Re: [netconf] [Technical Errata Reported] RFC8040 (6271)

[Kent Watsen <[e-mail]>]

I don’t think that I’m able to edit this errata for Section 4.5.  If I could, I would submit again with the new text that fixes the *entries* language.

I think that the AD needs to “reject” this Errata (EID 6271).  We could file separate errata to fix the “entries” language in Sections 4.5, 4.8.5, and 4.8.6, but I’m unsure it’s worth it.

So I just filed a new errata https://www.rfc-editor.org/errata/eid6277 <https://www.rfc-editor.org/errata/eid6277> to fix just the “insert” query parameter "default" issue.

K.


> On Sep 3, 2020, at 1:37 AM, Martin Björklund <[e-mail]> wrote:
> 
> Kent Watsen <[e-mail]> wrote:
>> 
>> 
>>> On Sep 2, 2020, at 12:24 PM, Kent Watsen <[e-mail]> wrote:
>>> 
>>>    NEW:
>>> 
>>>       The default value is “last”, with exception to when used with PUT
>>>       and the target resource already exists, in which case the default
>>>       is to replace the target resource without altering its position in the
>>>       "ordered-by user” list or leaf-list.
>> 
>> The text above allows/suggests that are four “insert” values MAY
>> always be explicitly passed (and presumably processed).  Do we want to
>> support the case where a PUT is used only to move an element?
>> i.e. the passed “replacement” data is identical to what exists on the
>> server, only the insert/point have an effect?
>> 
>> Or should this strategy be that the insert/point can always be passed
>> (in a PUT for an "ordered-by user” list or leaf-list), but they only
>> have meaning (i.e., are processed) when/if the object is being
>> created, and otherwise ignored?
> 
> No, move (and possibly modify the entry) is supported.  Section 4.8.6
> says:
> 
>   The "point" query parameter is used to specify the insertion point
>   for a data resource that is being created or moved within an
>                                             ^^^^^^^^
>   "ordered-by user" list or leaf-list.
> 
> 
> /martin