Re: [netmod] Fwd: New Version Notification for draft-lengyel-netmod-yang-instance-data-01.txt

[Robert Wilton <[e-mail]>]

On 26/06/2018 16:20, Balazs Lengyel wrote:
>
>
>
> On 6/26/2018 4:07 PM, Robert Wilton wrote:
>>>>>>        7) It might want to include a semantic version number for an
>>>>>>        instance-data-set, depending on whether the YANG 
>>>>>> versioning discussions
>>>>>>        ends up.
>>>>>>
>>>>>>      BALAZS: Yes I would like to. However I am not exactly clear 
>>>>>> on what does
>>>>>>      backwards compatibility mean for instance data.
>>>>>>      Data MAY NOT be removed or changed only added.  ???
>>>>>>      Who would use the semver numbers ???
>>>>> What does the version number mean? Every change of instance data 
>>>>> in an
>>>>> instacne-data-set leads to a new version number? What is a bug fix in
>>>>> this sense? What is a non-backwards compatible change of instance 
>>>>> data?
>>>>> I am left a bit puzzled.
>>>> Probably I don't mean semantic version.  But often files have 
>>>> versioning, or
>>>> revision, information associated with them.  This can be muxed into 
>>>> the file
>>>> name/path, but it also seems like potentially useful metadata, and 
>>>> being
>>>> able to handle this generically in a consistent way might be 
>>>> beneficial.
>>>>
>>>> As an example, perhaps the capability information related to S/W 
>>>> release
>>>> 1.2.4, etc.
>>> We need to separate a version number of the instance data from the a
>>> version number (or version context) that may be needed in order to
>>> understand what a YANG (module, path) tuple means if we move to a
>>> different YANG versioning scheme. The later I agree would be needed,
>>> the former I am less sure of - at least if we talk semantic version
>>> numbers.
>> Yes, I agree with the need for the latter, that may entail listing 
>> the versions of the modules in the meta-data.
>>
>> But what I was actually referring to was the former, and thinking 
>> more of it is just being a string field in the meta-data with a well 
>> defined name, that is optional to populate.  Actually, I see the 
>> current definition already has a revision with a date and description 
>> string.  This is roughly along the lines of what I was thinking of 
>> for versioning (although I'm not sure why this should be a list 
>> rather than just a pair of leaves).
>>
>> Thanks,
>> Rob
> BALAZS:  IMHO if we want to use versioning (semver) to version the 
> instance-data-set itself we first need to understand what backwards 
> compatibility means for an instance data set.
So, I'm suggesting versioning, but not semver.  Just a string version 
field.  E.g. for XR I would just put in the XR release number "R4.5.1" 
or similar.

>
> Revision is a list because as you develop your instance-data-set you 
> may go through multiple revisions. Just as you may have multiple 
> revision statements in a YANG Module (YAM) the same way you may need 
> multiple revision list entries in an instance data set; to keep history.
But I don't see that every instance data document wants to necessarily 
publish that history, although of course it could just publish a list 
containing a single (the latest) entry.

Thanks,
Rob


>
> (And following your advice, that is one very good reason to define the 
> metadata about the set as normal YANG instead of yang-meta-data which 
> does not have the concept of meta-data-array)
> regards Balazs
> -- 
> Balazs Lengyel                       Ericsson Hungary Ltd.
> Senior Specialist
> Mobile: [phone]              email:[e-mail]  
> .