[patch] expose more of ssl_engine_ext.c

["MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1)" <ma...@hp.com>]

Hi,
	Here's a patch which exposes more of the ssl_engine_ext.c code. The
ap_hook_* functionality is still not ported to the 2.0 style. Pl. let me
know if you have any comments / suggestions..

Thanks
-Madhu & Julius


 <<patch_ext.txt>> 

Re: [patch] expose more of ssl_engine_ext.c

["William A. Rowe, Jr." <wr...@rowe-clan.net>]

----- Original Message ----- 
From: "MATHIHALLI,MADHUSUDAN (HP-Cupertino,ex1)" <ma...@hp.com>
To: <ne...@apache.org>
Sent: Monday, July 30, 2001 10:31 PM
Subject: [patch] expose more of ssl_engine_ext.c


> Hi,
> Here's a patch which exposes more of the ssl_engine_ext.c code. The
> ap_hook_* functionality is still not ported to the 2.0 style. Pl. let me
> know if you have any comments / suggestions..

Finally, a patch I really can't apply as-is ... that's a compliment, btw, your last
several patches were excellent!

Comments inline...


--- mod_ssl.h 2001/07/30 22:35:33 1.22
+++ mod_ssl.h 2001/07/31 00:07:19
@@ -703,9 +701,7 @@
 
 /*  Extensions  */
 void         ssl_ext_register(apr_pool_t *p);
-#if 0 /* XXX */
 void         ssl_ext_unregister(void);
-#endif


How is this?  I'm not clear where/when we will be 'unregistering' fns in apache 2.0.
I'd rather leave that alone until we need it.


--- ssl_engine_ext.c 2001/07/30 22:35:33 1.5
+++ ssl_engine_ext.c 2001/07/31 02:02:59
@@ -69,7 +69,6 @@
 
 #include "..\..\modules\loggers\mod_log_config.h"
 static void  ssl_ext_mlc_register(apr_pool_t *p);
-#if 0 /* XXX */
 static void  ssl_ext_mlc_unregister(void);
 static void  ssl_ext_mr_register(void);
 static void  ssl_ext_mr_unregister(void);
@@ -77,31 +76,25 @@
 static void  ssl_ext_mp_unregister(void);
 static void  ssl_ext_ms_register(void);
 static void  ssl_ext_ms_unregister(void);
-#endif /* XXX */


I'm definately against re-registering this group of empty functions into the
namespace.  below, we effectively call a bunch of open-ended 'feature extensions'
that won't (necessarily) be implemented in the same way.  I would much rather
see us toggle these individually, as a feature is implemented.  WE DON'T NEED
BLANK PLACEHOLDERS!  [toggles snipped...]


 
@@ -128,16 +121,16 @@
     return;
 }
 
-#if 0 /* XXX - We don't really need this (do we???) */
 static void ssl_ext_mlc_unregister(void)
 {
+#if 0 /* XXX */
     ap_hook_unregister("ap::mod_log_config::log_c",
                        ssl_ext_mlc_log_c);
     ap_hook_unregister("ap::mod_log_config::log_x",
                        ssl_ext_mlc_log_x);
+#endif /* XXX */
     return;
 }
-#endif /* XXX */


As I said earlier, empty cruft.  No need for this function until it accomplishes something.
-1 on this part of the patch.





@@ -279,7 +279,7 @@
  * SSL proxy initialization
  */
 #ifdef SSL_EXPERIMENTAL_PROXY
-static void ssl_ext_mp_init(server_rec *s, pool *p)
+static void ssl_ext_mp_init(server_rec *s, apr_pool_t *p)
 {
     SSLSrvConfigRec *sc;
     char *cpVHostID;


The experimental stuff needs to be offered up as a seperate patch, and most importantly,
needs to be reviewied in the context of the httpd-2.0 proxy.  I don't know where we are
going with this, and would prefer not to touch it till the rest of the port is complete.

Cleaning up fn and type changes is fine, please posit that as it's own patch.

And thank you both for your hard efforts, I've attributed them all to Madhu since I don't
have the other pertinant details, sorry.  Please provide any other attributions in the
body of your message, so they aren't missed.  I'll hold off a bit on updating CHANGES, and
roll these in together, since the _end_user_ doesn't need every last bit of gory detail, 
only effective changes in the way they use/rely on the code.

Bill