Re: [Netconf] mbj's WGLC review of yang-push-17

[Qin Wu <[e-mail]>]

-----邮件原件-----
发件人: Juergen Schoenwaelder [mailto:[e-mail]] 
发送时间: 2018年9月26日 14:32
收件人: Qin Wu
抄送: Kent Watsen; Netconf
主题: Re: [Netconf] mbj's WGLC review of yang-push-17

On Wed, Sep 26, 2018 at 01:25:54AM +0000, Qin Wu wrote:
> 
> To address this, one solution we proposed in draft-ietf-lime-yang-connectionless-oam-18:
> Is to define ietf-lime-time-types which can be reused in other 
> modules. In ietf-lime-time-types, we define define time-unit-type as identity.
> But maybe typedef is even better since both time-unit and value are defined in one.
>

The good old practice was to think about the number range that makes sense and then to use a numeric type with a suitable unit that covers the number range.

I wonder whether you really need a uint32 range of hours in the OAM models. 
	
[Qin]: not necessary for OAM model, but It may be used in some other models that require unint32 range of hours.

Note that a single uint64 of nanoseconds would give you up to
584 years. If oam models do indeed need delay values that range from nanoseconds to hundreds of years, I would rather go with a single
uint64 instead of a combination of an uint32 and a unit identifier.

[Qin]: Not advocating using such combination, just point out such common type are also needed in some other models, that's why a separate ietf-lime-time-types was defined.

/js

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: [phone]         Campus Ring 1 | 28759 Bremen | Germany
Fax:   [phone]         <https://www.jacobs-university.de/>