Making Lists in vB XHTML 1.0 Strict Compliant

[Faruk]

For those who care, here is a way to have LISTS in vBulletin be strict standards compliant (XHTML 1), and it requires only a little bit of hacking

Currently:

[ list ]
[ * ] boo
[ / list ]

would create:

<ul>
<li> boo
</ul>

which will not pass the validator for lacking a </li> tag. If you'd hack the [ list ] code replacement bits, you could do this:

[ list ]    -> <ul><li>
[ * ]      -> </li><li>
[ / list ]  -> </li></ul>

Now you would have this:

<ul>
<li></li>
<li> boo </li>
</ul>

Then you just add a replacement for "<li></li>" to "". That would create:

<ul>
<li> boo </li>
</ul>

Voila, XHTML 1.0 Strict Standards Compliant Lists.

[Lumina]

With this :
[ list ]
[ * ] boo
[ / list ]

You mean you would have this :
<list><li><br /></li><li>boo<br /></li></list>

You need to remove <li></li>, but <li><br /></li> and <li><br /><br /></li> too. In short, remove <li>(<br />)*</li>.

Also, adding the easy [li] tag will help you with missing line breaks.

[Faruk]

With this :
[ list ]
[ * ] boo
[ / list ]

You mean you would have this :
<list><li><br /></li><li>boo<br /></li></list>

You need to remove <li></li>, but <li><br /></li> and <li><br /><br /></li> too. In short, remove <li>(<br />)*</li>.

Also, adding the easy [li] tag will help you with missing line breaks.

Hmyeah, good point about the removal and <br />'s.

However, using the [li] tag will only be of use WHEN YOUR USERS ACTUALLY USE IT, and since it's much easier for them to use[*] or the vBQuick Code buttons for Lists (which uses[*] as well), you shouldn't rely on it. Also, it doesn't save you from existing posts that have[*].

If all of my users would only be able to use [li] & [/li] for lists, I'd consider it, but right now, I prefer my slight alteration to the parser to make sure, 100%, that it's valid

[Faruk]

Hmm, it depends on where the nl2br() is done, of course... I should look into that...

[Lumina]

If you find a way to prevent <br /> from being removed before a[*], please tell me how you changed the code !

[Faruk]

owkies

mind you I haven't actually LOOKED at the code of this yet, ever.

[Lumina]

Got a solution to resolve the issue with lists in Beta 5.
Watch closely.

File functions_bbcodeparse.php

//Comment the two following lines to avoid lines being removed before block elements

		$whitespacefind = array(
[B][COLOR=green]//[/COLOR][/B]			'#([\r\n]+)( )*(\[\*\]|\[/list|\[list|\[indent|\[quote|\[code|\[php)#si',
			'#(/list\]|/indent\]|/quote\]|/code\]|/php\])( )*([\r\n]+)#si'
		);
		$whitespacereplace = array(
[B][COLOR=green]//[/COLOR][/B]			'\3',
			'\1'
		);

File functions_wysiwyg.php

//The IE Wysiwyg has a bug, it displays too much lines.
//So we remove a line before each[*] and[/list]

[B][COLOR=green]	$html = preg_replace('#((\r\n)|\n|\r)(\[\*\]|\[/list\])#si', '\3', $html);[/COLOR][/B]
	return parse_bbcode($html, $forumid, $allowsmilie, '', 0, 1);

//Remove the use of rtrim() to avoid line breaks being removed at the end of each element

[B][COLOR=green]//[/COLOR][/B]	return '[*]' . rtrim($text);
[B][COLOR=green]	return '[*]' . $text;[/COLOR][/B]

//This line looks really useless, we remove it
//Then we put back line breaks that we removed before[*] and[/list]

[B][COLOR=green]//[/COLOR][/B]	$text = preg_replace('#<li>(.*)((?=<li>)|</li>)#iU', "\\1\n", $text);
[B][COLOR=green]	$text = preg_replace('#(\[\*\]|\[/list\])#si', "\n\\1", $text);
	$text = preg_replace('#\[list\]\n\[\*\]#si', "[list][*]", $text);[/COLOR][/B]

[Stadler]

Well, anything between [ list ] and the first [ * ] after that could simply be stripped, I suppose:

$text = preg_replace("#(\[list\]).*(\[\*\])#iU","\\1\\2",$text); 

should do the trick