Skip to content

model

modeling your graphql types is done with pydantic, and making the model compatible with the ql library, use lq.model

the @ql.model and how it works

this decorator defines couple of attributes that are required for graphql apis and are used across the library.

Danger

do not access those properties directly because they might be change in the future, the ql library provides standard ways to access those properties values in a expected manner

__ql_query_name__

this property contains the model query name, since we likely want to query object from graphql with different name then what we defined in python.

the default value is the class name

Example 
import ql
from pydantic import BaseModel


@ql.model
class Foo(BaseModel):
    field: str


@ql.model(query_name="different")
class MyClass(BaseModel):
    field: str


assert Foo.__ql_query_name__ == "Foo"
assert MyClass.__ql_query_name__ == "different"

# value is used when querying they model
assert ql.query((Foo, (ql._(Foo).field,))) == "{Foo{field,__typename}}"
assert ql.query((MyClass, (ql._(MyClass).field,))) == "{different{field,__typename}}"

__ql_typename__

graphql has the builtin-field __typename which returns the name of the returned object.

graphql has the option for interfaces, that different types can implement, thus we can query an interface but the concrete type is different

__ql_implements__

graphql supports interfaces that we can think about it as inheritance in python, so we define a base class (interface) and we have different types that inherite from the base class.

the __ql_implements__ will return list of models that implement the class

Example 

import ql
from pydantic import BaseModel


@ql.model
class Human(BaseModel):
    first_name: str
    last_name: str


@ql.model
class Male(Human):
    pass


@ql.model
class Female(Human):
    pregnant: bool

# -- (1)
assert ql.implements(Human) == (Male, Female)

  1. api documentation for implements

__ql_query_fields_enum__

this attribute is an enum, the keys are the configured field names, the value is the configured query name with ql.metadata

Example 

class ql
from typing import Annotated
from pydantic import BaseModel


@ql.model
class Foo(BaseModel):
  something: Annotated[str, ql.metadata(query_name="different")]  # -- (1)


@ql.model
class Foo2(BaseModel):
  another: str

# -- (2)
assert ql.model_queryable_fields(Foo).something.value == "different"
assert ql._(Foo).something.value == "different"
assert ql._(Foo2).another.value == "another"

  1. notice that we change the query_name
  2. a safe way to get the information; for api information click here