Hi, I’m trying to create a set of buckets driven by a config file. In this list,

buckets:
  - name: a
    is_primary: true
    region: <region-1>
    lifecycle_rules:
      - <rules_config>
    replication:
      region: <region-2>
      storage_class: STANDARD
      name: a-replica
    versioning:
      enabled: true
  - name: a-replica
    region: <region-2>
    lifecycle_rules:
        - <rules_config>
    replication:
      region: <region-1>
      name: a

where the first item replicates to the second and vice-versa. There could be more buckets configured like this… I’m trying to follow the bi-directional replication sample here. The sample captures each resource returned to reference it in another spot in the sample. (EX east_bucket_v2 = … and later east_bucket_v2.id is used. Is there a way to do this driven by the config file, and in a loop where there can be multiple sets of replicas defined? Ex. (if I add another 2 bucket definitions like above for a total of 4 buckets)?

Yes, it's ok, it's just a normal programming problem. You can store the already-created buckets in a map of name-to-bucket. However you should do this only for learning purposes. This technique is vulnerable to unintended reordering which will cause deletes and re-creates. Also, nearly all production-like use cases would depend on you knowing in advance that you have a certain set of buckets, so you would configure them in your chosen language. Configuring them in a YAML file brings complexities that are not needed.