It would also help when testing, to be able to comment out large blocks of the mapping to reduce resync times while testing JQ in search-type relations.

+1. Our mapping file for Kubernetes is 500+ lines long with a lot of logic that is difficult to decipher.