Re: [NGO] external module properties

Hello,
mandatory versioning was the most frequent request I got from potential users of YANG, so I 
agree with Andy it is needed.
Balazs

Andy Bierman wrote:
> Hi,
> 
> There is so much focus on the contents of a NETCONF data module,
> that little attention has been paid to the 'external' properties.
> I have already brought up module/submodule name uniqueness.
> 
> YANG does not have a concept of a module version, which is a mistake.
> SMIv2 has LAST-UPDATED.  IMO, all YANG modules should have at
> least 3 standard external properties:
> 
>    - name (matches import or include directive)
>    - version (derived from most recent 'revision')
>    - authoritative online module repository location
> 
> This implies that at least one revision-stmt MUST be present.
> IMO, especially thinking for the long-term, it is not acceptable
> to use ad-hoc non-standard mechanisms to identify the actual
> module version.  LAST-UPDATED and REVISION are optional in SMIv2.
> This is a mistake that should be fixed in YANG.
> 
> Another controversial external module property is 'schemaLocation' ;-)
> 
_______________________________________________
NGO mailing list
NGO@ietf.org
https://www.ietf.org/mailman/listinfo/ngo