preg_replace() dollar sign in replacement issue

[Brendan_Nolan]

For some reason the following

echo preg_replace("{<!-- *?SC:body:SC *?-->}", 'text text $687 text text', 'text <!-- SC:body:SC --> text text');

returns

  text text text 7 text text text text

instead of the expected

  text text text $687 text text text text

I understand that $ is a special word in the regular expressions world, but I thought that the regex is the first argument, and then the next two should be treated as plain text?

Anyway, I think that it is stupid - It has caused me a lot of trouble!

Is there any reason for the second and third arguments running through the regex engine?

The soloution - preg_quote()

[Brendan_Nolan]

Actually IT IS NOT the solution...

It did prevent the $ from causing problems.

But it looks like preg_match() may only be messing around with the $

Since it preg_quote() escapes all the other special characters as well it is leaving a whole heap of backslashes in my document.

Can somebody please tell me what I am doing wrong? I had carets, dots, exclamation marks, equals signs and question marks in there and they worked fine - just not the damn dollar sign!

[Brendan_Nolan]

Okay, I have done some tests, and it looks like all that I need to do is escape the dollar signs and backslashes. The other metacharacters seem fine, and when I escape them the backslash still remains in the string after the preg_replace().

However I am still confident that just escaping \ and $ is the best solution... How am I meant to to it?