graphql-spec
graphql-spec copied to clipboard
graphql
GraphQL is a query language and execution engine tied to any backend service.
This formalizes the proprosed feature within `graphql-js` whereby the internal method provided by JavaScript for runtime type resolution is allowed to return an intermediate interface. See: Issue: https://github.com/graphql/graphql-js/issues/3253 PR: https://github.com/graphql/graphql-js/pull/3599...
Namespaces
At the moment all GraphQL types share one global namespace. This is also true for all of the fields in mutation/subscription type. It can be a concern for bigger projects...
A [recent talk about Relay 2](https://speakerdeck.com/wincent/relay-2-simpler-faster-more-predictable) suggested adding arguments to query fragments via directives. I think first-class support for arguments in fragments would give a few benefits: - Smaller variable...
Hi folks, I tried asking this both in slack and discord but got no replies. I'm new to GraphQL and am trying to figure out the correct way of implementing...
It makes sense often to have a field that could return, for example, an `Int` or a `String`. It would be convenient to be able to use a `resolveType`-like mechanism...
Follow up of ~~[the @oneField directive](https://github.com/graphql/graphql-spec/pull/586)~~ and ~~[the Tagged type](https://github.com/graphql/graphql-spec/pull/733)~~. **Introducing**: OneOf Input Objects. **OneOf Input Objects** are a special variant of Input Objects where the type system asserts that...
This is currently disallowed by the specification: ```graphql enum Foo { bar baz } extend enum Foo { baz @myDirective } ``` From https://spec.graphql.org/draft/#sec-Enum-Extensions.Type-Validation: > 3. All values of an...
## TL;DR It would be really nice to have a set of GraphQL Documents which collectively express the entire GraphQL grammar. This would include atypical constructions such as random insertions...
This provides definitions at the introduction of "Type System" for _leaf type_, _composite type_, and _abstract type_. Then, these definitions are used throughout the spec. --- Inspired by #957 which...
Whilst reviewing #966 I spotted this; I don't believe this is a breaking change since it can be derived from: > The `errors` entry in the response is a **non-empty**...