typegql 2.0.1

A Python GraphQL library that makes use of type hinting and concurrency support with the new async/await syntax.

TypeGQL

A Python GraphQL library that makes use of type hinting and concurrency support with the new async/await syntax.

DISCLAIMER

This library is still in it’s infancy, so use with caution and feel free to contribute.

Installation

pip install typegql

Usage

The following demonstrates how to use typegql for implementing a GraphQL API for a library of books. The example can be found in typegql/core/examples and you can run it with Sanic by executing python <path_to_example>/server.py

Define your query

from typing import List
from typegql.core.graph import Graph, Connection
from typegql.examples.library.types import Author, Category
from typegql.examples.library.types import Book
from typegql.examples.library import db

class Query(Graph):
    books: List[Book] = Field()
    authors: List[Author] = Field()
    categories: List[Category] = Field()

    books_connection: Connection[Book] = Field(description='Relay connection')

    async def resolve_authors(self, info, **kwargs):
        return db.get('authors')

    async def resolve_books(self, info, **kwargs):
        return db.get('books')

    async def resolve_categories(self, info, **kwargs):
        return db.get('categories')

   async def resolve_books_connection(self, info, **kwargs):
        data = db.get('books')
        return {
            'edges': [{
                'node': node
            } for node in data]}

Define your types

from dataclasses import dataclass
from datetime import datetime
from decimal import Decimal
from enum import Enum
from typing import List

from typegql import Field, ID, OptionalField, ReadonlyField
from typegql.core.graph import Graph
from examples.library import db


class Gender(Enum):
    MALE = 'male'
    FEMALE = 'female'


class GeoLocation(Graph):
latitude: Decimal = Field()
longitude: Decimal = Field()

def __init__(self, latitude, longitude):
    self.latitude = latitude
    self.longitude = longitude


@dataclass
class Author(Graph):
    """Person that is usually a writer"""

    id: ID = ReadonlyField()
    name: str = Field()
    gender: Gender = OptionalField()
    geo: GeoLocation = OptionalField()


@dataclass
class Category(Graph):
    id: ID = ReadonlyField()
    name: str = Field()


@dataclass
class Book(Graph):
    """A book... for reading :|"""

    id: ID = ReadonlyField()
    author_id: ID = Field()
    title: str = OptionalField()
    author: Author = ReadonlyField(description='The author of this book')
    categories: List[Category] = OptionalField()
    published: datetime = OptionalField()
    tags: List[str] = OptionalField()

    def __post_init__(self):
        self.published = datetime.strptime(self.published, '%Y-%m-%d %H:%M:%S')

    async def resolve_author(self, info):
        data = filter(lambda x: x['id'] == self.author_id, db.get('authors'))
        data = next(data)
        author = Author(**data)
        author.gender = Gender[author.gender.upper()].value
        if 'geo' in data:
            author.geo = GeoLocation(**data.get('geo'))
        return author

    async def resolve_categories(self, selections, name=None):
        data = filter(lambda x: x['id'] in self.categories, db.get('categories'))
        for d in data:  # showcasing async generator
            yield Category(**d)

    def resolve_tags(self, selections):
        return ['testing', 'purpose']

Using Fields instead

You can use the following fields to define your GraphQL schema:

Field, InputField, RequiredField, OptionalField

For example:

from typegql import Field, Connection, OptionalField


class Query(Graph):
    authors: Author = Field()
    categories: Category = Field(description="what's this?")
    books_connection: Connection[Book] = OptionalField()

Run your query

from typegql.core.schema import Schema
from examples.library.query import Query


schema = Schema(Query)
query = '''
query BooksConnection {
  books_connection {
    edges {
      node {
        id
        title
        published
        author {
          id
          name
        }
      }
    }
  }
}
'''

async def run():
    result = await schema.run(query)

Client

TypeGQL supports DSL client for working with a GraphQL API. The client automatically converts snake to camelcase. set camelcase=False if this is not desired

pip install typegql[client]

For example:

from typegql.client import Client

async with Client(url, camelcase=True) as client:
    await client.introspection()
    dsl = client.dsl
    query = dsl.Query.books_connection.select(dsl.BooksConnection.total_count)
    doc = dsl.query(query)

    status, result = await client.execute(doc)

Change Log

2.0.1 [2019-01-19]

• all properties that don’t have a Field instance assigned to them will be ignored by the Schema

• updates docs & example to reflect 2.0 changes

• fixed a bug when using a List argument in mutations

1.0.7 [2018-12-09]

• bug fixing

• adds support for camelcase in Client

1.0.1 [2018-11-19]

• adds support for client DSL

Initial

• added graphql-core-next as a baseline for all GraphQL operations

TODO

• testing

• travis

• more testing

• please help with testing :|