django+graphql的基础配置

官网地址

一、开发环境

• 1、python3.6
• 2、django2.0
• 3、window10

二、项目搭建

• 1、创建一个虚拟空间mkvirtualenv 空间名

• 2、创建一个django项目

• 3、安装graphql的依赖包

  pip install graphene-django
  

• 4、创建一个组件blog

• 5、把组件blog及graphene_django注入到app中

• 6、在settings.py中配置mysql数据库连接

三、书写blog的内容

• 1、在models.py中写上数据模型

  from django.db import models
  
  
  # Create your models here.
  class User(models.Model):
      name = models.CharField(max_length=100, verbose_name="博主名字")
      gender = models.CharField(max_length=6, choices=(('male', u'男'), ('female', '女')), default='female',
                                verbose_name='性别')
      create_at = models.DateTimeField(auto_now_add=True, verbose_name='创建时间')
  
  
  class Blog(models.Model):
      title = models.CharField(max_length=100, verbose_name='标题')
      user = models.ForeignKey(User, null=True, blank=True, on_delete=models.SET_NULL, verbose_name='博主名字')
      content = models.TextField(verbose_name='博客内容')
      create_at = models.DateTimeField(auto_now_add=True, verbose_name='创建时间')
      update_at = models.DateTimeField(auto_now=True, verbose_name='更新时间')
  

• 2、新建一个schema.py文件

  #!/usr/bin/env python
  # encoding: utf-8
  
  import graphene
  from graphene_django.types import DjangoObjectType
  from .models import User, Blog
  
  
  class UserType(DjangoObjectType):
      class Meta:
          model = User
  
  
  class BlogType(DjangoObjectType):
      class Meta:
          model = Blog
  
  
  # 定义动作约素输入类型
  class UserInput(graphene.InputObjectType):
      name = graphene.String(required=True)
      gender = graphene.String(required=True)
  
  
  class BlogInput(graphene.InputObjectType):
      title = graphene.String(required=True)
      user = graphene.Int(required=True)
      content = graphene.String(required=True)
  
  
  # 定义一个创建user的mutation
  class CreateUser(graphene.Mutation):
      # api的输入参数
      class Arguments:
          user_data = UserInput(required=True)
  
      # api的响应参数
      ok = graphene.Boolean()
      user = graphene.Field(UserType)
  
      # api的相应操作，这里是create
      def mutate(self, info, user_data):
          user = User.objects.create(name=user_data['name'], gender=user_data['gender'])
          ok = True
          return CreateUser(user=user, ok=ok)
  
  
  # 定义一个创建博客的mutation
  class CreateBlog(graphene.Mutation):
      class Arguments:
          blog_data = BlogInput(required=True)
  
      blog = graphene.Field(BlogType)
  
      def mutate(self, info, blog_data):
          # 插入到数据库中
          blog = Blog.objects.create(title=blog_data['title'], user_id=blog_data['user'], content=blog_data['content'])
          return CreateBlog(blog=blog)
  
  # 定义一个查询语句
  class Query(object):
      all_user = graphene.List(UserType)
      all_blog = graphene.List(BlogType)
  
      def resolve_all_user(self, info, **kwargs):
          # 查询所有book的逻辑
          return User.objects.all()
  
      def resolve_all_blog(self, info, **kwargs):
          # 查询所有title的逻辑
          return Blog.objects.all()
  
  

• 3、在跟目录(和settings.py同级)创建一个项目的总schema.py

  import graphene
  import book.schema, blog.schema
  
  
  class Query(blog.schema.Query, graphene.ObjectType):
      # 总的Schema的query入口
      pass
  
  
  class Mutations(graphene.ObjectType):
      # 总的Schema的mutations入口
      create_user = blog.schema.CreateUser.Field()
      create_blog = blog.schema.CreateBlog.Field()
  
  
  schema = graphene.Schema(query=Query, mutation=Mutations)
  

• 4、配置url地址

  from django.contrib import admin
  from django.urls import path
  from graphene_django.views import GraphQLView
  from .schema import schema
  urlpatterns = [
      path('admin/', admin.site.urls),
      path('graphql/', GraphQLView.as_view(graphiql=True, schema=schema)),
  ]
  

• 5、生成数据库映射及启动项目,直接在浏览器上访问

四、可以对上面的代码调整

• 1、把Mutations也单独定义在各自的schema.py中

  # 定义一个总的mutation出口
  class Mutation(graphene.AbstractType):
      create_user = CreateUser.Field()
      create_blog = CreateBlog.Field()
  

• 2、在总的schema.py中引入类型Query一样的操作

  class Mutations(blog.schema.Mutation, graphene.ObjectType):
      # 总的Schema的mutations入口
      pass
  

• 3、输入数据类型可以直接定义在mutation里面

  class CreateUser(graphene.Mutation):
      # api的输入参数(类名可以随便定义)
      class Arguments:
          name = graphene.String(required=True)
          gender = graphene.String(required=True)
  
      # api的响应参数
      ok = graphene.Boolean()
      user = graphene.Field(UserType)
  
      # api的相应操作，这里是create
      def mutate(self, info, name, gender):
          user = User.objects.create(name=name, gender=gender)
          ok = True
          return CreateUser(user=user, ok=ok)
  

五、Query语句中使用条件查询

• 1、app的schema(官方案例)

  import graphene
  from graphene_django.types import DjangoObjectType
  from .models import Category, Ingredient
  
  
  class CategoryType(DjangoObjectType):
      class Meta:
          model = Category
  
  
  class IngredientType(DjangoObjectType):
      class Meta:
          model = Ingredient
  
  
  # 定义一个查询
  class Query(object):
      # 定义一个根据id或者name查询的
      category = graphene.Field(CategoryType,
                                id=graphene.Int(),
                                name=graphene.String())
      # 查询全部的
      all_categories = graphene.List(CategoryType)
      # 根据条件查询
      ingredient = graphene.Field(IngredientType,
                                  id=graphene.Int(),
                                  name=graphene.String())
      # 查询全部的
      all_ingredients = graphene.List(IngredientType)
  
      def resolve_all_categories(self, info, **kwargs):
          return Category.objects.all()
  
      def resolve_all_ingredients(self, info, **kwargs):
          # We can easily optimize query count in the resolve method
          return Ingredient.objects.select_related('category').all()
  
      # 定义查询语句
      def resolve_category(self, info, **kwargs):
          id = kwargs.get('id')
          name = kwargs.get('name')
  
          if id is not None:
              return Category.objects.get(pk=id)
  
          if name is not None:
              return Category.objects.get(name=name)
  
          return None
  
      def resolve_ingredient(self, info, **kwargs):
          id = kwargs.get('id')
          name = kwargs.get('name')
  
          if id is not None:
              return Ingredient.objects.get(pk=id)
  
          if name is not None:
              return Ingredient.objects.get(name=name)
  
          return None
  

六、源码下载

七、查看博主更多文章