Re: [netmod] [Anima] mcr's YANG question raised during the ANIMA WG session

[Michael Richardson <[e-mail]>]

Hi,  I completed two examples, one using the augment method that we were
using, which you pointed out was broken.   Thank you for helping with this,
and I hope you have a moment to review my test files... probably there is a
*third* way that works better.

What we want is to be able to write extensions in different RFCs, and then be
able to compose them together easily.

And yes, I get errors:

MODULE D
module-D2.yang:14: warning: imported module "module-A" not used
module-D2.yang:37: error: the 'yang-data' extension must have exactly one child that is a container
module-D2.yang:39: error: there is already a child node to "module-D2-artifact" at module-D2.yang:37 with the name "module-A-things" defined at module-D2.yang:39 (at module-A.yang:30)
module-D2.yang:47: error: node module-D2::module-B2-things is not found
module-D2.yang:55: error: there is already a child node to "module-D2-grouping" at module-D2.yang:42 with the name "module-A-things" defined at module-D2.yang:45 (at module-A.yang:30)
module-D2.yang:56: error: node module-D2::module-C2-things is not found

My first example (D1) created new grouping, which I import separately, but do not augment.
But, the results has quite a bit of unwanted "structure":

MODULE D
module: module-D1

  grouping module-D1-grouping:
    +-- module-D1-things
       +-- module-A-things
       |  +-- attribute-A-Alpha?   binary
       +-- module-B-things
       |  +-- module-A-things
       |  |  +-- attribute-A-Alpha?   binary
       |  +-- attribute-B-Alpha?   binary
       +-- module-C1-things
       |  +-- module-A-things
       |  |  +-- attribute-A-Alpha?   binary
       |  +-- attribute-C1-Gamma?   binary
       +-- attribute-D-Delta?   binary

Maybe there is a way to fix this to be saner.
It's all in a git repo,   https://github.com/mcr/yang-augment-test

The file "practice1.sh" uses the method with multiple uses, which I think it
was has been recommended, but looks weird above:

  grouping module-D1-grouping {
    description "Grouping for module D1";
    container module-D1-things {
      uses vA:module-A-grouping;
      uses vB:module-B-grouping;
      uses vC1:module-C1-grouping;
      leaf attribute-D-Delta {
        type binary;
        description "Delta";
      }
    }
  }

The file practice2.sh uses the old method we were using with augment, which
does *not* work.



--
Michael Richardson <[e-mail]>, Sandelman Software Works
 -= IPv6 IoT consulting =-