详解Django的model查询操作与查询性能优化

1 如何 在做orm查询时 查看sql的执行情况

(1) 最底层的 django.db.connection

在 django shell 中使用  python manage.py shell

>>> from django.db import connection
>>> books.objects.all()
>>> connection.queries  ## 可以查看查询时间
[{'sql': 'select "testsql_books"."id", "testsql_books"."name", "testsql_books"."author_id" from "testsql_books" limi
t 21', 'time': '0.002'}]

(2) django-extensions 插件 

pip install django-extensions

 installed_apps = (
    ...
    'django_extensions',
    ...
    )

在 django shell 中使用  python manage.py shell_plus  --print-sql （extensions 强化）

这样每次查询都会 有sql 输出

>>> from testsql.models import books
>>> books.objects.all()
  select "testsql_books"."id", "testsql_books"."name", "testsql_books"."author_id" from "testsql_books" limit 21

execution time: 0.002000s [database: default]

<queryset [<books: books object>, <books: books object>, <books: books object>]>

2 orm查询操作 以及优化

基本操作

增

models.tb1.objects.create(c1='xx', c2='oo') 增加一条数据，可以接受字典类型数据 **kwargs

obj = models.tb1(c1='xx', c2='oo')
obj.save()

 查

models.tb1.objects.get(id=123)     # 获取单条数据，不存在则报错（不建议）
models.tb1.objects.all()        # 获取全部
models.tb1.objects.filter(name='seven') # 获取指定条件的数据
models.tb1.objects.exclude(name='seven') # 获取指定条件的数据

 删

models.tb1.objects.filter(name='seven').delete() # 删除指定条件的数据

 改
models.tb1.objects.filter(name='seven').update(gender='0') # 将指定条件的数据更新，均支持 **kwargs
obj = models.tb1.objects.get(id=1)
obj.c1 = '111'
obj.save()                         # 修改单条数据

查询简单操作

获取个数

  models.tb1.objects.filter(name='seven').count()

大于，小于

  models.tb1.objects.filter(id__gt=1)       # 获取id大于1的值
  models.tb1.objects.filter(id__gte=1)       # 获取id大于等于1的值
  models.tb1.objects.filter(id__lt=10)       # 获取id小于10的值
  models.tb1.objects.filter(id__lte=10)       # 获取id小于10的值
  models.tb1.objects.filter(id__lt=10, id__gt=1)  # 获取id大于1 且 小于10的值

in

  models.tb1.objects.filter(id__in=[11, 22, 33])  # 获取id等于11、22、33的数据
  models.tb1.objects.exclude(id__in=[11, 22, 33]) # not in

isnull
  entry.objects.filter(pub_date__isnull=true)

contains

  models.tb1.objects.filter(name__contains="ven")
  models.tb1.objects.filter(name__icontains="ven") # icontains大小写不敏感
  models.tb1.objects.exclude(name__icontains="ven")

range

  models.tb1.objects.filter(id__range=[1, 2])  # 范围bettwen and

其他类似

  startswith，istartswith, endswith, iendswith,

order by

  models.tb1.objects.filter(name='seven').order_by('id')  # asc
  models.tb1.objects.filter(name='seven').order_by('-id')  # desc

group by--annotate

  from django.db.models import count, min, max, sum
  models.tb1.objects.filter(c1=1).values('id').annotate(c=count('num'))
  select "app01_tb1"."id", count("app01_tb1"."num") as "c" from "app01_tb1" where "app01_tb1"."c1" = 1 group by "app01_tb1"."id"

limit 、offset

  models.tb1.objects.all()[10:20]

regex正则匹配，iregex 不区分大小写

  entry.objects.get(title__regex=r'^(an?|the) +')
  entry.objects.get(title__iregex=r'^(an?|the) +')

date

  entry.objects.filter(pub_date__date=datetime.date(2005, 1, 1))
  entry.objects.filter(pub_date__date__gt=datetime.date(2005, 1, 1))

year

  entry.objects.filter(pub_date__year=2005)
  entry.objects.filter(pub_date__year__gte=2005)

month

  entry.objects.filter(pub_date__month=12)
  entry.objects.filter(pub_date__month__gte=6)

day

  entry.objects.filter(pub_date__day=3)
  entry.objects.filter(pub_date__day__gte=3)

week_day

  entry.objects.filter(pub_date__week_day=2)
  entry.objects.filter(pub_date__week_day__gte=2)

hour

  event.objects.filter(timestamp__hour=23)
  event.objects.filter(time__hour=5)
  event.objects.filter(timestamp__hour__gte=12)

minute

  event.objects.filter(timestamp__minute=29)
  event.objects.filter(time__minute=46)
  event.objects.filter(timestamp__minute__gte=29)

second

  event.objects.filter(timestamp__second=31)
  event.objects.filter(time__second=2)
  event.objects.filter(timestamp__second__gte=31)

查询复杂操作

fk foreign key 使用的原因:

• 约束
• 节省硬盘

但是多表查询会降低速度，大型程序反而不使用外键，而是用单表(约束的时候，通过代码判断)

extra

  extra(self, select=none, where=none, params=none, tables=none, order_by=none, select_params=none)
    entry.objects.extra(select={'new_id': "select col from sometable where othercol > %s"}, select_params=(1,))
    entry.objects.extra(where=['headline=%s'], params=['lennon'])
    entry.objects.extra(where=["foo='a' or bar = 'a'", "baz = 'a'"])
    entry.objects.extra(select={'new_id': "select id from tb where id > %s"}, select_params=(1,), order_by=['-nid'])

f

  from django.db.models import f
  models.tb1.objects.update(num=f('num')+1)

q

  方式一：
  q(nid__gt=10)
  q(nid=8) | q(nid__gt=10)
  q(q(nid=8) | q(nid__gt=10)) & q(caption='root')

  方式二：
  con = q()
  q1 = q()
  q1.connector = 'or'
  q1.children.append(('id', 1))
  q1.children.append(('id', 10))
  q1.children.append(('id', 9))
  q2 = q()
  q2.connector = 'or'
  q2.children.append(('c1', 1))
  q2.children.append(('c1', 10))
  q2.children.append(('c1', 9))
  con.add(q1, 'and')
  con.add(q2, 'and')

  models.tb1.objects.filter(con)

exclude(self, *args, **kwargs)

  # 条件查询
  # 条件可以是：参数，字典，q

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。
   model.tb.objects.all().select_related()
   model.tb.objects.all().select_related('外键字段')
   model.tb.objects.all().select_related('外键字段__外键字段')

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 在内存中做关联，而不会再做连表查询
      # 第一次 获取所有用户表
      # 第二次 获取用户类型表where id in (用户表中的查到的所有用户id)
      models.userinfo.objects.prefetch_related('外键字段')

annotate(self, *args, **kwargs)

# 用于实现聚合group by查询

  from django.db.models import count, avg, max, min, sum

  v = models.userinfo.objects.values('u_id').annotate(uid=count('u_id'))
  # select u_id, count(ui) as `uid` from userinfo group by u_id

  v = models.userinfo.objects.values('u_id').annotate(uid=count('u_id')).filter(uid__gt=1)
  # select u_id, count(ui_id) as `uid` from userinfo group by u_id having count(u_id) > 1

  v = models.userinfo.objects.values('u_id').annotate(uid=count('u_id',distinct=true)).filter(uid__gt=1)
  # select u_id, count( distinct ui_id) as `uid` from userinfo group by u_id having count(u_id) > 1

extra(self, select=none, where=none, params=none, tables=none, order_by=none, select_params=none)

 # 构造额外的查询条件或者映射，如：子查询

    entry.objects.extra(select={'new_id': "select col from sometable where othercol > %s"}, select_params=(1,))
    entry.objects.extra(where=['headline=%s'], params=['lennon'])
    entry.objects.extra(where=["foo='a' or bar = 'a'", "baz = 'a'"])
    entry.objects.extra(select={'new_id': "select id from tb where id > %s"}, select_params=(1,), order_by=['-nid'])

reverse(self):

# 倒序
models.userinfo.objects.all().order_by('-nid').reverse()
# 注：如果存在order_by，reverse则是倒序，如果多个排序则一一倒序

下面两个 取到的是对象，并且注意 取到的对象可以 获取其他字段（这样会再去查找该字段降低性能
defer(self, *fields):

 models.userinfo.objects.defer('username','id')
或
models.userinfo.objects.filter(...).defer('username','id')
# 映射中排除某列数据

only(self, *fields):

# 仅取某个表中的数据
models.userinfo.objects.only('username','id')
或
models.userinfo.objects.filter(...).only('username','id')

执行原生sql

1.connection
from django.db import connection, connections
cursor = connection.cursor() 
# cursor = connections['default'].cursor()
django的settings中的db配置 ' default'，指定数据库
cursor.execute("""select * from auth_user where id = %s""", [1])
row = cursor.fetchone()

2 .extra
entry.objects.extra(select={'new_id': "select id from tb where id > %s"}, select_params=(1,), order_by=['-nid'])

3 . raw     
name_map = {'a':'a','b':'b'}
models.userinfo.objects.raw('select * from xxxx',translations=name_map)

以上就是本文的全部内容，希望对大家的学习有所帮助，也希望大家多多支持。