I'm trying to write a utility to parse some PHP (w/HTML) files using HTML::TokeParser::Simple (HTS). I need to keep track of any problems I find by line number so I've opted loop through the file and parse each line with HTS using the string constructor. (This is the only way to know where you are within the file.) One of the first things I tried to detect is when the file goes into PHP mode by looking for processing instructions (is_pi). When I do this, it only detects 2 of the 5 occcurances. But if I rewrite it to parse the file via HTS's file constructor it finds all 5! I then wrote a quick test using HTML::TokeParser which indicates that the 2 missing PI's are being detected as comments. How is it that HTS using a file constructor works properly but yet the HTS string constructor and TokeParser don't work? At least if it was consistent that would make sense, but it's not. Here's some code that illustrates this situation:
#!/usr/bin/perl -w
use strict;
use warnings;
use Data::Dumper;
use HTML::TokeParser::Simple;
$|++;
#--- Variables
my $file = 'test2.php';
my $line_number;
print "#--- TokeParser::Simple String\n";
$line_number = 0;
open (FH, "<$file") or die "Unable to open $file $!\n";
while (<FH>)
{
chomp;
$line_number++;
my $p = HTML::TokeParser::Simple->new(string => $_);
while (my $token = $p->get_token) {
if ($token->is_pi) {
print $line_number . 'P: ' . $token->get_token0 . "\n";
}
if ($token->is_comment) {
print $line_number . 'C: ' . $token->as_is . "\n";
}
}
}
close (FH);
##############################
print "\n#--- TokeParser::Filehandle\n";
my $fh;
open ($fh, "<$file") or die "Unable to open $file $!\n";
$line_number++;
my $p = HTML::TokeParser::Simple->new(handle => $fh);
while (my $token = $p->get_token) {
if ($token->is_pi) {
print $line_number . 'P: ' . $token->get_token0() . "\n";
}
if ($token->is_comment) {
print $line_number . 'Pc: ' . $token->as_is . "\n";
}
}
close ($fh);
##############################
print "\n#--- TokeParser\n";
$line_number = 0;
open (FH, "<$file") or die "Unable to open $file $!\n";
while (<FH>)
{
chomp;
$line_number++;
print "LINE: $line_number ********\n";
my $p = HTML::TokeParser->new(\$_);
while (my $token = $p->get_token) {
print Dumper($token) . "\n";
}
}
close (FH);
__END__
Here is the code I'm running it against (test2.php):
<? /**** $Id$ ****/
?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http:/
+/www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<HTML>
<BODY>
<P>Some text<BR />
</P>
<p><? echo U($date);?></p>
<?
echo 'This is from PHP.';
?>
<h1><? echo $h1; ?></h1>
<?php
?>
</body>
</html>
And the output I get:
#--- TokeParser::Simple String
1C: <? /**** $Id$ ****/
9P: echo U($date);?
10C: <?
13P: echo $h1; ?
14C: <?php
#--- TokeParser::Filehandle
18P: /**** $Id$ ****/
?
18P: echo U($date);?
18P:
echo 'This is from PHP.';
?
18P: echo $h1; ?
18P: php
?
#--- TokeParser
LINE: 1 ********
$VAR1 = [
'C',
'<? /**** $Id$ ****/'
];
--- SNIP! ---
LINE: 10 ********
$VAR1 = [
'C',
'<?'
];
--- SNIP! ---
LINE: 14 ********
$VAR1 = [
'C',
'<?php'
];
--- SNIP! ---
This shows HTML::TokeParser sees those three lines as comments not as
+PI's.
2005-02-24 Janitored by Arunbear - added readmore tags, as per Monastery guidelines
