Return-Path: 
 <dev-return-22117-apmail-httpd-dev-archive=httpd.apache.org@httpd.apache.org>
Delivered-To: apmail-httpd-dev-archive@httpd.apache.org
Received: (qmail 5567 invoked by uid 500); 12 Oct 2001 18:17:00 -0000
Mailing-List: contact dev-help@httpd.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@httpd.apache.org
list-help: <mailto:dev-help@httpd.apache.org>
list-unsubscribe: <mailto:dev-unsubscribe@httpd.apache.org>
list-post: <mailto:dev@httpd.apache.org>
Delivered-To: mailing list dev@httpd.apache.org
Received: (qmail 5544 invoked from network); 12 Oct 2001 18:17:00 -0000
Date: Fri, 12 Oct 2001 11:17:03 -0700 (PDT)
From: Dirk-Willem van Gulik <dirkx@covalent.net>
X-X-Sender: dirkx@titatovenaar.sfo.covalent.net
To: dev@httpd.apache.org
Subject: Portign 1.3 modules - should I still use handlers ?
Message-ID: 
 <Pine.OSX.4.40.0110121109000.23488-100000@titatovenaar.sfo.covalent.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N


While porting a 1.3 module to 2.0 - which is totally handler based; I
used to have:

	static const handler_rec s99_handlers[] = {
	    {"s99hk", s99_hk_handler},
	    {"s99get", s99_get_handler},
	    {"s99cadget", s99_cadget_handler},
	    .....<lots and lots more of those>.....
	    {NULL, NULL}
	};

and now in 2.0 I am doing:

	static void register_hooks(apr_pool_t *p)
	{
	    ap_hook_handler(s99_hk_handler, NULL, NULL, APR_HOOK_MIDDLE);
	    ap_hook_handler(s99_get_handler, NULL, NULL, APR_HOOK_MIDDLE);
	    ....

	static int s99_hk_handler(...)
	{
	    struct s99_module_data * m = ap_get_module..
	    .....
	    if (strcmp(r->handler, "s99hk"))
        	return DECLINED;

	static int s99_get_handler(...)
	{
	    struct s99_module_data * m = ap_get_module..
	    .....
	    if (strcmp(r->handler, "s99get"))
        	return DECLINED;

Now is this the right thing to do ? i.e. have apache go into each of the
handlers and ask it wether it wants to run or decline.

Or is there a way I can keep this deceision process closer to the core in
some tight inner loop ?

Or to ask the question in other words - are handlers still the right thing
to use for this sort of 'fan out' handling or is there another technique
more optimal in 2.0 ?

Dw