Replacing Characters On Condition

megamanXplosion

Hi you guys 🙂

I have a message board (invision board) and I've been trying to enable javascripting for admins, so that I can write some tutorials and be able to implement working examples inside of the posts. I've figured part of it out, but there's still a problem.

I have figured out how to get the parser to replace [script] and [/script] to <script> and </script>. Now, here's the problem.

I need to be able to replace all linebreaks \n with <br> but not in between the [script] and [script] tags. Here's what converts all line breaks....

$txt = preg_replace( "/\n/", "<br>", $txt );

could someone please rewrite that so that it won't replace anything inside of [script] and [/script]?

Thanks.

megamanXplosion

I think you misunderstood me. Let's say that I was going to write this code into a post....

[script]
document.write("Hello World!");
[/script]

when it is parsed and submitted, the code would be outputted like so....

<script><br>
document.write("Hello World");<br>
</script>

I am wanting to have all of the line breaks replaced with the <*br> tag, except for those that are inbetween the [script] tags so that it won't cause any errors when someone views the post.

megamanXplosion

anyone come up with anything?

lanbor

suppose your code just like this:

[script]
document.write("Hello World!");
[/script]

when it is parsed and submitted, the code would be outputted like so....

then you can replace all <br> linebreaks with space which in "<script>.....</script>".

sorry for my poor english. 🙂

Weedpacket

I'm guessing that it could be done with a single regexp, but I'm guessing it will be messy and I'm a bit wary of trying right now.

I thought of doing this (after converting [script] to <script>:

$inside_script = false;
$lines = explode("\n",$text);
foreach($lines as $key => $line)
{   if($inside_script && strpos($line, '</script>')!==false)
        $inside_script = false;
    elseif(!$inside_script && strpos($line, '<script')!==false) // no >, in case of attributes
        $inside_script = true;                             // if you don't use them, then
                                                           // it can go in
    $lines[$key] .= $inside_script ? "\n" : '<br>';
}
$text = join('',$lines);

I just dashed that off the top of my head - might not be the most
efficient, might not work 🙂. - edit, use strpos instead of preg_match

Andy_B2

Okay, looked at this for a challenge - and it is.

Best I could do was:

$str = "
hello
hi
script
some lines
hi again
script
more lines
and more...
";

echo preg_replace( "/(?<=script(.))\n(?=(.)script)/", '<br>', $str);

(?<=script(.))\n(?=(.)script)

This means:
(?<= if Looking back
script(.) I match script then anything
\n and now a new line
(?= and looking forward
(.)script I match something then 'script'

then match.

This doesn't work though, as the look back does not have a fixed length. Perl only supports fixed length look backs.

Best bet, as suggested above, swap all \n to <br>, then back the <br> in a script tags and make them \n again. Or you might be able to use the above, modify it in some way.

Weedpacket

Originally posted by Andy B

Best bet, as suggested above, swap all \n to <br>, then back the <br> in a script tags and make them \n again. Or you might be able to use the above, modify it in some way.

But this would clobber any <br> tags already inside <script> tags (string literals?). I'm trying to think of a situation where one might want to use </script> inside a <script>...</script> pair for (who uses client-side scripts to write client-side script entities?)...

choppsta

Here's something I used to convert all newlines to BRs in a wodge of text ($details) except for anything enclosed in <html> tags.

It works by stripping out all the <html> tags and what's inbetween, replacing them with markers (), converting the newlines and then plonking the html code back in.

if (preg_match_all("{\\<html\\>(.*)\\<\\/html\\>}siU", $details, $temp)) {
	while (list($k, $v) = each($temp[0])) {
		$details = str_replace($v, '<!-- h'.$k.'h -->', $details);
		$html[$k] = $temp[1][$k];
	}
	$details = str_replace("\n", "<br>", $details);
	while (list($k, $v) = each($html)) {
		$details = str_replace('<!-- h'.$k.'h -->', $v, $details);
	}
}
else {
	$details = str_replace("\n", "<br>", $details);
}

There's probably better ways but it works!

megamanXplosion

I've tried everything on this page and nothing works. I'm going to attach the post_parser.php code in this post and hopefully one of you guys can rewrite it a little bit to make it work. The part where it converts line breaks is on line 259.

choppsta

Maybe I misunderstood what you're trying to achieve, but I'm sure my code works, use the script below to test it:

<?php

function parsescript($details) {
	if (preg_match_all("{\\[script\\](.*)\\[\\/script\\]}siU", $details, $temp)) {
		while (list($k, $v) = each($temp[0])) {
			$details = str_replace($v, '<!-- h'.$k.'h -->', $details);
			$html[$k] = $temp[1][$k];
		}
		$details = str_replace("\n", "<br>", $details);
		while (list($k, $v) = each($html)) {
			$details = str_replace('<!-- h'.$k.'h -->', '<script>'.$v.'</script>', $details);
		}
	}
	else {
		$details = str_replace("\n", "<br>", $details);
	}
	return $details;
}

if ($details) {
	$orig = $details;
	$details = parsescript($details);
}

echo '<html><body><form action="'.$PHP_SELF.'" method="POST">
<b>Results:</b><br>'.htmlspecialchars($details).'<br><br>
<textarea name="details" cols="50" rows="20">'.$orig.'</textarea><br>
<input type="submit"></form></body></html>';

?>

Andy_B2

Yep, that's useful. Thanks. I'm going to snip that one...

KoshNaranek

what about nl2br() ?? 😃

choppsta

While we're on the subject, the following should really be used because of the differing newline characters on different operating systems:

$details = ereg_replace("(\r\n|\n|\r)", "<br>", $details);