[Openh323gk-developer] =?gb2312?B?UmU6IFJlOiBbT3BlbmgzMjNnay1kZXZlbG9wZXJdIE51bWJlcmljIEgzMjMgQWxpYXMgcHJv YmxlbQ==?=

[<yt...@16...>]

Dear Chih-Wei,
Maybe you misunderstood me, I mean the caller endpoint has no idea about the user dialed digits is e164 or h323-id, but GK knows the alias type of registered endpoints.

ITU-T H.323 *does* allow endpoint to set h323-ID to all number string, and *does* allow to set e164 and h323-ID to the same, here is the definition from H.225

AliasAddress ::= CHOICE
{
    e164 IA5String (SIZE (1..128)) (FROM ("0123456789#*,")),
    h323-ID BMPString (SIZE (1..256)), -- Basic ISO/IEC 10646-1 (Unicode)...,
    url-ID IA5String (SIZE(1..512)), -- URL style addresstransportID TransportAddress,
    email-ID IA5String (SIZE(1..512)), -- rfc822-compliant email address
    partyNumber PartyNumber
}
There will be no ambiguous if you have different type of aliases but the same strings, even with gnugk, you can try set e164 and h323-id to the same string, will get RCF from gnugk.

I've tested the "456" like h323-ID with other gatekeepers such as opengk from openh323.org and all commercial GKs I have, they all work well.

The point is you can't prevent the endpoint from using all number h323-ID, gnugk must handle this situation just like other GK does.

> Yu Tan wrote:
> > But actually the endpoint has no idea a dialed-digits is a E164 number or is
> > a H323_ID, only GK knows that.
> 
> Of course not.
> The GK has *no idea* what the alias type is.
> The type of alias is specified by the endpoint.
> The GK just match the alias with the correct type
> in the RegistrationTable.
> If the GK try to match aliases by ignoring the types,
> there will be ambiguous if you have different type of aliases
> but the same strings.
> (Note the GK have to assure the uniqueness of an alias)
> If the GK automatically match an alias with different type,
> why ITU-T defined such types?
> 
> > what's more, most endpoints can't manually specify the alias type in
> > H323Alias or E164 number except Openh323.I can't find this kind of setting
> > in Polycom, Tandberg, VCON, Picturetel, Netmeeting or any other terminals.
> 
> That's the problem of the endpoint.
> 
> In the real world, most of the endpoints(also openh323 client)
> automatically set the type of aliases without asking the user to specify them.
> If an alias is all numbers (e.g., 456), it is set as a dialedDigits(e164),
> or it is an H323ID.
> If you'd like to set 456 as an H323ID, you have to ask
> the vendor of the endpoint to provide this feature.
> 
> In summary, why did you register an alias 456 as an H323ID?
> You are confusing yourself.
> The rule is, always register a number alias as a dialedDigits(e164).
> 
> -- 
>    ~     Chih-Wei Huang (cw...@li...)
>   'v'    CLDP Project : http://www.linux.org.tw/CLDP/ (Coordinator)
>  // \\   CLE  Project : http://cle.linux.org.tw/CLE/  (Developer)
> /(   )\  I18N Project : http://i18n.linux.org.tw/     (Translator)
>  ^`~'^   HomePage     : http://www.cwhuang.idv.tw/
> 
> 
> ----------------------------------------------------------------------------
>                    Bringing you mounds of caffeinated joy
>                    >>>     http://thinkgeek.com/sf    <<<
> 
> _______________________________________________
> Openh323gk-developer mailing list
> Ope...@li...
> https://lists.sourceforge.net/lists/listinfo/openh323gk-developer
>