How to remove map entries from selected nested maps properties

[james-lim-ev]

Hi yq community, I have a question here.

I have a yaml file looks kind of like this:

...
.
.
.
    TypeA:
      type: object
      field:
        - title: Field A
          properties:
            propA:
              ref: 'some/reference'
        - title: Field B
          properties:
            propB:
              ref: 'some/reference'
      properties:
        propA:
          ref: 'some/reference'
        propB:
          ref: 'some/reference'
.
.
.
    TypeB:
      type: object
      field:
        - title: Field C
          properties:
            propA:
              ref: 'some/reference'
        - title: Field D
          properties:
            propB:
              ref: 'some/reference'
      properties:
        propA:
          ref: 'some/reference'
        propB:
          ref: 'some/reference'
        propC:
          ref: 'some/reference'
        propD:
          ref: 'some/reference'
.
.
.
...

My objective is to remove the duplicated properties in the top properties section if the field section has already contain the same properties.

My desire output:

...
.
.
.
    TypeA:
      type: object
      field:
        - title: Field A
          properties:
            propA:
              ref: 'some/reference'
        - title: Field B
          properties:
            propB:
              ref: 'some/reference'
.
.
.
    TypeB:
      type: object
      field:
        - title: Field C
          properties:
            propA:
              ref: 'some/reference'
        - title: Field D
          properties:
            propB:
              ref: 'some/reference'
      properties:
        propC:
          ref: 'some/reference'
        propD:
          ref: 'some/reference'
.
.
.
...

How do I do this with yq? I've tried multiple approaches with pick and omit but can't get anywhere I want it to go.

I'd really appreciate it if someone could figure out any working approaches.

[james-lim-ev]

so the closest approach I came about is

yq '.. | select(has("field") and has("properties")).properties |= omit(["propA","propB"])' myfile.yaml

but I still need to specify the string array. Is there a way that I can plug field[].properties | keys as a string array into omit ?

I tried

yq '.. | select(has("field") and has("properties")).properties |= omit(select(has("field") and has("properties")).field[].properties | keys)' myfile.yaml

but that didn't work for me.

[mikefarah]

yq 'with(.. | select(kind == "map" and .type == "object"); 
    ([.field[].properties | keys] | flatten ) as $k | .properties |= omit($k) ) ' 

• with(.. | select(kind == "map" and .type == "object"); <<exp>> will iterate over every type: object node
• Then we need to get an array of the keys the we want to omit. ([.field[].properties | keys] | flatten ) and save that as $k
• And finally we update the properties map to omit those keys .properties |= omit($k)

[mikefarah]

TypeA:
  type: object
  field:
    - title: Field A
      properties:
        propA:
          ref: 'some/reference'
    - title: Field B
      properties:
        propB:
          ref: 'some/reference'
  properties: {}
TypeB:
  type: object
  field:
    - title: Field C
      properties:
        propA:
          ref: 'some/reference'
    - title: Field D
      properties:
        propB:
          ref: 'some/reference'
  properties:
    propC:
      ref: 'some/reference'
    propD:
      ref: 'some/reference'

[mikefarah]

If you also want to delete empty property maps:

yq 'with(.. | select(kind == "map" and .type == "object");  (  
       [.field[].properties | keys] | flatten ) as $k | .properties |= omit($k) 
   ) |  del(.. | select(has("properties") and .properties | length == 0).properties)'  data1.yaml

[james-lim-ev]

Thank you very much @mikefarah !