Module: sip-router
Branch: master
Commit: 16b47a9a8830fc5c9d2b310536a7a881e233b9ac
URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=16b47a9…
Author: Daniel-Constantin Mierla <miconda(a)gmail.com>
Committer: Daniel-Constantin Mierla <miconda(a)gmail.com>
Date: Mon Apr 27 13:00:44 2009 +0200
define ANY_ROUTE
- defined as 0xffffffff to match any route type
---
route.h | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/route.h b/route.h
index 221fb4c..cb86ba4 100644
--- a/route.h
+++ b/route.h
@@ -52,6 +52,7 @@
#define ONSEND_ROUTE (1 << 4)
#define ERROR_ROUTE (1 << 5)
#define LOCAL_ROUTE (1 << 6)
+#define ANY_ROUTE (0xFFFFFFFF)
/* The value of this variable is one of the route types defined above and it
* determines the type of the route being executed, module functions can use
Module: sip-router
Branch: master
Commit: 89bfe93c18d4ec77ff8c1819d365bda2a85613fe
URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=89bfe93…
Author: Andrei Pelinescu-Onciul <andrei(a)iptel.org>
Committer: Andrei Pelinescu-Onciul <andrei(a)iptel.org>
Date: Fri Apr 24 10:52:46 2009 +0200
tm: hooks comments update
- updated description for TMCB_RESPONSE_OUT and
TMCB_LOCAL_RESPONSE_OUT.
---
modules/tm/t_hooks.h | 34 ++++++++++++++++++++++++----------
1 files changed, 24 insertions(+), 10 deletions(-)
diff --git a/modules/tm/t_hooks.h b/modules/tm/t_hooks.h
index b396737..de20c51 100644
--- a/modules/tm/t_hooks.h
+++ b/modules/tm/t_hooks.h
@@ -167,16 +167,27 @@ struct cell;
* distinguish between the two, one would need to look at the method in
* Cseq (look at t_reply.c:1630 (reply_received()) for an example).
*
- * TMCB_RESPONSE_OUT -- a final reply was sent out (either local
- * or proxied) -- there is nothing more you can change from
- * the callback, it is good for accounting-like uses. No lock is held.
- * Known BUGS: it's called also for provisional replies for relayed replies.
- * For local replies it's called only for the final reply.
- *
- * Note: the message passed to callback may also have
+ * TMCB_RESPONSE_OUT -- a final or provisional reply was sent out
+ * successfully (either a local reply or a proxied one).
+ * For final replies is called only for the first one (it's not called
+ * for retransmissions).
+ * For non-local replies (proxied) is called also for provisional responses
+ * (NOTE: this might change and in the future it might be called only
+ * for final replies --andrei).
+ * For local replies is called _only_ for the final reply.
+ * There is nothing more you can change from the callback, it is good for
+ * accounting-like uses. No lock is held.
+ * Known oddities: it's called for provisional replies for relayed replies,
+ * but not for local responses (see NOTE above).
+ * Note: if the send fails or via cannot be resolved, this callback is
+ * _not_ called.
+ * Note: local reply means locally generated reply (via t_reply() & friends)
+ * and not local transaction.
+ *
+ * Note: the message passed to the callback may also have
* value FAKED_REPLY (like other reply callbacks) which
- * indicates a pseudo_reply caused by a timer. Check for
- * this value before deferring -- you will cause a segfault
+ * indicates a local reply caused by a timer, calling t_reply() a.s.o.
+* Check for this value before deferring -- you will cause a segfault
* otherwise. Check for t->uas.request validity too if you
* need it ... locally initiated UAC transactions set it to 0.
*
@@ -249,9 +260,12 @@ struct cell;
* transaction arrived. Message may be FAKED_REPLY. Can be called multiple
* times, no lock is held.
*
- * TMCB_LOCAL_REPONSE_OUT -- provisional reply for localy initiated
+ * TMCB_LOCAL_RESPONSE_OUT -- provisional reply for localy initiated
* transaction. The message may be a FAKED_REPLY and the callback might be
* called multiple time quasi-simultaneously. No lock is held.
+ * Note: depends on tm.pass_provisional_replies.
+ * Note: the name is very unfortunate and it will probably be changed
+ * (e.g. TMCB_LOCAL_PROVISIONAL).
*
* TMCB_NEG_ACK_IN -- an ACK to a negative reply was received, thus ending
* the transaction (this happens only when the final reply sent by tm is