File Coverage

blib/lib/Regexp/IgnoreHTML.pm

Criterion	Covered	Total	%
statement	39	39	100.0
branch	9	10	90.0
condition	7	12	58.3
subroutine	4	4	100.0
pod	3	3	100.0
total	62	68	91.1

package Regexp::IgnoreHTML;
use Regexp::Ignore;
our @ISA = ("Regexp::Ignore"); # inherit from Regexp::Ignore class

########################
# new
########################
sub new {
    my $proto = shift;
    my $class = ref($proto) || $proto;
    my $self = $class->SUPER::new(@_);
    # by default it does not add spaces
    $self->{SPACE_AFTER_NON_TEXT_CHARACTERISTICS_HTML} = 0;
    return $self;
} # of new

#############################################
# space_after_non_text_characteristics_html
#############################################
sub space_after_non_text_characteristics_html {
    my $self = shift;
    if (@_) { $self->{SPACE_AFTER_NON_TEXT_CHARACTERISTICS_HTML} = shift }
    return $self->{SPACE_AFTER_NON_TEXT_CHARACTERISTICS_HTML};
} # of space_after_non_text_characteristics_html

###########################################
#
#
#
########################
# get_tokens
########################
sub get_tokens {
    my $self = shift;

    my $tokens = [];
    my $flags = [];
    my $index = 0;
    # we should create tokens from the TEXT.
    my $text = $self->text();

    # the regular expression will try to match:
    #  - HTML remarks - all the remark will be matched.
    #  - HTML tags
    my $re1 = qr/(<\!\-\-[\s\S]+?\-\->)|(<\/?[^\>]*?>)/is;

    my $re2;
    if ($self->space_after_non_text_characteristics_html()) {
	# if the tag that we found is one of the following, we do not
	# put space after it: B, BASEFONT, BIG, BLINK, CITE, CODE, EM,
	# FONT, I, KBD, PLAINTEXT, S, SMALL, STRIKE, STRONG, SUB, SUP,
	# TT, U, VAR, A, SPAN, WBR
	$re2 = '<\!\-\-.+?\-\->|'.
	    '\<\!\[[^\]]*?\]\>|'.
	    '<\/?\s*B(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*BASEFONT(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*BIG(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*BLINK(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*CITE(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*CODE(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*EM(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*FONT(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*I(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*KBD(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*PLAINTEXT(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*S(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*SMALL(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*STRIKE(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*STRONG(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*SUB(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*SUP(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*TT(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*U(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*VAR(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*A(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*SPAN(\s[^>]*?>|\s*>)|'.
 	    '<\/?\s*WBR(\s[^>]*?>|\s*>)|'.
	    '<\/?\s*[OVWXP]\:[^>]*?>';
	$re2 = qr/$re2/is;
    }

    while (defined($text) && $text =~ /$re1/) {
	if ($`) { # if there is a text before, take it as clean
	    $tokens->[$index] = $`;
	    $flags->[$index] = 1; # the text before the match is clean.
	    $index++; # increment the index
	}
	$tokens->[$index] = $&;
	$flags->[$index] = 0; # the match itself is unwanted.
	$text = $'; # update the original text to after the match.
	$index++; # increment the index again
	# check if we should add space after the text
	if ($self->space_after_non_text_characteristics_html() &&
	    $tokens->[$index - 1] !~ /$re2/) { # this tag is not text
	                                       # characteristic tag
	    # we add a space token after this tag
	    $tokens->[$index] = " ";
	    $flags->[$index] = 1;
	    $index++;
	}
    }

    # if we had no match, check if there is still something in the
    # $text. this will be also a clean text.
    if (defined($text) && $text) {
	$tokens->[$index] = $text;
	$flags->[$index] = 1;
    }
    # return the two lists
    return ($tokens, $flags);
} # of get_tokens

1; # make perl happy

__END__