browsers used

today do not support it, or lack the required fonts.

Printing HTML documents is possible, but the results vary widely

between platforms and browsers. The results are miles removed from

the quality we have come to expect in the L

X world.

There have been many attempts to create translators from L

X to

HTML. Some were even quite successful in the sense that they are able

to produce legible web pages from a standard L

X input ﬁle. But all of

them cut corners left and right to get the job done. As soon as you start

using more complex L

X features and external packages things tend to

fall apart. Authors wishing to preserve the unique typographic quality of

their documents even when publishing on the web turn to PDF (Portable

Document Format), which preserves the layout of the document and permits

hypertext navigation. Most modern browsers come with plugins that allow

the direct display of PDF documents.

Even though there are DVI and PS viewers for almost every platform,

you will ﬁnd that Acrobat Reader and Xpdf for viewing PDF documents are

more widely deployed

. So providing PDF versions of your documents will

make them much more accessible to your potential readers.

4.7.1 PDF Documents for the Web

The creation of a PDF ﬁle from L

X source is very simple, thanks to the

pdfT

X program developed by Hàn Th

ê Thành. pdfT

X produces PDF

output where normal T

X produces DVI. There is also a pdfL

X, which

produces PDF output from L

X sources.

Both pdfT

X and pdfL

X are installed automatically by most modern

X distributions, such as teT

X, fpT

X, MikT

X, T

XLive and CMacT

To produce a PDF instead of DVI, it is suﬃcient to replace the command

latex file.tex

pdflatex file.tex

. On systems where L

X is not

called from the command line, you may ﬁnd a special button in the T

GUI.

Set the paper size with an optional documentclass argument such as

a4paper

letterpaper

. This works in pdfL

X too, but on top of this

pdfT

X also needs to know the physical size of the paper to determine the

physical size of the pages in the pdf ﬁle. If you use the hyperref package (see

page 93), the papersize will be adjusted automatically. Otherwise you have

http://pdfreaders.org

92 Specialities

to do this manually by putting the following lines into the preamble of the

document:

\pdfpagewidth=\paperwidth

\pdfpageheight=\paperheight

The following section will go into more detail regarding the diﬀerences

between normal L

X and pdfL

X. The main diﬀerences concern three

areas: the fonts to use, the format of images to include, and the manual

conﬁguration of hyperlinks.

4.7.2 The Fonts

pdfL

X can deal with all sorts of fonts (PK bitmaps, TrueType, PostScript

type 1. . . ) but the normal L

X font format, the bitmap PK fonts produce

very ugly results when the document is displayed with Acrobat Reader. It is

best to use PostScript Type 1 fonts exclusively to produce documents that

display well. Modern TeX installations will be set up so that this happens

automatically. Best is to try. If it works for you, just skip this whole section.

The Type 1 font set most widely used today is called Latin Modern (LM).

If you have a recent T

X installation, chances are that you already have a

copy of them installed; all you need to do is to add

\usepackage{lmodern}

\usepackage[T1]{fontenc}

\usepackage{textcomp}

to the preamble of your document and you are all set for creating excellent

PDF output with full support for the full Latin character set. If you are

working with a stripped down setup, you may have to add the lm fonts

explicitly.

For the Russian language you may want to use C1 virtual fonts, available

ftp://ftp.vsu.ru/pub/tex/font-packs/c1fonts

. These fonts com-

bine the standard CM type 1 fonts from Bluesky collection and CMCYR

type 1 fonts from the Paradissa and BaKoMa collection, all available on

CTAN. Because Paradissa fonts contain only Russian letters, C1 fonts are

missing other Cyrillic glyphs.

Another solution is to switch to other PostScript type 1 fonts. Actually,

some of them are even included with every copy of Acrobat Reader. Because

these fonts have diﬀerent character sizes, the text layout on your pages will

change. Generally these other fonts will use more space than the CM fonts,

which are very space-eﬃcient. Also, the overall visual coherence of your

document will suﬀer because Times, Helvetica and Courier (the primary

candidates for such a replacement job) have not been designed to work in

harmony in a single document.

4.7 Working with pdfL

X 93

Two ready-made font sets are available for this purpose: pxfonts, which is

based on Palatino as its main text body font, and the txfonts package, which

is based on Times. To use them it is suﬃcient to put the following lines into

the preamble of your document:

\usepackage[T1]{fontenc}

\usepackage{pxfonts}

You may ﬁnd lines like

Warning: pdftex (file eurmo10): Font eur... not found

in the

.log

ﬁle after compiling your input ﬁle. They mean that some font

used in the document has not been found. Make sure you identify and ﬁx

the oﬀending parts of your document, as the resulting PDF document may

not display the pages with the missing characters at all.

4.7.3 Using Graphics

Including graphics into a document works best with the graphicx package

(see page 83):

\usepackage{xcolor,graphicx}

In the sample above I have included the color package, as using color in

documents displayed on the web comes quite naturally.

So much for the good news. The bad news is that graphics in Encapsulated

PostScript format do not work with pdfL

X. If you don’t deﬁne a ﬁle

extension in the

\includegraphics

command, graphicx will go looking for a

suitable ﬁle on its own, depending on the setting of the driver option. For

pdftex

this is formats

.png

.pdf

.jpg

and

.mps

(

METAPOST

)—but not

.eps.

The simple way out of this problem is to just convert your EPS ﬁles into

PDF format using the

epstopdf

utility found on many systems. For vector

graphics (drawings) this is a great solution. For bitmaps (photos, scans) this

is not ideal, because the PDF format natively supports the inclusion of PNG

and JPEG images. PNG is good for screenshots and other images with few

colours. JPEG is great for photos, as it is very space-eﬃcient.

It may even be desirable not to draw certain geometric ﬁgures, but rather

describe the ﬁgure with a specialized command language, such as