[core] Using Yang to describe LWM2M ... was Re: ? WG adoption of draft-veillette-core-yang-cbor-mapping-00

[Hannes Tschofenig <[e-mail]>]

Hi Juergen,

On 04/25/2016 12:33 PM, Juergen Schoenwaelder wrote:
> Hannes,
> 
> whether this YANG fragment makes sense or not I simply do not know
> without knowing LWM2M and since there are no open specs there is
> little help you can expect. I have no clue about the interaction model
> of OMA LWM2M nor do I know what LWM2M_Device is etc.

The specs are publically available for download at no cost.

Here is the link:
http://technical.openmobilealliance.org/Technical/technical-information/release-program/current-releases/oma-lightweightm2m-v1-0

You want to look at the 'Technical Specification'.

The registry for the objects are also public and available in a nicely
written page (similar to what IANA provides for the IETF):
http://technical.openmobilealliance.org/Technical/technical-information/omna/lightweight-m2m-lwm2m-object-registry

> 
> The claim was made that there is a 'simple' conversion (and I read
> that as 'simple algorithm' to do conversion) and all I am asking is
> whether someone actually went through the details of this. All I saw
> so far are sketches of examples, thats good for a start but still
> likely a longer way from a 'simple algorithm' (if that was the
> intention).

Nobody has gone through the details. This is one of my complaints as
well. There is this argument that we can do things easily with Yang
since it is such a powerful language but then nobody spends the time to
look at the already available solutions.

> 
> But we should really be using a different thread

Changed the name of the Thread.

> - or simply wait
> until someone managed to work out the details and got permission from
> OMA to write it down without violating their legal constraints they
> put on their registration wall.

There is no permission that needs to be obtained IMHO, if I understand
the OMA copyright policy correctly.

Ciao
Hannes

> 
> /js
> 
> On Mon, Apr 25, 2016 at 12:19:54PM +0200, Hannes Tschofenig wrote:
>> Hi Juergen, Hi Carsten,
>>
>> The data modelling language used by OMA LWM2M is an XML schema and the
>> OMA only uses it for two purposes:
>>
>>  * Adding new objects and resources to the registry allows automatic
>> validation (of the XML instance documents against the XML schema).
>>
>>  * The XML schema really only models the data and does not attempt to
>> describe the interaction.
>>
>> There has been no attempt to generate code from the XML instance
>> documents (as far as I know).
>>
>> I have tried to take a subset of the device object and to describe it in
>> Yang. To my knowledge nobody has tried to describe all currently
>> registered objects with Yang.
>>
>> In any case, here is what I have tried:
>>
>> ----
>>
>> module LWM2M_Device {
>> 	// header information
>> 	yang-version "1";
>> 	namespace "foobar1";			
>> 	prefix "LWM2M_Device";
>> 	
>> 	// linkage statements
>> 	import extensions {
>> 		prefix extensions;
>> 		[e-mail];
>> 	}
>> 	
>> 	// meta information
>> 	organization "OMA";
>> 	contact "foobar2";
>> 	description
>> 		"This object provides a range of device related information which can
>> be queried by the LWM2M Server, and a device reboot and factory reset
>> function.";
>> 	reference
>> "http://technical.openmobilealliance.org/tech/profiles/LWM2M_Device-v1_0.xml";
>> 	
>> 	// revision history
>> 	revision 2016-04-20 {
>> 		description "Initial revision.";
>> 		reference "";
>> 	}
>> 	
>> 	// module definitions
>> 	extensions:LWM2M_Object_ID "3";
>> 	
>> 	// typedef statements
>> 	
>> 	// data statements
>> 	
>> 	leaf Manufacturer {
>> 		extensions:LWM2M_Resource_ID "0";
>>         type string;
>> 		config true;
>> 		description "Human readable manufacturer name";
>> 	}
>> 	
>> 	leaf ModelNumber {
>> 		extensions:LWM2M_Resource_ID "1";
>> 		type String;
>> 		config true;
>> 		description "A model identifier
>>                 (manufacturer specified string).";
>> 	}
>> 	
>> 	leaf SerialNumber {
>> 		extensions:LWM2M_Resource_ID "2";
>> 		type String;
>> 		config true;
>> 		description "Serial Number.";
>> 	}
>> 	
>> 	leaf AvailablePowerSources {
>> 		extensions:LWM2M_Resource_ID "6";
>> 		type int32 {
>>                   range "0..7";
>>                 };		
>> 		config true;
>> 		description "
>>  0 – DC power
>>  1 – Internal Battery
>>  2 – External Battery
>>  4 – Power over Ethernet
>>  5 – USB
>>  6 – AC (Mains) power
>>  7 – Solar";
>> 	}
>> 	
>> 	leaf BatteryLevel {
>> 		extensions:LWM2M_Resource_ID "9";
>> 		type int32 {
>>             range "0..100";
>>         };
>> 		units percent;
>> 		config false;
>> 		description "Contains the current battery level as a percentage (with
>> a range from 0 to 100). This value is only valid when the value of
>> Available Power Sources Resource is 1.";
>> 	}
>> 	
>> 	leaf MemoryFree {
>> 		extensions:LWM2M_Resource_ID "10";
>> 		type int32;
>> 		units KB;
>> 		config false;
>> 		description "Estimated current available amount of storage space which
>> can store data and software in the LWM2M Device (expressed in kilobytes).";
>> 	}
>> 	
>> 	leaf CurrentTime {
>> 		extensions:LWM2M_Resource_ID "13";
>> 		type timestamp;
>> 		units KB;
>> 		config false;
>> 		description "Current UNIX time of the LWM2M Client. The LWM2M Client
>> should be responsible to increase this time value as every second
>> elapses. The LWM2M Server is able to write this Resource to make the
>> LWM2M Client synchronized with the LWM2M Server.";
>> 	}
>> 	
>> }
>>
>> ----
>>
>> The problem I have is to find out whether Yang is able to also describe
>> the interaction model of OMA LWM2M. I believe the current understanding
>> is that this is not possible without some extensions. Please correct me.
>>
>> Ciao
>> Hannes
>>
>>
>> On 04/24/2016 09:28 PM, Carsten Bormann wrote:
>>> Hi Juergen,
>>>
>>> trying to read your message -- are you just commenting on the side
>>> conversation here or is this also a comment on the adoption call?
>>>
>>> Grüße, Carsten
>>>
>>>
>>> Juergen Schoenwaelder wrote:
>>>> On Thu, Apr 21, 2016 at 05:34:50PM +0000, Michel Veillette wrote:
>>>>> First, LWM2M have its own modeling language encoded in xml.
>>>>> A file like "OMA-SUP-XML_LWM2M_Security-V1_0-20131210-C" is not fundamentally different than something than can be named security.yang.
>>>>> A simple xml transform can probably do the conversion between the two without any lost.
>>>>> LWM2M just have a simpler (subset) modeling language.
>>>>>
>>>>
>>>> These are pretty bold statements. Claiming something is simple and
>>>> knowing something is simple are sometimes different things. Have you
>>>> worked throught the details? Is there a decent public definition of
>>>> the 'simpler (subset) modeling language'? And with public I mean
>>>> public, not hidden behind all sorts of registration walls.
>>>>
>>>> /js
>>>>
>>>
>>> _______________________________________________
>>> core mailing list
>>> [e-mail]
>>> https://www.ietf.org/mailman/listinfo/core
>>>
>>
> 
> 
>