王大阳_

Django Rest Framework框架呼啦圈项目实战

文章目录

知识点

1. 什么是restful规范
2. 什么是drf？
3. drf如何帮助我们快速开发的？drf提供了那些功能？
4. 序列化：
5. 序列化：展示特殊的数据（choices、FK、M2M）可使用
drf--serializers的序列化与表单验证
drf与django小结
实现对文章的增删改查

部分字段更新
添加返回的字段（choices）

两种分页

drf PageNumberPagination
drf LimitOffsetPagination
drf 内置ListAPIView显示列表

1.呼啦圈

1.1 表结构设计
2. 筛选

drf的组件：内置了筛选的功能

3. drf视图应用

resful规范补充

request请求对象封装
API版本

局部版本控制
全局

知识点

1. 什么是restful规范

是一套规则，用于程序之间进行数据交换的约定。
他规定了一些协议，对我们感受最直接的的是，以前写增删改查需要写4个接口，restful规范的就是1个接口，根据method的不同做不同的操作，比如：get/post/delete/put/patch/delete.
初次之外，resetful规范还规定了：
	- 数据传输通过json
	
扩展：前后端分离、app开发、程序之间（与编程语言无关）

JSON:
    {
        name:'alex',
        age:18,
        gender:'男'
    }

以前用webservice，数据传输格式xml。
XML
	alex
	alex
	男

2. 什么是drf？

drf是一个基于django开发的组件，本质是一个django的app。
drf可以办我们快速开发出一个遵循restful规范的程序。

3. drf如何帮助我们快速开发的？drf提供了那些功能？

- 视图，APIView用处还不知道。
- 解析器，根据用户请求体格式不同进行数据解析，解析之后放在request.data中。
		在进行解析时候，drf会读取http请求头 content-type.
		如果content-type:x-www-urlencoded，那么drf会根据 & 符号分割的形式去处理请求体。
			user=wang&age=19
		 
		如果content-type:application/json,那么drf会根据 json 形式去处理请求体。
		 {"user":"wang","age":19}
- 序列化，可以对QuerySet进行序列化，也可以对用户提交的数据进行校验。
- 渲染器，可以帮我们把json数据渲染到页面上进行友好的展示。（内部会根据请求设备不同做不同的展示）

4. 序列化：

many=True or False

5. 序列化：展示特殊的数据（choices、FK、M2M）可使用

depth
source，无需加括号，在源码内部会去判断是否可执行，如果可执行自动加括号。【fk/choice】
SerializerMethodField，定义钩子方法。【m2m】

## 扩展
1. restful规范

URL中一般用名词：
http://www.luffycity.com/article/ (面向资源编程，网络上东西都视为资源)
根据请求不同做不同操作：GET/POST/PUT/DELTE/PATCH
筛选条件，在URL参数中进行传递：
http://www.luffycity.com/article/?page=1&category=1

一般传输的数据格式都是JSON


2. drf组件

…


3. 潜规则：类的约束

4. drf的配置

setting.py

REST_FRAMEWORK = {
…
}


5. 分页

- page
- offset

# 呼啦圈项目
## 创建Django程序

source ~/.bashrc
mkvirtualenv hula_django_framework
workon hula_django_framework
pip install djangorestframework


## drf的配置
![在这里插入图片描述](https://img-blog.csdnimg.cn/20200618160524166.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80Mzc0NjQzMw==,size_16,color_FFFFFF,t_70)
![在这里插入图片描述](https://img-blog.csdnimg.cn/20200618161606203.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80Mzc0NjQzMw==,size_16,color_FFFFFF,t_70)
## 创建程序并初始化数据库

python manage.py makemigrations
python manage.py migrate


![在这里插入图片描述](https://img-blog.csdnimg.cn/20200618184953117.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80Mzc0NjQzMw==,size_16,color_FFFFFF,t_70)
测试api
![在这里插入图片描述](https://img-blog.csdnimg.cn/20200618183714147.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3dlaXhpbl80Mzc0NjQzMw==,size_16,color_FFFFFF,t_70)
## drf 增删改查

```py
from django.shortcuts import render

# Create your views here.

from rest_framework.views import APIView
from rest_framework.response import Response
from api import models
from django.forms.models import model_to_dict


class DrfCategory(APIView):
    pass


class DrfCategoryView(APIView):
    def get(self, request, *args, **kwargs):
        """
        获取所有文章列表
        """
        pk = kwargs.get('pk')
        if not pk:

            data = models.Category.objects.all().values('id', 'name')
            data_list = list(data)

            return Response(data_list)
        category_object = models.Category.objects.filter(id=pk).first()
        data = model_to_dict(category_object)
        return Response(data)

    def post(self, request, *args, **kwargs):
        """
        增加一条分类信息
        """
        # name = request.POST.get('name')
        print(request.body)
        print(request.data)
        models.Category.objects.create(**request.data)
        return Response('成功')

    def delete(self, request, *args, **kwargs):
        """
        删除一条分类信息
        """
        pk = kwargs.get('pk')
        if not pk:
            return Response('请选择删除的 id')
        data = models.Category.objects.filter(id=pk).delete()
        if data[0] == 0:
            return Response('删除--无此数据')
        return Response('删除成功')

    def put(self, request, *args, **kwargs):
        """
        更新一条分类信息
        """
        pk = kwargs.get('pk')
        if not pk:
            return Response('请选择删除的 id')
        data = models.Category.objects.filter(id=pk).update(**request.data)
        print(data)
        if data == 0:
            return Response('更新--无此数据')
        return Response('更新成功')

到目前为止，使用drf与django不通在于取数据的方式

drf–serializers的序列化与表单验证

我们把变量从内存中变成可存储或传输的过程称之为序列化
把变量内容从序列化的对象重新读到内存里称之为反序列化

class NewCategorySerializer(serializers.ModelSerializer):
    """
    序列化
    """
    class Meta:
        model = models.Category
        # fields = "__all__"
        fields = ['id', 'name']


class NewCategoryView(APIView):
    def get(self, request, *args, **kwargs):
        pk = kwargs.get('pk')
        print(pk)
        if not pk:
            queryset = models.Category.objects.all()
            ser = NewCategorySerializer(instance=queryset, many=True)
            return Response(ser.data)
        else:
            model_object = models.Category.objects.filter(id=pk).first()
            ser = NewCategorySerializer(instance=model_object, many=False)
            return Response(ser.data)

    def post(self, request, *args, **kwargs):
        ser = NewCategorySerializer(data=request.data)
        if ser.is_valid():
            ser.save()
            return Response(ser.data)
        return Response(ser.errors)
    
    def put(self, request, *args, **kwargs):
        pk = kwargs.get('pk')
        category_object = models.Category.objects.filter(id=pk).first()
        ser = NewCategorySerializer(instance=category_object,data=request.data)
        if ser.is_valid():
            ser.save()
            return Response(ser.data)
        return Response(ser.errors)

    def delete(self, request, *args, **kwargs):
        """
        删除一条分类信息
        """
        pk = kwargs.get('pk')
        if not pk:
            return Response('请选择删除的 id')
        data = models.Category.objects.filter(id=pk).delete()
        if data[0] == 0:
            return Response('删除--无此数据')
        return Response('删除成功')

可以通过继承rest_framework.generics的类来实现上述功能

class NewCategorySerializer(serializers.ModelSerializer):
    """
    序列化
    """
    class Meta:
        model = models.Category
        # fields = "__all__"
        fields = ['id', 'name']

from rest_framework.generics import ListAPIView, RetrieveAPIView, CreateAPIView, UpdateAPIView, DestroyAPIView
class NewCategoryView(ListAPIView, RetrieveAPIView, CreateAPIView, UpdateAPIView, DestroyAPIView):

    queryset = models.Category.objects.all()
    serializer_class = NewCategorySerializer

drf与django小结

实现对文章的增删改查

部分字段更新

添加返回的字段（choices）

from rest_framework import serializers
from api import models


class ArticleSerializer(serializers.ModelSerializer):
    """
    序列化
    """
    # 2.添加一个category分类文本字段
    category_name = serializers.CharField(source='category.name', required=False)

    # 3 get_x1()
    x1 = serializers.SerializerMethodField()
    x2 = serializers.SerializerMethodField()
 
    # ststus 2.变为中文
    status_txt = serializers.CharField(source='get_status_display', required=False)

    class Meta:
        model = models.Article
        fields = ['id', 'title', 'summary', 'content', 'category',
                  'category_name', 'x1', 'status', 'status_txt', 'x2']  # category category_name可覆盖
        # fields = "__all__"
        # depth = 1  # 1. 关联表1级关联，2为2级，但不推荐，有多余数据

    def get_x1(self, obj):
        # obj为Article对象，将category.name返回给x1中
        return obj.category.name

     # ststus 3.变为中文
    def get_x2(self, obj):
        return obj.get_status_display()

class ArticleView(APIView):
    def get(self, request, *args, **kwargs):
        pk = kwargs.get('pk')
        if not pk:
            queryset = models.Article.objects.all()
            ser = serializer.ArticleSerializer(instance=queryset, many=True)
            return Response(ser.data)
        article_object = models.Article.objects.filter(id=pk).first()
        ser = serializer.ArticleSerializer(
            instance=article_object, many=False)  # 单条记录
        return Response(ser.data)

    def post(self, request, *args, **kwargs):
        ser = serializer.ArticleSerializer(data=request.data)
        if ser.is_valid():
            ser.save()
            return Response(ser.data)
        return Response(ser.errors)

    # put为全部字段更新
    def put(self, request, *args, **kwargs):
        pk = kwargs.get('pk')
        article_object = models.Article.objects.filter(id=pk).first()
        ser = serializer.ArticleSerializer(
            instance=article_object, data=request.data)
        if ser.is_valid():
            ser.save()
            return Response(ser.data)
        return Response(ser.errors)

    def patch(self, request, *args, **kwargs):
        pk = kwargs.get('pk')
        article_object = models.Article.objects.filter(id=pk).first()
        ser = serializer.ArticleSerializer(
            instance=article_object, data=request.data, partial=True)
        if ser.is_valid():
            ser.save()
            return Response(ser.data)
        return Response(ser.errors)

    def delete(self, request, *args, **kwargs):
        """
        删除一条分类信息
        """
        pk = kwargs.get('pk')
        if not pk:
            return Response('请选择删除的 id')
        data = models.Article.objects.filter(id=pk).delete()
        if data[0] == 0:
            return Response('删除--无此数据')
        return Response('删除成功')

两种分页

drf PageNumberPagination

drf LimitOffsetPagination

 # 分页
    url(r'^page/article/$', views.PageArticleView.as_view()),

settings.py设置

# 分页
REST_FRAMEWORK = {
    "PAGE_SIZE": 2,
    # 引入类的路径
    "DEFAULT_PAGINATION_CLASS": "rest_framework.pagination.PageNumberPagination"
}

class HulaLimitOffsetPagination(LimitOffsetPagination):
    max_limit = 2


class PageArticleView(APIView):
    def get(self, request, *args, **kwargs):
        queryset = models.Article.objects.all()

        # 方式一：仅数据
        """
        page_object = PageNumberPagination()
        result = page_object.paginate_queryset(queryset,request,self)
        ser = PageArticleSerializer(instance=result,many=True)
        return Response(ser.data)
        """
        # 方式二：数据 + 分页信息
        """
        page_object = PageNumberPagination()
        result = page_object.paginate_queryset(queryset, request, self)
        ser = PageArticleSerializer(instance=result, many=True)
        return page_object.get_paginated_response(ser.data)
        """
        # 方式三：数据 + 部分分页信息, 如分页页数等等
        """
        page_object = PageNumberPagination()
        result = page_object.paginate_queryset(queryset, request, self)
        ser = PageArticleSerializer(instance=result, many=True)
        return Response({'count':page_object.page.paginator.count,'result':ser.data})
        """
        
        # offset=1&limit=2, 从第一条数据后、取两条数据
        page_object = HulaLimitOffsetPagination()
        result = page_object.paginate_queryset(queryset, request, self)
        ser = PageArticleSerializer(instance=result, many=True)
        return Response(ser.data)

drf 内置ListAPIView显示列表

urls.py

    url(r'^page/view/article/$', views.PageViewArticleView.as_view()),

不再继承drf的APIView, 内部会调用分页

from rest_framework.generics import ListAPIView

class PageViewArticleSerializer(serializers.ModelSerializer):
    class Meta:
        model = models.Article
        fields = "__all__"

class PageViewArticleView(ListAPIView):
    queryset = models.Article.objects.all()
    serializer_class = PageViewArticleSerializer

REST_FRAMEWORK = { "PAGE_SIZE":2, "DEFAULT_PAGINATION_CLASS":"rest_framework.pagination.PageNumberPagination" }

1.呼啦圈

1.1 表结构设计

不会经常变化的值放在内存：choices形式，避免跨表性能低。
分表：如果表中列太多/大量内容可以选择水平分表
表自关联
外链图片转存失败,源站可能有防盗链机制,建议将图片保存下来直接上传

from django.db import models

class UserInfo(models.Model):
    """ 用户表 """
    username = models.CharField(verbose_name='用户名',max_length=32)
    password = models.CharField(verbose_name='密码',max_length=64)


class Article(models.Model):
    """ 文章表 """
    category_choices = (
        (1,'咨询'),
        (2,'公司动态'),
        (3,'分享'),
        (4,'答疑'),
        (5,'其他'),
    )
    category = models.IntegerField(verbose_name='分类',choices=category_choices)
    title = models.CharField(verbose_name='标题',max_length=32)
    image = models.CharField(verbose_name='图片路径',max_length=128) # /media/upload/....
    summary = models.CharField(verbose_name='简介',max_length=255)

    comment_count = models.IntegerField(verbose_name='评论数',default=0)
    read_count = models.IntegerField(verbose_name='浏览数',default=0)

    author = models.ForeignKey(verbose_name='作者',to='UserInfo')
    date = models.DateTimeField(verbose_name='创建时间',auto_now_add=True)

class ArticleDetail(models.Model):
    article = models.OneToOneField(verbose_name='文章表',to='Article')
    content = models.TextField(verbose_name='内容')


class Comment(models.Model):
    """ 评论表 """
    article = models.ForeignKey(verbose_name='文章',to='Article')
    content = models.TextField(verbose_name='评论')
    user = models.ForeignKey(verbose_name='评论者',to='UserInfo')
    # 评论自关联
    # parent = models.ForeignKey(verbose_name='回复',to='self', null=True,blank=True)

2. 筛选

案例：在文章列表时候，添加筛选功能。

全部：http://127.0.0.1:8000/hg/article/
筛选：http://127.0.0.1:8000/hg/article/?category=2

class ArticleView(APIView):
    """ 文章视图类 """

    def get(self,request,*args,**kwargs):
        """ 获取文章列表 """
        pk = kwargs.get('pk')
        if not pk:
            condition = {}
            category = request.query_params.get('category')
            if category:
                condition['category'] = category
            queryset = models.Article.objects.filter(**condition).order_by('-date')
            
            pager = PageNumberPagination()
            result = pager.paginate_queryset(queryset,request,self)
            ser = ArticleListSerializer(instance=result,many=True)
            return Response(ser.data)
        article_object = models.Article.objects.filter(id=pk).first()
        ser = PageArticleSerializer(instance=article_object,many=False)
        return Response(ser.data)

drf的组件：内置了筛选的功能

from django.shortcuts import render
from rest_framework.views import APIView
from rest_framework.response import Response
from . import models

from rest_framework.filters import BaseFilterBackend

class MyFilterBackend(BaseFilterBackend):

    def filter_queryset(self, request, queryset, view):
        val = request.query_params.get('cagetory')
        return queryset.filter(category_id=val)
    

class IndexView(APIView):

    def get(self,request,*args,**kwargs):
        # http://www.xx.com/cx/index/
        # models.News.objects.all()

        # http://www.xx.com/cx/index/?category=1
        # models.News.objects.filter(category=1)

        # http://www.xx.com/cx/index/?category=1
        queryset = models.News.objects.all()
        obj = MyFilterBackend()
        result = obj.filter_queryset(request,queryset,self)
        # print(result)
        
        return Response('...')

3. drf视图应用

from rest_framework.generics import GenericAPIView, ListAPIView, CreateAPIView, UpdateAPIView, DestroyAPIView, RetrieveAPIView
from rest_framework.pagination import PageNumberPagination
from rest_framework import serializers
from django.shortcuts import render
from rest_framework.views import APIView
from rest_framework.response import Response
from . import models

from rest_framework.filters import BaseFilterBackend


class MyFilterBackend(BaseFilterBackend):

    def filter_queryset(self, request, queryset, view):
        val = request.query_params.get('cagetory')
        return queryset.filter(category_id=val)


class IndexView(APIView):

    def get(self, request, *args, **kwargs):
        # http://www.xx.com/cx/index/
        # models.News.objects.all()

        # http://www.xx.com/cx/index/?category=1
        # models.News.objects.filter(category=1)

        # http://www.xx.com/cx/index/?category=1
        # queryset = models.News.objects.all()
        # obj = MyFilterBackend()
        # result = obj.filter_queryset(request,queryset,self)
        # print(result)

        return Response('...')


class NewFilterBackend(BaseFilterBackend):

    def filter_queryset(self, request, queryset, view):
        val = request.query_params.get('cagetory')
        return queryset.filter(category_id=val)


class NewSerializers(serializers.ModelSerializer):
    class Meta:
        model = models.News
        fields = "__all__"


"""
class NewsView(GenericAPIView):
    queryset = models.News.objects.all()
    filter_backends = [NewFilterBackend,]
    serializer_class = NewSerializers
    pagination_class = PageNumberPagination
    
    def get(self,request,*args,**kwargs):
        v = self.get_queryset()
        queryset = self.filter_queryset(v)
        data = self.paginate_queryset(queryset)
        ser = self.get_serializer(instance=data,many=True)
        return Response(ser.data)
"""


class NewsView(ListAPIView):
    queryset = models.News.objects.all()
    filter_backends = [NewFilterBackend, ]
    serializer_class = NewSerializers
    pagination_class = PageNumberPagination


# #########################################################
class TagSer(serializers.ModelSerializer):
    class Meta:
        model = models.Tag
        fields = "__all__"


class TagView(ListAPIView, CreateAPIView):
    queryset = models.Tag.objects.all()
    serializer_class = TagSer

    # def get_serializer_class(self):
    #     # self.request
    #     # self.args
    #     # self.kwargs
    #     if self.request.method == 'GET':
    #         return TagSer
    #     elif self.request.method == 'POST':
    #         return OtherTagSer
    # def perform_create(self,serializer):
    #     serializer.save(author=1)


class TagDetailView(RetrieveAPIView, UpdateAPIView, DestroyAPIView):
    queryset = models.Tag.objects.all()
    serializer_class = TagSer

resful规范补充

建议用https代替http
在URL中体现api，添加api标识
- https://www.cnblogs.com/xwgblog/p/11812244.html # 错误
- https://www.cnblogs.com/api/xwgblog/p/11812244.html # 正确
- https://api.cnblogs.com/xwgblog/p/11812244.html # 正确
- 建议：https://www.cnblogs.com/api/…
在URL中要体现版
- https://www.cnblogs.com/api/v1/userinfo/
- https://www.cnblogs.com/api/v2/userinfo/
一般情况下对于api接口，用名词不用动词。 https://www.cnblogs.com/api/v1/userinfo/
如果有条件的话，在URL后面进行传递。 https://www.cnblogs.com/api/v1/userinfo/?page=1&category=2
根据method不同做不同操作 get/post/put/patch/delete

request请求对象封装

API版本

局部版本控制