Gould, James <jgo...@verisign.com> 2017-03-31 17:25 > Thanks for the response. My feedback is embedded below. I’ve updated the > proposal based on your feedback: > > > > 1. The <contact:chg> sub-elements do have replace semantics > > a. Existing sub-element data deleted first and then set with > updated data. > > b. This includes the <contact:postalInfo>, <contact:voice>, > <contact:fax>, <contact:email>, <contact:authInfo>, and the > <contact:disclose> elements. > > c. Servers with special policies regarding contact data > modifications should check the new data for any actual changes in relevant > fields. > > 2. The typed <contact:postalInfo> elements (“int” or “loc”) are > treated independently > > a. Exclusion of a <contact:postalInfo> type (“int” or “loc”) does > not implicitly delete it > > 3. The typed <contact:postalInfo> element (“int” or “loc”) is > deleted explicitly via an empty element > > a. <contact:postalInfo type=”int”/> or <contact:postalInfo > type=”loc”/> > > b. The same applies to deleting the voice or the fax via an empty > element (<contact:voice/> or <contact:fax/>) > > > > Let me know if any other changes are needed. Hello James,
I agree with your mini specification. I would suggest to add some wording about the contact:street nodes, especially if their number changes with the update done. It would probably be superfluous but I believe in this case it is better to over specify things instead of under-specify since this contact:update issue is quite old, so it is a great idea and work to specify it clearly and completely now once for all. -- Patrick Mevzek _______________________________________________ regext mailing list regext@ietf.org https://www.ietf.org/mailman/listinfo/regext
