Inconsistent gRPC REST endpoint format

[ryanchristo]

Summary

We currently have no clear guidelines on how to write gRPC REST endpoints and we are using a few different formats.

When we query for projects by credit class (we use classes as the root and add projects after {class_id}):

"/regen/ecocredit/v1/classes/{class_id}/projects"

When we query for projects by reference id (we use projects as the root and then by-reference-id):

"/regen/ecocredit/v1/projects/by-reference-id/{reference_id}"

When we query for batches by class (we use batches as the root and then class):

"/regen/ecocredit/v1/batches/class/{class_id}"

Problem Definition

Using a consistent format for gRPC REST endpoints will provide a better user experience.

Proposal

[ work in progress ]

---

For Admin Use

• Not duplicate issue
• Appropriate labels applied
• Appropriate contributors tagged
• Contributor assigned/self-assigned

[ryanchristo]

It looks like other sdk projects are using the format <object>_by_<value> (the sdk and osmosis) and this seems like the right direction but best practices for REST endpoints is to use - rather than _.

The data module in general could use more descriptive query method names and I'm not sure we want to make all the changes currently in #1156 but I thought it would be good to demonstrate what more descriptive endpoints might look like.

[ryanchristo]

Also, we may want to make use of additional_bindings in some cases. In particular, the first example in this issue shouldn't have to be one or the other but rather both of the following:

"/regen/ecocredit/v1/classes/{class_id}/projects"

"/regen/ecocredit/v1/projects-by-class/{class_id}"