Hi bogdan, If it's my cpl outgoing script configuration, from USER A ****************************************************** <cpl> <outgoing> <address-switch field="destination" subfield="user"> <address is="B"> <proxy /> </address> </address-switch> </outgoing> <cpl> ******************************************************* When I make a call from A to B, then the script has called, but in debug I can see that CPL rewrites R-URI with source address (A@A_ip_address) instead of destination address (B@B_ip_address). So the message goes back to user A. regards, Daniel Bogdan-Andrei Iancu wrote: > Hi Daniel, > > there are no differences - the implementation follows the RFC. For > incoming calls, the location set is empty, but it can be populated by > the CPL script during execution. So, when you do "proxy", you may > have multiple entries in the location set. > > The comment in the cpl_sig.c file refers to some implementation > details - how to process the available location set, depeding if it > is the first time doing proxy (for the call) or a second time. > > regards, > bogdan > > daniel grotti wrote: >> Hi all, >> this is what CPL_SIG.c says about forwarding a given message to the >> given location set: >> >> " forwards the msg to the given location set; if flags has set the >> CPL_PROXY_DONE, all locations will be added as branches, >> otherwise, the first one will set as RURI (this is ha case when >> this is the first proxy of the message) >> The given list of location will be freed, returning 0 instead. >> Returns: 0 - OK " >> >> >> My question is : which and what is "the first one" location ? Can >> you explain it better? >> Because RFC 3880 say that : "For the incoming top-level call >> processing action, the location set is initialized to the empty set. >> For the outgoing action, it is initialized to the destination >> address of the call." >> >> So RFC is very clear. I think there are considerable differences >> between RFC and the behavior of CPL interpreter. >> >> Thanks. >> Daniel >>