Mario,
Thanks for reviewing this. My replies are in-line.
------ Original Message ------
From: "Mario Loffredo" <mario.loffr...@iit.cnr.it>
To: "Andrew Newton" <a...@hxr.us>
Cc: "Registration Protocols Extensions" <regext@ietf.org>
Sent: 10/9/2017 11:41:21 AM
Subject: Re: [regext] Preparing for IETF 100
Hi Andy,
it seems to me that some corrections should be made:
- the last three properties of "secureDNS" definition should be
moved inside the "domain_mixin" definition;
Good catch! Thanks.
- the "autnum_mixin" definition should contain the "country"
property;
Another good catch. Thanks.
- the "event" definition should be splitted in two like the
following:
$noActorEvent = {
"eventAction" : string,
"eventDate" : datetime,
$links,
$lang
}
$event = {
$noActorEvent,
"eventActor" : string,
}
The noActorEvent type should be used in the definition of
"asEventActor" property
because RFC7483 states that each object in the array must not have
an "eventActor" member.
Yes, I agree. I'll also add some comments to explain it. Thanks.
In addition, there is a case of redundancy but I don't know if it
is an error or it is allowed by JCR overriding rules.
The presence of the "lang" property in a lookup response can be
derived by following two paths. For example, in a domain lookup
response lang can be derived both by
domain_response->response_mixin->lang
and domain_response->domain_mixin->common_mixin->lang
Since lang is optional, there is no harm. However, I see no reason why
it should be in both places. Thanks for point this out.
I'll make these corrections on the next version of the draft.
-andy
_______________________________________________
regext mailing list
regext@ietf.org
https://www.ietf.org/mailman/listinfo/regext
