How to clean version control $Id$ comment

This might be a common question, but search engines are horrible at searching for non-alphanumeric characters.

I have a bunch of code which contains the standard version control $Id$ tag/comment. I need an easy way to remove (clean, strip), all of these tags from every file in a directory structure.

Specifically this is a PHP script (phpBB to be specific), and so looks like this:

<?php
/**
*
* @package acp
* @version $Id: acp_attachments.php 8479 2008-03-29 00:22:48Z naderman $
*
*/
?开发者_Go百科>

The line I want to change is this:

* @version $Id: acp_attachments.php 8479 2008-03-29 00:22:48Z naderman $

So the line becomes:

* @version $Id$

Keep in mind, I'm on Windows, so I can't use a Linux command for find/replace. But I do have the ability to run PHP to act on all my files.

On windows you can use grepWin using this regex:

Search for: \$Id\: .+ \$$

Replace with: \$Id\$

NOTE: first make a backup of all files before doing this

Can you use Perl? Here's my cvs-clean script that I use for this:

#!/usr/bin/perl

use strict;
use warnings;

while (<>) {
    s/\$([A-Z][a-z]+):[^\$]*\$/\$$1:\$/g;
    print;
}

It filters stdin to stdout. It's not exactly what you asked for; it changes $Id: blah $ to $Id:$. Delete the last : in the s/// command to change that.

If you can't use Perl, translating this to PHP is left as an exercise.