delete entire document from multi-document file/input

[travisghansen]

I would like to remove an entire document (preferably by some selector, but by index would be crude enough) from the input.

apiVersion: v1
kind: Namespace
metadata:
  labels:
    app.kubernetes.io/component: rabbitmq-operator
    app.kubernetes.io/name: rabbitmq-system
    app.kubernetes.io/part-of: rabbitmq
  name: rabbitmq-system
---
apiVersion: apiextensions.k8s.io/v1
kind: CustomResourceDefinition
metadata:
  annotations:
    controller-gen.kubebuilder.io/version: v0.15.0
  name: bindings.rabbitmq.com
spec:
  group: rabbitmq.com
  names:
    categories:
    - rabbitmq
    kind: Binding
    listKind: BindingList
    plural: bindings
    singular: binding
  scope: Namespaced
---
foo: bar
---
bar: baz

I would like to remove any document for example which contains kind: Namespace, or document at index 0.

[ccoVeille]

I would use

$ find -name  "*.y*ml" -print0 | xargs -0 -r -- fgrep -L -v "kind: Namespace"

You will get the list of files that do not contain "kind: Namespace"

But it's possible someone could do it with yq

[travisghansen]

I could have been more clear, it’s a single file with multiple yaml documents inside. It’s not multiple files. Although that may be an option to find a tool which will write all the documents in a single file to a file-per-doc situation.

[ccoVeille]

OK it's clearer. Let's wait and see what other people suggests

[mikefarah]

Not currently on a machine with yq - but you should be able to do it like so

yq 'del(select(.kind == "Namespace"))' data.yaml

Explanation:

• select(.kind == "Namespace") returns all the top level documents that match
• wrapping that in del( <exp>) will delete those matches
• See del for more details
• You can also use the documentIndex/di as a filter: del(select(di == 0)), see document index for more info :)

[travisghansen]

Worked perfectly, thanks!