Lifecycle of a request

[!NOTE]

You can see this YouTube video after you read this doc or before it to get the most out of them:

Eric Baer - GraphQL Under the Hood.

1. Document parsing:
   • AKA Lexing and Parsing.
   • Transform the query (which is string) into something it understands.
2. Validation using the type system: is it valid when checked against the API’s schema?
   • Validation errors you might encounter.
3. Execution.
4. Mirror the returned data from resolver into the requested shape.
   • GraphQL place each resolved field into a key-value map with the field name (or alias) as the key and the resolved value as the value.
   • It starts from leaves and works its way back up to the root (this why our responses resemble our requests).
   • Scalar coercion converts values into the types expected by the schema (e.g. enums will be coerced).
   • GraphQL waits for promises to be resolved before trying to construct the response (just look at my createUser resolver, I do NOT have await and it is working just fine).
5. Return response:
   • GraphQL returns a JSON as a response for query operation which contains:
     • A data key:
       • This is mandated by GraphQL spec (ref).
       • May include partial data for the requested fields if errors were raised during the execution of some field resolvers (learn more here).
     • An error key:
       • Learn about different reasons that an error might occur here.
     • An extension key:
       • No restriction on what can goes inside it.
       • Spec only tell that it can be sent back as a top-level key.
       • Depends on the GraphQL implementation.
       • Can contain additional info about response, things like:
         • Rate limit consumption.
         • Telemetry data.
   • At least one of data or errors will be present on the response.