The Perl Advent Calendar needs more articles for 2022. Submit your idea today!


LaTeX::Encode - encode characters for LaTeX formatting


  use LaTeX::Encode ':all', add => { '@' => 'AT' }, remove => [ '$' ];

  $latex_string  = latex_encode($text, %options);

  %old_encodings = add_latex_encodings( chr(0x2002) => '\\hspace{.6em}' );
  %old_encodings = remove_latex_encodings( '<', '>' );



This manual page describes version 0.091.5 of the LaTeX::Encode module.


This module provides a function to encode text that is to be formatted with LaTeX. It encodes characters that are special to LaTeX or that are represented in LaTeX by LaTeX text-mode commands.

The special characters are: \ (command character), { (open group), } (end group), & (table column separator), # (parameter specifier), % (comment character), _ (subscript), ^ (superscript), ~ (non-breakable space), $ (mathematics mode).

Note that some of the LaTeX commands for characters are defined in the LaTeX textcomp package. If your text includes such characters, you will need to include the following lines in the preamble to your LaTeX document.


The function is useful for encoding data that is interpolated into LaTeX document templates, say with Template::Plugin::Latex (shameless plug!).


Note that latex_encode() will encode a UTF8 string (a string with the UTF8 flag set) or a non-UTF8 string, which will normally be regarded as ISO-8859-1 (Latin 1) and will be upgraded to UTF8. The UTF8 flag indicates whether the contents of a string are regarded as a sequence of Unicode characters or as a string of bytes. Refer to the Unicode Support in Perl, Perl Unicode Introduction and Perl Unicode Tutorial manual pages for more details.

If you are seeing spurious LaTeX commands in the output of latex_encode() then it may be that you are reading from a UTF-8 input or have data with UTF-8 characters in a literal but the UTF8 flag is not being set correctly. The fact that your programs are dealing with UTF-8 characters on a byte-by-byte basis may not be apparent normally as the terminal may make no distinction and happily display the byte sequence in the program's output as the UTF-8 characters they represent, however in a Perl program that deals with individual characters, what happens is that the individual bytes that make up multi-byte characters are regarded as separate characters; if the strings are promoted to UTF8 strings then the individual bytes are converted separately to UTF8. This is termed double encoding. latex_encode() will then map the double-encoded characters.

If the input text is Western European text then what you are likely to see in the output from latex_encode() is spurious sequences of {\^A} or {\~A} followed by the mapping of an apparently random character (or the right character if it is a symbol such as the Sterling POUND sign, i.e. "£" will map to {\^A}\textsterling); this is because the initial byte of a two-byte UTF-8 character in the LATIN1 range will either be 0xC2 or 0xC3 and the next byte will always have the top two bits set to 10 to indicate that it is a continuation byte.


latex_encode($text, %options)

Encodes the specified text such that it is suitable for processing with LaTeX. The behaviour of the filter is modified by the options:


Lists the characters that should be excluded from encoding. By default no special characters are excluded, but it may be useful to specify except = "\\{}" to allow the input string to contain LaTeX commands such as "this is \\textbf{bold} text" (the doubled backslashes in the strings represent Perl escapes, and will be evaluated to single backslashes).


If true then single or double quotes around words will be changed to LaTeX single or double quotes; double quotes around a phrase will be converted to "``" and "''" and single quotes to "`" and "'". This is sometimes called "intelligent quotes"


If passed a reference to a hash latex_encode() will update the hash with names of LaTeX packages that are required for typesetting the encoded string.


Adds a set of new or modified encodings. Returns a hash of any encodings that were modified.


Removes a set of encodings. Returns a hash of the removed encodings.


Resets the LaTeX encodings to the state that they were when the module was loaded (including any additions and removals specified on the 'use' statement), or to the standard set of encodings if $forget_import_specifiers is true.


The following snippet shows how data from a database can be encoded and inserted into a LaTeX table, the source of which is generated with LaTeX::Table.

    my $sth = $dbh->prepare('select col1, col2, col3 from table where $expr');
    while (my $href = $sth->fetchrow_hashref) {
        my @row;
        foreach my $col (qw(col1 col2 col3)) {
            push(@row, latex_encode($href->{$col}));
        push @data, \@row;

    my $headings = [ [ 'Col1', 'Col2', 'Col3' ] ];

    my $table = LaTeX::Table->new( { caption => 'My caption',
                                     label   => 'table:caption',
                                     type    => 'xtab',
                                     header  => $header,
                                     data    => \@data } );

    my $table_text = $table->generate_string;    

Now $table_text can be interpolated into a LaTeX document template.


None. You could probably break the latex_encode function by passing it an array reference as the options, but there are no checks for that.


Not applicable.


The HTML::Entities and Pod::LaTeX modules were used for building the encoding table but this is not rebuilt at installation time. The LaTeX::Driver module is used for formatting the character encodings reference document.


None known.


Not all LaTeX special characters are included in the encoding tables (more may be added when I track down the definitions).


Andrew Ford <>


Copyright (C) 2007-2012 Andrew Ford. All Rights Reserved.

This module is free software; you can redistribute it and/or modify it under the same terms as Perl itself.

This software 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.



Unicode Support in Perl

Perl Unicode Introduction

Perl Unicode Tutorial