spec: JSON:API Graphs

[runspired]

As EmberData builds out its JSON:API support and Schema driven behaviors, we've noticed that some useful capabilities are less usable by JSON:API users due to that spec not yet considering the overall ergonomics of querying and consuming graphs.

This adds a specification with an extension and profile that APIs may implement to make use of these capabilities.

Rendered

[dsmyda-ab]

for exclusion could do something like:

{
  -name
}

and if the only fields are exclusion fields in the brackets ({}) then you receive all fields minus excludes. You may not mix included and excluded fields.

[vlascik]

Great idea, JSON:API needs this.

IMO though, for a query language to be really useful it should also contain comparison operators, such as:

eq (equals)
neq (not equals)
gt (greater than)
lt (less than)
gte (greater than or equal to)
lte (less than or equal to), 
in (list)
not_in (list), 
contains (string) or like

and perhaps some basic grouping and logic (and, or, not) as in https://dgraph.io/docs/graphql/queries/and-or-not/ or https://hasura.io/docs/latest/queries/postgres/filters/boolean-operators/
and maybe even aggregate functions (sum, avg, min, max) as in https://dgraph.io/docs/graphql/queries/aggregate/
and multi-field ordering (order_by: ['date': 'desc', 'name': 'asc'], or something similar to https://dgraph.io/docs/graphql/queries/order-page/