Regex $1, $2, etc

Question

I've been trying to do some regex operations in PHP, and I'm not very skilled in this area. It seems that when I use a regex function like preg_replace on a string, I can access the regex-replaced strings by some sort of variables named $1, $2, and so on. What is this called and how can I use it?

Answer 1

These are known in regex terminology as backreferences (more on that here). You use them to refer to capture groups (or subpatterns, surrounded by ()) within your regex or in the replacement string.

An example:

/*   * Replaces abcd123 with 123abcd, or asdf789 with 789asdf.  *   * The $1 here refers to the capture group ([a-z]+),  * and the $2 refers to the capture group ([0-9]+).  */ preg_replace('/([a-z]+)([0-9]+)/', '$2$1', $str); 

Answer 2

They are called backreferences and match grouped elements within the regexp.

If you surround a section of the regexp with brackets, then you can refer to it in the replace section (or indeed later in the same regexp, by the backreference which corresponds to its position.

Slash form, or dollar form can be used in replacements:

\1, \2 == $1, $2