view release on metacpan or  search on metacpan

lib/XML/Parser/Lite.pm  view on Meta::CPAN

# ======================================================================
#
# Copyright (C) 2000-2007 Paul Kulchenko (paulclinger@yahoo.com)
# Copyright (C) 2008 Martin Kutter (martin.kutter@fen-net.de)
# XML::Parser::Lite is free software; you can redistribute it
# and/or modify it under the same terms as Perl itself.
#
# ======================================================================

package XML::Parser::Lite;

use 5.006;
use strict;
use warnings;

our $VERSION = '0.722';

sub new {
    my $class = shift;

    return $class if ref $class;
    my $self = bless {} => $class;

    my %parameters = @_;
    $self->setHandlers(); # clear first
    $self->setHandlers(%{$parameters{Handlers} || {}});

    return $self;
}

sub setHandlers {
    my $self = shift;

    # allow symbolic refs, avoid "subroutine redefined" warnings
    no strict 'refs';  ## no critic
    no warnings qw(redefine);
    # clear all handlers if called without parameters
    if (not @_) {
        for (qw(Start End Char Final Init Comment Doctype XMLDecl)) {
            *$_ = sub {}
        }
    }

    # we could use each here, too...
    while (@_) {
        my($name, $func) = splice(@_, 0, 2);
        *$name = defined $func
            ? $func
            : sub {}
    }
    return $self;
}

sub _regexp {
    my $patch = shift || '';
    my $package = __PACKAGE__;

    # This parser is based on "shallow parser" http://www.cs.sfu.ca/~cameron/REX.html

    # Robert D. Cameron "REX: XML Shallow Parsing with Regular Expressions",
    # Technical Report TR 1998-17, School of Computing Science, Simon Fraser University, November, 1998.
    # Copyright (c) 1998, Robert D. Cameron.
    # The following code may be freely used and distributed provided that
    # this copyright and citation notice remains intact and that modifications
    # or additions are clearly identified.

    # Modifications may be tracked on XML::Parser::Lite's source code repository at
    # https://github.com/redhotpenguin/perl-XML-Parser-Lite
    #
    use re 'eval';
    my $TextSE = "[^<]+";
    my $UntilHyphen = "[^-]*-";
    my $Until2Hyphens = "([^-]*)-(?:[^-]$[^-]*-)*-";
    #my $CommentCE = "$Until2Hyphens(?{${package}::comment(\$2)})>?";
    my $CommentCE = "(.+)--(?{${package}::comment(\$2)})>?";
#    my $Until2Hyphens = "$UntilHyphen(?:[^-]$UntilHyphen)*-";
#    my $CommentCE = "$Until2Hyphens>?";
    my $UntilRSBs = "[^\\]]*](?:[^\\]]+])*]+";
    my $CDATA_CE = "$UntilRSBs(?:[^\\]>]$UntilRSBs)*>";
    my $S = "[ \\n\\t\\r]+";
    my $NameStrt = "[A-Za-z_:]|[^\\x00-\\x7F]";
    my $NameChar = "[A-Za-z0-9_:.-]|[^\\x00-\\x7F]";