File Coverage

blib/lib/DOM/Tiny/Entities.pm

Criterion	Covered	Total	%
statement	20	20	100.0
branch	6	6	100.0
condition			n/a
subroutine	6	6	100.0
pod	2	2	100.0
total	34	34	100.0

package DOM::Tiny::Entities;

use strict;
use warnings;
use Exporter 'import';

our $VERSION = '0.003';

our @EXPORT_OK = qw(html_escape html_unescape);

# To generate a new HTML entity table run this command
# perl examples/entities.pl
my %ENTITIES;
for my $line (split "\n", join('', )) {
  next unless $line =~ /^(\S+)\s+U\+(\S+)(?:\s+U\+(\S+))?/;
  $ENTITIES{$1} = defined $3 ? (chr(hex $2) . chr(hex $3)) : chr(hex $2);
}

# Characters that should be escaped in HTML/XML
my %ESCAPE = (
  '&'  => '&',
  '<'  => '<',
  '>'  => '>',
  '"'  => '"',
  '\'' => '''
);

sub html_escape {
  my $str = shift;
  $str =~ s/([&<>"'])/$ESCAPE{$1}/ge;
  return $str;
}

sub html_unescape {
  my $str = shift;
  $str =~ s/&(?:\#((?:[0-9]{1,7}|x[0-9a-fA-F]{1,6}));|(\w+;))/_decode($1, $2)/ge;
  return $str;
}

sub _decode {
  my ($point, $name) = @_;

  # Code point
  return chr($point !~ /^x/ ? $point : hex $point) unless defined $name;

  # Named character reference
  return exists $ENTITIES{$name} ? $ENTITIES{$name} : "&$name";
}

1;

=encoding utf8

=head1 NAME

DOM::Tiny::Entities - Escape or unescape HTML entities in strings

=head1 SYNOPSIS

  use DOM::Tiny::Entities qw(html_escape html_unescape);

  my $str = 'foo & bar';
  $str = html_unescape $str; # "foo & bar"
  $str = html_escape $str; # "foo & bar"

=head1 DESCRIPTION

L contains functions for escaping and unescaping HTML
entities for L, based on functions from L. All functions
are exported on demand.

=head1 FUNCTIONS

=head2 html_escape

 my $escaped = html_escape $str;

Escape unsafe characters C<&>, C<< < >>, C<< > >>, C<">, and C<'> in string.

 html_escape ''; # "<div>"

=head2 html_unescape

 my $str = html_unescape $escaped;

Unescape all HTML entities in string, according to the
L.

 html_unescape '<div> # ""

=head1 BUGS

Report any issues on the public bugtracker.

=head1 AUTHOR

Dan Book

=head1 COPYRIGHT AND LICENSE

This software is Copyright (c) 2015 by Dan Book.

This is free software, licensed under:

  The Artistic License 2.0 (GPL Compatible)

=head1 SEE ALSO

L

=cut

__DATA__