#! perl -w
$rc=is_well_formed ("<memo>
  <to>self</to>
  <message>Don't forget to mow the car and wash the
  lawn.</message>
</memo>");
print $rc ? 'success' : 'failure';
sub is_well_formed {
    local $\ = "\n";
    my $text = shift;                     # XML text to check

    # match patterns
    my $ident = qr/[:_A-Za-z][:A-Za-z0-9\-\._]*/;   # identifier
    my $optsp = qr/\s*/;                            # optional whitespace
    my $att1 = qr/$ident$optsp=$optsp"[^"]*"/;      # attribute
    my $att2 = qr/$ident$optsp=$optsp'[^']*'/;      # attr. variant
    my $att = qr/$att1|$att2/;                      # any attribute

    my @elements = ( );                    # stack of open elems
print "Identifier $ident";
print "optsp $optsp";
print "att $att";

    # loop through the string to pull out XML markup objects
    while( length($text) ) {
print "Inside Loop";
        # match an empty element
        if( $text =~ s/^<($ident)(\s+$att)*\s*\/>//o ) {

        # match an element start tag
        } elsif( $text =~ s/^<($ident)(\s+$att)*\s*>//o ) {
            push( @elements, $1 );

        # match an element end tag
        } elsif( $text =~ s/^<\/($ident)\s*>//o ) {
            return unless( $1 eq pop( @elements ));

        # match a comment
        } elsif( $text =~ s/^<!--//o ) {
            # bite off the rest of the comment
            if( $text =~ s/^.*?--(>?)//os ) {
                return if(!$1);  # comments can't contain '--'
            } else {
                return;
            }


        # match extra whitespace
        # (in case there is space outside the root element)
        } elsif( $text =~ s/^\s+//o ) {

        # match character data
        } elsif( $text =~ s/^([^<&>]+)//o ) {
print "char data";
            my $data = $1;
            # make sure the data is inside an element
            return if( $data =~ /\S/ and not( @elements ));

        # match entity reference
        } elsif( $text =~ s/^&$ident;//o ) {

        # match numerical entity reference
        # added by bart
        } elsif( $text =~ s/^&#(?:\d+|x[0-9a-f]+);//io ) {

        # something unexpected
        } else {
            return;
        }
    }
    return if( @elements );     # the stack should be empty
    return 1;

}

##</code><code>##

Identifier (?-xism:[:_A-Za-z][:A-Za-z0-9\-\._]*)
optsp (?-xism:\s*)
att (?-xism:(?-xism:(?-xism:[:_A-Za-z][:A-Za-z0-9\-\._]*)(?-xism:\s*)=(?-xism:\s*)"[^"]*")|(?-xism:(?-xism:[:_A-Za-z][:A-Za-z0-9\-\._]*)(?-xism:\s*)=(?-xism:\s*)'[^']*'))
Inside Loop
Inside Loop
Inside Loop
Inside Loop
char data
Inside Loop
Inside Loop
Inside Loop
Inside Loop
char data
Inside Loop
Inside Loop
Inside Loop
success