Support namespace filtering when reading from files

[Treenhan]

The idea is when running ingress2gateway command, we should be able to specify which namespace we want to use. If a namespace is explicitly specified, the command will only process the Ingress resources within that namespace. However, if no namespace is provided, the command will process Ingress resources across all namespaces.

/kind feature
/kind test

[k8s-ci-robot]

@Treenhan: The label(s) kind/test cannot be applied, because the repository doesn't have them.

Details

In response to this:

The idea is when running ingress2gateway command, we should be able to specify which namespace we want to use. If a namespace is explicitly specified, the command will only process the Ingress resources within that namespace. However, if no namespace is provided, the command will process Ingress resources across all namespaces.

/kind feature
/kind test

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.

[LiorLieberman]

/assign @Treenhan

[levikobi]

Hi @Treenhan,
I think we should keep the same functionality for reading from a file as reading from the cluster.
At the moment, when reading from the cluster without specifying a namespace you will only process objects from the current active namespace, as opposed to all the namespaces.
If you want to read from all namespaces, you have a --all-namespaces flag.

For reference, #30

[LiorLieberman]

@Treenhan whats the status of it? do you have any blockers you need help with?

[Treenhan]

I'm still deciding what the default behavior should be if no namespace is specified in the YAML/JSON files. I assume that not all machines converting these files will have access to cluster information, or they don't have a cluster at all but just want to convert the files. So they won't be able to grab the default namespace from the cluster context.

My thinking is that the default behavior should be to read everything, unless users want to choose a specific namespace. The reason is that without cluster context, there's no way to know what the default namespace should be.

@LiorLieberman @levikobi any ideas?

[Treenhan]

I think I will go with this:

• Without any flag, use the namespace in the active context, if there is no context, it will convert all (this allows users to use this tool without having kubernetes running)
• Use --all-namespaces to convert all
• Use --namespace X to filter out specific namespaces

[levikobi]

I think I will go with this:

• Without any flag, use the namespace in the active context, if there is no context, it will convert all (this allows users to use this tool without having kubernetes running)
• Use --all-namespaces to convert all
• Use --namespace X to filter out specific namespaces

Sounds good to me @Treenhan !

[LiorLieberman]

+1 sgtm

[Treenhan]

This is now done in #43