Adding allowDefaultResolve option to makeExecutableSchema to allo…

[sethtippetts]

Adding allowDefaultResolve option to makeExecutableSchema to allow non-scalar types to use the GraphQL default resolve function.

Issue: #44

[apollo-cla]

@sethtippetts: Thank you for submitting a pull request! Before we can merge it, you'll need to sign the Meteor Contributor Agreement here: https://contribute.meteor.com/

[helfer]

Rather than passing in allowDefaultResolve to assertResolveFunctionsPresent, how about we just have an on/off switch? I think that will be easier for folks to understand because either they get all of the checks or they get none. There will be other ways to check arguments, and I can already foresee someone wanting to turn off checks for resolve functions present in resolvers but not in the schema. Having too many fine-grained controls may make it harder for people to understand what's really going to happen when they toggle a combination of options.

If we do that, we can just call the argument checkResolvers, and do something like this:

if (checkResolvers) {
  assertResolveFunctionsPresent(schema);
}

While we're at it, we should also pass that in to addResolveFunctionsToSchema, and turn off the checks there that make sure the types / fields defined in the resolvers actually exist.

What do you think? Happy to discuss this further!

[helfer]

@sethtippetts Thanks a lot for the PR, I could merge it as is, but I think we have an opportunity here to kill two birds with one stone (see in-line comment). Let me know what you think!

[helfer]

Talked about this a little more with @stubailo and another option would be to pass in a resolverValidationOptions parameter, which is an object:

resolverValidationOptions = {
  requireResolversForNonNull: true,
  requireResolversForArgs: true,
  rejectExtraResolvers: true,
}

[sethtippetts]

@helfer I like the resolverValidationOptions a lot! I initially wrapped assertResolveFunctionsPresent in an if but thought people may still want to validate a resolver when the schema says it accepts arguments.

I've updated the PR to use resolverValidationOptions

[helfer]

@sethtippetts Awesome! I'll open a separate issue to keep track of the fact that we'd like to add the 'rejectExtraResolvers' option at some point.