ORM分组操作示例（与SQL语句的比较）

class Employee(models.Model):
    name = models.CharField(max_length=16)
    age = models.IntegerField()
    salary = models.IntegerField()
    province = models.CharField(max_length=32)
    dept = models.CharField(max_length=16)

    def __str__(self):
        return self.name

    class Meta:
        db_table = "employee"

  操作：

我们使用原生sql语句，按照部分分组求平均工资：

select dept,AVG(salary) from employee group by dept;

ORM语句与SQL语句对应关系：

ORM查询：

  ret = models.Employee.objects.all()
    print(ret)#, , , ]>
              #(0.003) SELECT `employee`.`id`, `employee`.`name`, `employee`.`age`, `employee`.`salary`, `employee`.`province`, `employee`.`dept` 
　　　　　　　　　FROM `employee` LIMIT 21; args=()

    ret = models.Employee.objects.values("dept")
    print(ret)
    # (0.002)  SELECT `employee`.`dept` FROM `employee` LIMIT  21; args = ()
    # < QuerySet[{‘dept‘: ‘保安部‘}, {‘dept‘: ‘影视部‘}, {‘dept‘: ‘影视部‘}, {‘dept‘: ‘福利部‘}] >

 ret = models.Employee.objects.values("dept").annotate(avg=Avg("salary")).values("dept","avg")
    print(ret)
    #(0.068) SELECT `employee`.`dept`, AVG(`employee`.`salary`) AS `avg` FROM `employee` GROUP BY `employee`.`dept` ORDER BY NULL LIMIT 21; 
    # 

多表操作

建表：

class Employee2(models.Model):
    name = models.CharField(max_length=16)
    age = models.IntegerField()
    salary = models.IntegerField()
    province = models.CharField(max_length=32)
    dept = models.ForeignKey(to="Dept")

    def __str__(self):
        return self.name

    class Meta:
        db_table = "employee2"


class Dept(models.Model):
    name = models.CharField(max_length=16, unique=True)

    def __str__(self):
        return self.name

    class Meta:
        db_table = "dept2"

  SQL查询：

select dept2.name,AVG(salary) from employee2 inner join dept2 on (employee2.dept_id=dept2.id) group by dept_id;

ORM查询：

from Django.db.models import Avg
ret = models.Employee2.objects.values("dept_id").annotate(avg=Avg("salary")).values("dept__name","avg")
print(ret)
# < QuerySet[{‘dept__name‘: ‘保安部‘, ‘avg‘: 2000.0}, {‘dept__name‘: ‘影视部‘, ‘avg‘: 6500.0}, {‘dept__name‘: ‘福利部‘, ‘avg‘: 8000.0}] >
# (0.089) SELECT `dept2`.`name`,AVG(`employee2`.`salary`) AS  `avg` FROM `employee2` INNER JOIN `dept2` ON(`employee2`.`dept_id` = `dept2`.id`) 
GROUP BY `employee2`.`dept_id`,`dept2`.`name` ORDER BY NULL LIMIT 21;args = ()

# 查所有的员工和部门名称
    ret = models.Employee2.objects.values("name", "dept__name")
    print(ret)
    #(0.012) SELECT `employee2`.`name`, `dept2`.`name` FROM `employee2` INNER JOIN `dept2` ON (`employee2`.`dept_id` = `dept2`.`id`) LIMIT 21;
    #    {‘name‘: ‘化工哥‘, ‘dept__name‘: ‘福利部‘}]>

select_related 和 prefetch_related 的使用

def select_related(self, *fields)
    性能相关：表之间进行join连表操作，一次性获取关联的数据。

    总结：
    1. select_related主要针一对一和多对一关系进行优化。
    2. select_related使用SQL的JOIN语句进行优化，通过减少SQL查询的次数来进行优化、提高性能。

def prefetch_related(self, *lookups)
    性能相关：多表连表操作时速度会慢，使用其执行多次SQL查询在python代码中实现连表操作。

    总结：
    1. 对于多对多字段（ManyToManyField）和一对多字段，可以使用prefetch_related()来进行优化。
    2. prefetch_related()的优化方式是分别查询每个表，然后用Python处理他们之间的关系。

select_related的使用示例

 #select_related的使用：表之间进行join连表操作，一次性获取关联的数据。
    ret = models.Employee2.objects.select_related()
    print(ret)
    #(0.019) SELECT `employee2`.`id`, `employee2`.`name`, `employee2`.`age`, `employee2`.`salary`, `employee2`.`province`, `employee2`.`dept_id`,
   `dept2`.`id`, `dept2`.`name` FROM `employee2` INNER JOIN `dept2` ON (`employee2`.`dept_id` = `dept2`.`id`) LIMIT 21; args=()
    #, , , ]>

    ret = models.Employee2.objects.select_related().values("name","dept__name")
    print(ret)
    #(0.020) SELECT `employee2`.`name`, `dept2`.`name` FROM `employee2` INNER JOIN `dept2` ON (`employee2`.`dept_id` = `dept2`.`id`) LIMIT 21; 
    #    {‘name‘: ‘化工哥‘, ‘dept__name‘: ‘福利部‘}]>

  建立多对多关系表：

class Author(models.Model):
    name = models.CharField(max_length=32)
    books = models.ManyToManyField(to="Book")

    def __str__(self):
        return self.name

    class Meta:
        db_table = "author"

class Book(models.Model):
    title = models.CharField(max_length=32)

    def __str__(self):
        return self.title

    class Meta:
        db_table = "book"

 ret = models.Author.objects.select_related("books__title").values("name", "books__title")
    print(ret)
    #(0.014) SELECT `author`.`name`, `book`.`title` FROM `author` LEFT OUTER JOIN `author_books` ON (`author`.`id` = `author_books`.`author_id`)
    LEFT OUTER JOIN `book` ON (`author_books`.`book_id` = `book`.`id`) LIMIT 21; args=()
    #    ‘books__title‘: ‘光子出版社‘}, {‘name‘: ‘小黄‘, ‘books__title‘: ‘光子出版社‘}, {‘name‘: ‘小黑‘, ‘books__title‘: ‘番茄物语‘}, 
    {‘name‘: ‘小白‘, ‘books__title‘: ‘番茄物语‘}, {‘name‘: ‘小黄‘, ‘books__title‘: ‘番茄物语‘}]>

批量操作

def bulk_create(self, objs, batch_size=None):
    # 批量插入
    # batch_size表示一次插入的个数
    objs = [
        models.DDD(name=‘r11‘),
        models.DDD(name=‘r22‘)
    ]
    models.DDD.objects.bulk_create(objs, 10)

示例：

    # 批量创建
    # 有100个书籍对象
    objs = [models.Book(title="沙河{}".format(i)) for i in range(6)]
    #
    # 在数据库中批量创建, 2次一提交
    models.Book.objects.bulk_create(objs, 2)

ORM分组操作示例（与SQL语句的比较）

原文地址：https://www.cnblogs.com/myws9898/p/13769419.html