mblog3.0xtc.com is a subdomain I set up to do my MagnaBlog 3 hacks and bugfixes on. If you’re using Magnablog 3 and you want to check out things of Magna-Blogmatical interest, it’s worth taking a look at.

Soooo…you want attributes in bbcode links, huh?

Take a deep breath…and let’s think about what we really want here. We want something that thinks like a human when it sees something. When a human sees something like “…url=http://magnablog.0xtc.com title=T’s site class=alink rel=Alternate Start…”, it knows that the link is to 0xtc.com, the title is “T’s site”, the class is “alink” and the “rel” attribute is “Alternate Start”…How ? Because humans are good at patterns. Humans know that in our case, “class” isn’t “alink rel=Alternate Start” because “rel” is an attribute of its own! That’s what we want our function to do: To recognize not only the right pattern, but also where an attribute starts and more importantly, where it STOPS!

And since I couldn’t get preg_replace to recognize negative RegEx patterns….I wrote functions

function url2ahref($code)
{
$code = trim($code);
$link_text = explode("]", $code);
$code = $link_text[0];
$link_text = $link_text[1];
$linkurl = explode(" ",$code);
$linkurl = $linkurl[0];
$rest = strstr($code," ");
$rest = preg_replace ("# ([a-zA-Z]+)=#is", "\" $1=\"",$rest);
$atts = $rest;
return "<a href=\"$linkurl$atts\">$link_text</a>";
} /*……*/

$pattern = "#\[url=(.*?)\[\/url\]#ise";
$replace = "url2ahref(’$1′)";
$content = preg_replace ($pattern, $replace, $content);

/*……*/

function href2url($code)
{
$code = trim($code);
$link_text = explode(">", $code);
$code = $link_text[0];
$link_text = $link_text[1];
$linkstring = str_replace("href=","url=", $code);
$linkstring = str_replace("\\\"","", $linkstring);
return "[".$linkstring."]".$link_text."[/url]";
}

/*……*/

$pattern = ‘/<a (.*?)<\/a>/ise’;
$replace = "href2url(’$1′)";
$content = preg_replace ($pattern, $replace, $content);

/*……*/