Skip to content Skip to sidebar Skip to footer

When To Use `parent` And When To Use `root` As First Argument In A Graphql-resolver Function

I've seen some tutorials and examples out there who are sometimes using parent and sometimes using root for their first argument in a graphql resolver. What would be the correct na

Solution 1:

There's not an established convention here. I've seen root, parent, and obj all used in different reference guides. At the end of the day you can call your function arguments whatever you want -- your code will function the same.

That said, I typically use the name of the type that field being resolved in part of. So for a type like

type RoomRental {
  user: User
  date: Date
}

the resolver signature looks like this:

user: (roomRental, args, context, info) => {
  //
}

I find that this makes it easier to reason about what your resolver is doing when you (or a teammate) are reading the code. Again, this is just my preference, but you may find it helps make your code a little more digestible.

For completion, I'd also add that GraphQL.js does have a configurable root value. This value is passed down as the first argument in the resolvers of root-level fields (that is, fields on types like Query or Mutation). There's not much utility in passing things through the root -- anything you might pass in that way should probably go in your context instead. However, the root value is part of the implementation, and may be why you see a lot of examples use root as the name for that first argument.

Post a Comment for "When To Use `parent` And When To Use `root` As First Argument In A Graphql-resolver Function"