Regex in preg_replace to detect url format and extract elements

Question

I need to replace certain user-entered URLs with embedded flash objects...and I'm having trouble with a regex that I'm using to match the url...I think mainly because the URLs are SEO-friendly and therefore a bit more difficult to parse

URL structure: http://www.site.com/item/item_title_that_can_include_1('_etc-32CHARACTERALPHANUMERICGUID

I need to both detect a match of an URL in that format and capture the 32CHARACTERALPHANUMERICGUID which is always placed after the - in the url

something like this:

$ret = preg_replace('#http://www\.site\.com/item/([^-])-([a-zA-Z0-9]+)#','<embed>itemid=$2</embed>', $ret);

For some reason, the above does not find a match for an URL in the specified format. I'm new to regexes, so I think I'm missing something fairly obvious.

Answer 1

You should check out parse_url().

Examine the results - it was made for parsing URLs. You'll be able to extract the data you require from the tokens returned.

If you are regex crazy, try this...

/^http:\/\/www\.site\.com\/item\/[^-]*\-([a-zA-Z0-9]{32})$/

Your example is almost there, but...

• When you do the not character range, i.e. [^-], you still need a quantifier. I placed *, or 0 or more.
• You don't seem to use the item title, so we won't bother capturing it.
• You should use beginning (^) and end ($) anchors if the string is always exactly like that.
• You say the GUID is 32 chars, so we may as well explicitly state that with the {32} quantifier.