DAG as template

[reibuildots]

Is it possible to declare 3 types of DAGS (d1, d2, d3) each DAG has different tasks to than create: d1 -> [d2, d3]
I tried to generate steps with DAG as a template but I run into this exception:
AttributeError: 'DAG' object has no attribute 'template'

Example for nested DAG:
https://github.com/argoproj/argo-workflows/blob/master/examples/dag-nested.yaml

[reibuildots]

This is how I solved it, Can you tell me if it is the best practice?

    with Workflow(
            generate_name="steps-",
            entrypoint="hello-hello-hello"
    ) as w:
        echo = Container(
            name="echo",
            image="alpine:3.7",
            command=["echo", "{{inputs.parameters.message}}"],
            inputs=[Parameter(name="message")],
        )
        with DAG(name="diamond1") as d1:
            A = echo(name="A", arguments={"message": "A"})
            B = echo(name="B", arguments={"message": "B"})
            C = echo(name="C", arguments={"message": "C"})
            D = echo(name="D", arguments={"message": "D"})
            A >> [B, C] >> D

        with DAG(name="diamond2") as d2:
            A = echo(name="A", arguments={"message": "A"})
            B = echo(name="B", arguments={"message": "B"})
            C = echo(name="C", arguments={"message": "C"})
            D = echo(name="D", arguments={"message": "D"})
            A >> [B, C] >> D

        with DAG(name="diamond3") as d3:
            A = echo(name="A", arguments={"message": "A"})
            B = echo(name="B", arguments={"message": "B"})
            C = echo(name="C", arguments={"message": "C"})
            D = echo(name="D", arguments={"message": "D"})
            A >> [B, C] >> D

        with Steps(name="hello-hello-hello") as s:
            Step(name="hello1", template=d1)

            with s.parallel():
                Step(name="hello2a", template=d2)
                Step(name="hello2b", template=d3)

[flaviuvadan]

Hey @reibuildots! Yea, that's the right way to do it for now! This PR makes the DAG callable so you will be able to express DAG dependencies exactly how you want: d1 -> [d2, d3]

[flaviuvadan]

#622