[h323plus] Some issues with h323pdu.cxx

Simon Horne s.horne at spranto.com
Fri Feb 14 07:18:17 EST 2014


Denis

 

Sorry there was a typo. I just checked in the fix. 

 

Simon

 

From: h323plus-bounces at lists.packetizer.com
[mailto:h323plus-bounces at lists.packetizer.com] On Behalf Of Denis Kochmashev
Sent: 14 February 2014 21:22
To: h323plus at lists.packetizer.com
Subject: Re: [h323plus] Some issues with h323pdu.cxx

 

Simon, excuse me, but you did it wrong way.

 

Variable "tag" must be set to the value of aliasAddressTypes[i].tag BEFORE
comparing it with H225_AliasAddress::e_h323_ID.

1.       tag = aliasAddressTypes[i].tag;

2.       if (tag == H225_AliasAddress::e_h323_ID .

That was in my patch.

 

In your variant variable "tag", when compared with
H225_AliasAddress::e_h323_ID, has the value supplied as function argument.

 

From: Simon Horne [mailto:s.horne at spranto.com] 
Sent: Friday, February 14, 2014 4:53 PM
To: 'Denis Kochmashev'; h323plus at lists.packetizer.com
Subject: RE: [h323plus] Some issues with h323pdu.cxx

 

Denis

 

Agreed and checked in.

 

Simon

 

From: h323plus-bounces at lists.packetizer.com
[mailto:h323plus-bounces at lists.packetizer.com] On Behalf Of Denis Kochmashev
Sent: 14 February 2014 20:41
To: 'Simon Horne'; h323plus at lists.packetizer.com
Subject: Re: [h323plus] Some issues with h323pdu.cxx

 

Right!

Do you agree with it?

 

From: Simon Horne [mailto:s.horne at spranto.com] 
Sent: Friday, February 14, 2014 4:33 PM
To: 'Denis Kochmashev'; h323plus at lists.packetizer.com
Subject: RE: [h323plus] Some issues with h323pdu.cxx

 

Denis

 

So the logic is 

Dialed                                      Alias Type

123456                                      dialedDigits

user                                           h323-id

h323:123456                           h323-id

h323:user                                h323-id

h323:user at domain.com  url-id

 

Is that the intended behavior?

 

Simon

 

From: h323plus-bounces at lists.packetizer.com
[mailto:h323plus-bounces at lists.packetizer.com] On Behalf Of Denis Kochmashev
Sent: 14 February 2014 17:16
To: h323plus at lists.packetizer.com
Subject: Re: [h323plus] Some issues with h323pdu.cxx

 

Simon,

 

I was just pointing to the problem, which I thought must be solved somehow.
And I was not sure that my solution is the right one. Now I'm sure that it
is wrong. I apologize for my mistake.

 

After reading chapters

-          "7.1.4 H.323 URL scheme";

-          "O.5 Encoding of H.323 URL in H.323 messages";

-          "O.6 Non-H.323 URLs and URIs within the context of H.323"

from "H.323 (12/2009)" I propose another one.

 

From: Simon Horne [mailto:s.horne at spranto.com] 
Sent: Thursday, February 13, 2014 10:56 PM
To: 'Denis Kochmashev'; h323plus at lists.packetizer.com
Subject: RE: [h323plus] Some issues with h323pdu.cxx

 

Denis

 

Patch applied. Thx

Note I have been reliably informed the h323: prefix means URL alias type but
I agree with you.

 

Simon

 

From: h323plus-bounces at lists.packetizer.com
[mailto:h323plus-bounces at lists.packetizer.com] On Behalf Of Denis Kochmashev
Sent: 13 February 2014 18:41
To: h323plus at lists.packetizer.com
Subject: [h323plus] Some issues with h323pdu.cxx

 

Hello!

 

Please correct me if I'm wrong, but

1.       It looks like there is a small mistake in h323pdu.cxx which doesn't
allow to build h323plus without H.235 being enabled.

2.       I've found that 'H323SetAliasAddress(const PString & _name,
H225_AliasAddress & alias, int tag)' doesn't allow to set AliasAddress of
type h323_ID if it looks like dialedDigits even if _name begins with
'h323:'. For example, if I specify _name = 'h323:12345', alias tag will all
the same be dialedDigits because it looks like them. I suggest a small
change to H323SetAliasAddress, because now it produces unpredictable
results.

 

Thanks,

Denis

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.packetizer.com/pipermail/h323plus/attachments/20140214/bf1f37c6/attachment-0002.html>


More information about the h323plus mailing list