Benjie
(The errorPaths would help a GraphQL client to know a) those positions are definitely errors, not custom scalars, and b) to know where the errors occurred without having to walk...
In my experience schemas that accept JSON do minimal validation on it; walking the entire tree to ensure it doesn't contain `__$$error` at any position at any depth might be...
Counterproposal: this should be a request option not a directive: https://github.com/graphql/nullability-wg/discussions/86
Spec edits (and reference implementation) of the request parameter are now complete: - #1153
Interesting find! I'd be open to someone opening a pull request to address this; but it's so rare for foreign key relations to have a default value that I'm not...
Is this specifically an issue with the Relay preset? I don’t think what you shared above would be non-nullable normally.
See Discord chat starting https://discord.com/channels/489127045289476126/498852330754801666/1319336710647189546
Depth of 5 is definitely very generous; for the GraphQLInt case that was suggested as a motivational example, I traced (without trying) this path: `GraphQLInt.mjs ---> GraphQLScalar.mjs ---> assertName.mjs --->...
You're taking me way off topic, Martin, but I've been thinking about this for a while, and we have: - RFC documents (now at https://github.com/graphql/graphql-wg/tree/main/rfcs but previously in other places...
Back on topic: I've also figured that RFCX can "override" other RFC labels, so I'm allowing items that go RFCX to retain their "highest level achieved" RFC status. So you...