Question

I can’t seem to figure out what I’m doing wrong…

I’m trying to find matches of

<cite>stuffhere</cite>

Is this right?

preg_match_all('<cite>(.*?)</cite>/ms', $str, $matches)

Answer 1

Your confusion is not your fault; PHP is notoriously weird in this area.

In most programming languages, you create a regex object one of two ways. If the language supports regexes as a first-class language element, you can use a regex literal:

var re = /<b>"\w+"<\/b>/;  // JavaScript

Here, the forward-slash (/) is the regex delimiter; if you want to match a literal /, you have to escape it with a backslash: \/.

In other languages, you have to write the regex in the form of a string literal, which you then pass to a constructor or a factory method:

Pattern p = Pattern.compile("<b>\"\\w+\"</b>");  // Java

The forward-slash doesn’t need to be escaped, but both the double-quote (") and backslash (\) do, because of their special meanings in string literals.

But PHP is unique: it doesn’t support regex literals, so you have to write the regex as a string, but the string has to look like a regex literal! That is, it has to have string delimiters (quotes) and regex delimiters. For example:

$re = '/<b>"\w+"<\/b>/';

It isn’t all bad; as you can see, you can use PHP’s single-quoted strings instead of double-quoted, so you don’t have to escape all backslashes and double-quotes. You can also choose different regex delimiters, so you don’t have to escape (for example) literal forward-slashes in your regex:

$re = '~<cite>(.*?)</cite>~s'

The modifiers (‘s’ for single-line, ‘i’ for ignore-case, etc.) go after the trailing regex delimiter, as in Perl or JavaScript. Almost any ASCII punctuation character can be used as a regex delimiter; ~ and # are popular choices.