use field options to refine replacements

[natasha41575]

#3492, #3772

Use delimiter, index, and create options.

@monopole I was hoping you could clarify what you intended for the encoding option to do (from this comment).

[natasha41575]

@monopole updated to allow prefixing/suffixing and added a test with a path

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: natasha41575

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 [natasha41575]

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

[KnVerey]

This also seems like invalid configuration to me that should result in an error. It makes sense to provide a way to target the last element regardless of how many there are, but I'd say that should be done with a specific value, i.e. -1, or for negative indexes to be supported in general. But I feel like > current length and < 0 both meaning "last" could get confusing, especially given that current length can change independently of the Kustomization this is written in.

Edit: making negative indexing select the end when specified in source and indicate a prefix in target is also confusing

[natasha41575]

I have updated it to throw an error if the source index is out of bounds

[monopole]

It's a good idea to be relaxed about inputs (and strict on outputs).

However, I cannot think of a no-surprises default behavior for an out-of-range index after the split when trying to get the source value.

So - seems that it should error (fail build). Meaning that a split resulting in an empty array is always an error.

[monopole]

/lgtm