git:master: modules:carrierroute Fix documentation when using cr_route/ t_relay in failure routes
Module: sip-router Branch: master Commit: c44529a8078f7c58d621e5116757da10084180f1 URL: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commit;h=c44529a8...
Author: Marius Zbihlei marius.zbihlei@1and1.ro Committer: Marius Zbihlei marius.zbihlei@1and1.ro Date: Thu Sep 30 17:04:00 2010 +0300
modules:carrierroute Fix documentation when using cr_route/t_relay in failure routes
Starting with 3.0 version, there is no need to call append_branch in failure routes before t_relay if the RURI is new(it is done automatically from t_relay).
---
modules/carrierroute/README | 69 +++++++++++------------ modules/carrierroute/doc/carrierroute_admin.xml | 6 +- 2 files changed, 36 insertions(+), 39 deletions(-)
Diff: http://git.sip-router.org/cgi-bin/gitweb.cgi/sip-router/?a=commitdiff;h=c445...
2010/9/30 Marius Zbihlei marius.zbihlei@1and1.ro:
Starting with 3.0 version, there is no need to call append_branch in failure routes before t_relay if the RURI is new(it is done automatically from t_relay).
Hi, and what does happen if append_branch() is called anyway?
Thanks.
On 10/01/2010 11:10 AM, Iñaki Baz Castillo wrote:
2010/9/30 Marius Zbihleimarius.zbihlei@1and1.ro:
Starting with 3.0 version, there is no need to call append_branch in failure routes before t_relay if the RURI is new(it is done automatically from t_relay).
Hi, and what does happen if append_branch() is called anyway?
Thanks.
Hello Iñaki,
Check this thread for the reasoning http://lists.sip-router.org/pipermail/sr-dev/2010-September/009144.html
It depends on patched version, but I think with 3.0, 2 identical branches will be created.
Marius
2010/10/1 marius zbihlei marius.zbihlei@1and1.ro:
Hi, and what does happen if append_branch() is called anyway?
Hello Iñaki,
Check this thread for the reasoning http://lists.sip-router.org/pipermail/sr-dev/2010-September/009144.html
It depends on patched version, but I think with 3.0, 2 identical branches will be created.
Thanks a lot.
On 10/1/10 11:25 AM, Iñaki Baz Castillo wrote:
2010/10/1 marius zbihleimarius.zbihlei@1and1.ro:
Hi, and what does happen if append_branch() is called anyway?
Hello Iñaki,
Check this thread for the reasoning http://lists.sip-router.org/pipermail/sr-dev/2010-September/009144.html
It depends on patched version, but I think with 3.0, 2 identical branches will be created.
Thanks a lot.
in 3.0 the auto-detection of new uri is not available, only in devel (3.1).
Cheers, Daniel
On Oct 01, 2010 at 11:28, marius zbihlei marius.zbihlei@1and1.ro wrote:
On 10/01/2010 11:10 AM, Iñaki Baz Castillo wrote:
2010/9/30 Marius Zbihleimarius.zbihlei@1and1.ro:
Starting with 3.0 version, there is no need to call append_branch in failure routes before t_relay if the RURI is new(it is done automatically from t_relay).
Hi, and what does happen if append_branch() is called anyway?
Thanks.
Hello Iñaki,
Check this thread for the reasoning http://lists.sip-router.org/pipermail/sr-dev/2010-September/009144.html
It depends on patched version, but I think with 3.0, 2 identical branches will be created.
Actually with 3.0 you still need append_branch() after changing RURI. With latest 3.1 nothing will happen (if you call it after changing RURI it won't add an extra branch).
Andrei
On Oct 01, 2010 at 11:30, Andrei Pelinescu-Onciul andrei@iptel.org wrote:
On Oct 01, 2010 at 11:28, marius zbihlei marius.zbihlei@1and1.ro wrote:
On 10/01/2010 11:10 AM, Iñaki Baz Castillo wrote:
2010/9/30 Marius Zbihleimarius.zbihlei@1and1.ro:
Starting with 3.0 version, there is no need to call append_branch in failure routes before t_relay if the RURI is new(it is done automatically from t_relay).
Hi, and what does happen if append_branch() is called anyway?
Thanks.
Hello Iñaki,
Check this thread for the reasoning http://lists.sip-router.org/pipermail/sr-dev/2010-September/009144.html
It depends on patched version, but I think with 3.0, 2 identical branches will be created.
Actually with 3.0 you still need append_branch() after changing RURI. With latest 3.1 nothing will happen (if you call it after changing RURI it won't add an extra branch).
One more clarification: the r-ruri change autodetection (3.1) works in all routes, not only in failure routes. E.g., if I have in the main route:
t_relay(); setuser("foo1"); t_relay(); setuser("foo2); t_relay();
3 branches will be created: 1 for the original uri, one for foo1@ and the last one for foo2@.
Andrei
2010/10/1 Andrei Pelinescu-Onciul andrei@iptel.org:
One more clarification: the r-ruri change autodetection (3.1) works in all routes, not only in failure routes. E.g., if I have in the main route:
t_relay(); setuser("foo1"); t_relay(); setuser("foo2); t_relay();
3 branches will be created: 1 for the original uri, one for foo1@ and the last one for foo2@.
Then the meaning of append_branch is confusing, it seems that with this new changes it's not needed at all so, what about removing it?
On Oct 01, 2010 at 11:49, Iñaki Baz Castillo ibc@aliax.net wrote:
2010/10/1 Andrei Pelinescu-Onciul andrei@iptel.org:
One more clarification: the r-ruri change autodetection (3.1) works in all routes, not only in failure routes. E.g., if I have in the main route:
t_relay(); setuser("foo1"); t_relay(); setuser("foo2); t_relay();
3 branches will be created: 1 for the original uri, one for foo1@ and the last one for foo2@.
Then the meaning of append_branch is confusing, it seems that with this new changes it's not needed at all so, what about removing it?
You can still add a branch without changing the uri: append_branch("sip:foo@bar").
You can use it to create a duplicate branch (though I admit I don't know what would this be usefull for), e.g.: setuser("foo"); append_branch(); # this does nothing append_branch(); # this will add a duplicate branch (same uri)
A more valid use, is to add branches first, and calling t_relay() only once at the end (classical parallel forking), e.g.: append_branch(); setuser("foo1"); append_branch(); setuser("foo2"); append_branch(); t_relay();
is equivalent with the first example.
Andrei
2010/10/1 Andrei Pelinescu-Onciul andrei@iptel.org:
A more valid use, is to add branches first, and calling t_relay() only once at the end (classical parallel forking), e.g.: append_branch(); setuser("foo1"); append_branch(); setuser("foo2"); append_branch(); t_relay();
I missed this obvius feature for which, of course, the classic append_branch() is the most expected solution :)
On 10/01/2010 12:45 PM, Andrei Pelinescu-Onciul wrote:
On Oct 01, 2010 at 11:30, Andrei Pelinescu-Onciulandrei@iptel.org wrote:
On Oct 01, 2010 at 11:28, marius zbihleimarius.zbihlei@1and1.ro wrote:
On 10/01/2010 11:10 AM, Iñaki Baz Castillo wrote:
2010/9/30 Marius Zbihleimarius.zbihlei@1and1.ro:
Starting with 3.0 version, there is no need to call append_branch in failure routes before t_relay if the RURI is new(it is done automatically from t_relay).
Hi, and what does happen if append_branch() is called anyway?
Thanks.
Hello Iñaki,
Check this thread for the reasoning http://lists.sip-router.org/pipermail/sr-dev/2010-September/009144.html
It depends on patched version, but I think with 3.0, 2 identical branches will be created.
Actually with 3.0 you still need append_branch() after changing RURI. With latest 3.1 nothing will happen (if you call it after changing RURI it won't add an extra branch).
One more clarification: the r-ruri change autodetection (3.1) works in all routes, not only in failure routes. E.g., if I have in the main route:
t_relay(); setuser("foo1"); t_relay(); setuser("foo2); t_relay();
3 branches will be created: 1 for the original uri, one for foo1@ and the last one for foo2@.
Andrei
Hello
Ah, so 3.0 is not affected, only master and upcoming 3.1. Thanks for clearing things up, because I didn't had the time to test with 3.0 and I only relied on the exchanged emails.
Marius
-
Andrei Pelinescu-Onciul -
Daniel-Constantin Mierla -
Iñaki Baz Castillo -
Marius Zbihlei