Add cluster and cluster bindings

[shashwathi]

Fixes #21

Addresses the following requirements in issue

• pipelineparams should have clusters
• pipelines should have clusterBindings
• tasks should have cluster inputs

/assign @bobcatfish
/assign @pivotal-nader-ziada

[knative-prow-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: shashwathi
To fully approve this pull request, please assign additional approvers.
We suggest the following additional approver: bobcatfish

If they are not already assigned, you can assign the PR to them by writing /assign @bobcatfish in a comment when ready.

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[nader-ziada]

@shashwathi it might conflict with the optional types PR #56 depending on what gets merged in first, but in general take a look at the api conventions to add the +optional comment

[shashwathi]

@pivotal-nader-ziada : I will update this PR to include optional comment for fields I have added.

/hold
Hold this PR until #56 is merged

[shashwathi]

/test pull-knative-build-pipeline-unit-tests

[shashwathi]

/hold cancel

Updated with optional tags

[nader-ziada]

/lgtm

[tejal29]

few nits and go coding conventions

[tejal29]

IIUC, inputName stands for what input this cluster is bound to? Can you rename it to boundTo or something?

[shashwathi]

This is auto generated code. If we change the definition, crds should change as well.

[tejal29]

Lets get ClusterBinding and SourceBinding structs fields in sync.
Shd we rename ClusterName to Key and also drop SourceKey to Key ?
SourceBinding.Key and ClusterBindings.Key avoids stuttering.

        clusterBindings:
          -Key: test

Can we consider renaming ClusterBinding.InputName and SourceBinding.Name to something like BoundTo.
We can also define a struct Binding and then include common fields in those.

[shashwathi]

Shd we rename ClusterName to Key and also drop SourceKey to Key ?

I agree with you to stay in sync across these bindings to make the pipeline yaml more readable. I like key for bindings. 👍

Can we consider renaming ClusterBinding.InputName and SourceBinding.Name to something like BoundTo

I think keyword bindings in ClusterBindings/ SourceBindings convey that this is templated to an another definition(like task). I think BoundTo might be redundant.

How about explicitly calling out taskInputName(since both are bound to tasks) or keep the original key inputName (more generic)?

[nader-ziada]

agreed key is a better name the key field

• ClusterBinding.InputName becomes ClusterBinding.Key
• SourceBinding.SourceKey becomes SourceBinding.Key (think this should be a separate PR since this PR here is about adding ClusterBinding)

But since we already have the type name including the word Binding I feel the field name should be enough to clarify what it is

eg:

apiVersion: pipeline.knative.dev/v1beta1
kind: Pipeline
metadata:
  name: wizzbang-pipeline
  namespace: default
spec:
    tasks:
      - name: 'test'
        taskRef:
          name: test-wizzbang-task
        inputsourceBindings:
          - name: repoUnderTest
            key: wizzbang   <- changed to key in different PR
        clusterBindings:
          - name: test-cluster  <- name of this cluster binding
            key: gke-cluster-27    <- key to refer to
    .........

[tejal29]

/lgtm

[nader-ziada]

/lgtm as well

@bobcatfish can you please take a look and merge if you think its all good, trying to get the PRs in before #57 makes big changes to the repo to reduce conflicts