84.经常使用的返回QuerySet对象的方法使用详解：select_related, prefetch_related

1.select_related: 只能用在一对多或者是一对一的关联模型之间，不能用在多对多或者是多对一的关联模型间，好比能够提早获取文章的做者，可是不能经过做者获取做者的文章，或者是经过某篇文章获取这个文章的全部的标签。

好比，想要获取与Book表经过外键的形式相关联的Author表中的数据，示例代码以下：

from django.db import connection
from django.http import HttpResponse
from .models import Article, Book, BookOrder
def index4(request):
    # 1. select_related()想要获取与Book表经过外键的形式相关联的Author表中的数据
    # 能够经过select_related()将相关联的表中的数据提取到内存中，减小到数据库中的查询
    books = Book.objects.select_related('author')
    for book in books:
        print("%s, %s" % (book.author.name, book.author.age))
    print(connection.queries) 
    # 最后django底层能够执行一条sql语句就将全部的查询操做完成

django底层执行的sql语句为：

[{'sql': 'SELECT @@SQL_AUTO_IS_NULL', 'time': '0.000'},
{'sql': 'SET SESSION TRANSACTION ISOLATION LEVEL READ COMMITTED', 'time': '0.000'},
{'sql': 'SELECT book.id, book.name, book.pages, book.price, book.rating, book.author_id, book.publisher_id, book.score, author.id, author.name, author.age, author.email FROM book INNER JOIN author ON (book.author_id = author.id)', 'time': '0.016'}]

由以上的sql语句，能够看出只进行了一次查询操做，因此这就大大提升了查询的效率了。所以若是在数据量不是太大，而且常常会使用到该表的相关联的表中的数据时，就能够采用select_related()方法首先将数据提取到内存中，以便以后在进行查找的时候能够减小向数据库中发起请求的次数。

2. 采用传统的方式进行查找，获取Book表经过外键的形式相关联的Author表中的数据，

示例代码以下：

books = Book.objects.all()
    # 采用all()方法获取每一个book的信息，而后向数据库中提出查询操做每一个book进行一次数据库的查询，提取出name，age字段的值
    for book in books:
        print("%s,%s" % (book.author.name, book.author.age))
    print(connection.queries)

django底层执行的sql语句为：
[{'sql': 'SELECT @@SQL_AUTO_IS_NULL', 'time': '0.000'},
{'sql': 'SET SESSION TRANSACTION ISOLATION LEVEL READ COMMITTED', 'time': '0.000'},
{'sql': 'SELECT book.id, book.name, book.pages, book.price, book.rating, book.author_id, book.publisher_id, book.score FROM book', 'time': '0.000'},
{'sql': 'SELECT author.id, author.name, author.age, author.email FROM author WHERE author.id = 5 LIMIT 21', 'time': '0.000'},
{'sql': 'SELECT author.id, author.name, author.age, author.email FROM author WHERE author.id = 1 LIMIT 21', 'time': '0.000'},
{'sql': 'SELECT author.id, author.name, author.age, author.email FROM author WHERE author.id = 4 LIMIT 21', 'time': '0.000'},
{'sql': 'SELECT author.id, author.name, author.age, author.email FROM author WHERE author.id = 3 LIMIT 21', 'time': '0.000'}]

由以上sql语句能够看出，采用传统的方式执行的sql语句较多，这会大大的下降查询的效率。因此若是数据量不是太大的状况下能够采用select_related()方法将数据提取到内存中。

3. prefetch_related(): 能够对没有多对多，多对一的关联模型执行查询操做。

好比，求每本书的订阅量。示例代码以下：

def index(request):
 # 3. prefetch_related()能够对多对多，多对一的关联模型操做
    # 好比，求每一个书的订阅量
    # 返回的books中的是包裹在QuerySet中的book对象
    books = Book.objects.prefetch_related('bookorder_set')
    print(type(books))
    # <class 'django.db.models.query.QuerySet'>
    for book in books:
        print(type(book))
    # <class 'front.models.Book'>, 以后就能够对返回的每一本书上的属性进行操做
    #   如下咱们能够提取出每本书上的全部订单信息
    # 注意：若是已经使用prefetch_related()进行了相关的查找，这里就不要使用filter()或者是其余的能够返回QuerySet的任何操做了。
    # book_nums = book.bookorder_set.filter(price__gte=80)
    # 这里可使用all()方法进行获取。若是对bookorder再次进行了查询就会增长n个查询语句,下降查询的效率
        book_nums = book.bookorder_set.all()
        for book_num in book_nums:
            print("%s, %s, %s" % (book_num.book.name, book_num.price, book_num.id))
    print(connection.queries)
    return HttpResponse("success!")

django底层执行的sql语句为：

[{'sql': 'SELECT @@SQL_AUTO_IS_NULL', 'time': '0.000'},
{'sql': 'SET SESSION TRANSACTION ISOLATION LEVEL READ COMMITTED', 'time': '0.000'},
{'sql': 'SELECT book.id, book.name, book.pages, book.price, book.rating, book.author_id, book.publisher_id, book.score FROM book', 'time': '0.000'},
{'sql': 'SELECT book_order.id, book_order.book_id, book_order.price, book_order.time FROM book_order WHERE book_order.book_id IN (1, 2, 3, 4)', 'time': '0.000'}]

总结：在多对多，多对一关联对象中执行查询操做，可使用prefetch_related()进行相关查询，这种方式中能够大大减小执行的sql语句，而且这种方式会产生两条查询语句。一样也可使用prefetch_related()对一对多或者是一对一的关联对象进行相关查询，一样也会在底层执行两条sql语句，若是使用select_related()进行查询的话只会执行一条查询语句。因此说若是是对一对一或者是一对多关联对象执行操做的话，建议使用select_related(); 若是是多对一或者是多对多关联模型执行操做时，就能够采用prefetch_related()

4. 采用传统的方式执行与prefetch_related()做用相同的操做，查看django底层执行的sql语句，

示例代码以下：

def index(request):
# 4. 使用传统的方法对多对多或者是多对一的关联模型表进行查询操做
    books = Book.objects.all()
    for book in books:
        print(type(book))
    #     # <class 'front.models.Book'>
        book_nums = book.bookorder_set.all()
        for book_num in book_nums:
            print(type(book_num))
    #         # <class 'front.models.BookOrder'>
            print("%s, %s, %s" % (book_num.book.name, book_num.price, book_num.id))
    print(connection.queries)
    return HttpResponse("success")

django底层执行的sql语句：
[{'sql': 'SELECT @@SQL_AUTO_IS_NULL', 'time': '0.000'},
{'sql': 'SET SESSION TRANSACTION ISOLATION LEVEL READ COMMITTED', 'time': '0.000'},
{'sql': 'SELECT book.id, book.name, book.pages, book.price, book.rating, book.author_id, book.publisher_id, book.score FROM book', 'time': '0.000'},
{'sql': 'SELECT book_order.id, book_order.book_id, book_order.price, book_order.time FROM book_order WHERE book_order.book_id = 1', 'time': '0.000'},
{'sql': 'SELECT book_order.id, book_order.book_id, book_order.price, book_order.time FROM book_order WHERE book_order.book_id = 2', 'time': '0.000'},
{'sql': 'SELECT book_order.id, book_order.book_id, book_order.price, book_order.time FROM book_order WHERE book_order.book_id = 3', 'time': '0.000'},
{'sql': 'SELECT book_order.id, book_order.book_id, book_order.price, book_order.time FROM book_order WHERE book_order.book_id = 4', 'time': '0.016'}]

由以上所执行的sql语句能够看出，采用传统的方式会使底层执行的sql语句的数量增大

5. 若是在某些状况下须要对预先查询的结果，再进行一些操做，可使用Prefetch()函数，

示例代码以下：

def index5(request):
    # 5.若是确实要对预先查询的结果，再进行一些操做的话，可使用Prefetch()函数
    # 能够将要进行的操做赋值给Prefetch函数中的queryset参数
    prefetch = Prefetch('bookorder_set', queryset=BookOrder.objects.filter(price__gte=90))
    books = Book.objects.prefetch_related(prefetch)
    for book in books:
            # <class 'front.models.BookOrder'>
        orders = book.bookorder_set.all()
        for order in orders:
            print("%s, %s, %s" % (order.book.name, order.price, order.id))
    print(connection.queries)
    return HttpResponse("success")

django底层执行的sql语句为：
水浒传, 95.0, 2
[{'sql': 'SELECT @@SQL_AUTO_IS_NULL', 'time': '0.000'},
{'sql': 'SET SESSION TRANSACTION ISOLATION LEVEL READ COMMITTED', 'time': '0.000'},
{'sql': 'SELECT book.id, book.name, book.pages, book.price, book.rating, book.author_id, book.publisher_id, book.score FROM book', 'time': '0.000'},
{'sql': 'SELECT book_order.id, book_order.book_id, book_order.price, book_order.time FROM book_order WHERE (book_order.price >= 90.0e0 AND book_order.book_id IN (1, 2, 3, 4))', 'time': '0.000'}]