[OpenSER-Users] Does somebody use $(rU{s.unescape.user}) to be RFC3261 compliant?

Johansson Olle E oej at edvina.net
Thu May 22 11:53:08 CEST 2008


22 maj 2008 kl. 11.48 skrev Iñaki Baz Castillo:

> El Thursday 22 May 2008 11:22:59 Johansson Olle E escribió:
>> 22 maj 2008 kl. 11.15 skrev Iñaki Baz Castillo:
>>> Hi, to be RFC3261 compliant a SIP proxy should accept hex encoded
>>> username in
>>> any URI, this is:
>>>
>>> sip:alice at domain == sip:%61l%69ce at domain
>>>
>>> For allowing this we must use, explicitely, the transformation
>>> $(rU{s.unescape.user}). Do OpenSer administrators allow this hex
>>> encoding?
>>
>> Equally interesting: Is OpenSER translating this before it hits the
>> configuration,
>> like all the database tables and comparisions - or do we have to do  
>> it
>> in the configuration?
>>
>> I would prefer "automatic" handling in comparisions with database
>> tables and such.
>> For one account, like "38#45" some clients URI-encode, some does not.
>
> Hi Olle. I've reported today a case in which OpenSer doesn't handle  
> those
> comparisions (it doesn't unescape hex values):
>
> https://sourceforge.net/tracker/index.php?func=detail&aid=1969445&group_id=139143&atid=743020

I've been having a lot of issues with this in Asterisk. If you turn  
pedantic=on in Asterisk (please don't
ask me about that stupid configuration) we will unescape everything  
before handling the message.

Now, a proxy is a different thing. When proxying a message, it needs  
to be the same coming out as
coming in. The issue is really what to do when comparing stuff, both  
in the script and internally,
like the check_xxx functions.

/O



More information about the sr-users mailing list