Queries are how we load our data from the database and into our application / into our models.

Aphrodite generates one query builder per Node in your schema. These query builders can be used to load data, traverse edges between data, join, union, etc.

Lets take a look at a Todo schema and its generated Query.

Todo node:

Todo as Node {
  id: ID<Todo>
  text: string
  completed: bool
}

Generated Todo Query:

export default class TodoQuery extends DerivedQuery<Todo> {
  static create(ctx: Context);
  static fromId(ctx: Context, id: SID_of<Todo>);

  whereId(p: Predicate<Data["id"]>);
  whereText(p: Predicate<Data["text"]>);
  whereCompleted(p: Predicate<Data["completed"]>);
}

Taking a look at the query builder, we have one where method per field that exists on the model. We can use these to filter the data being loaded. E.g.,

Todo.queryAll(ctx).whereText(P.equals(someText)).whereComplete(P.equals(false));

Edge Traversals

We've only seen loading and filtering data of a single type. Lets take a look at traversing edges between types.

To traverse edges between types, we'll need to add an edge on our schema. We'll extend the Todo schema to include a Person schema and an edge from the Person to their Todods.

Person as Node {
  id: ID<Person>
  name: string
} & OutboundEdges {
  todos: Edge<Todo.ownerId>
}

Todo as Node {
  id: ID<Todo>
  text: string
  completed: bool
  ownerId: ID<Person>
}

After running codegen, this'll create query builder for person that has a queryTodos method.

class PersonQuery {
  queryTodos(): TodoQuery;
}

The queryTodos method allows us to pivot from Person to their Todos.

Example -- getting all todos for all people:

Person.queryAll(ctx).queryTodos().whereComplete(P.equals(false));