django ORM操作

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

进阶操作

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
如果需要 in 很多连续值的话用range
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, null 和 空不是一个概念
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, 区别于 python range 的是 orm 的 range 的左右都是闭合标签
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

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)

其他操作

##################################################################
# PUBLIC METHODS THAT ALTER ATTRIBUTES AND RETURN A NEW QUERYSET #
##################################################################

def all(self)
    # 获取所有的数据对象

def filter(self, *args, **kwargs)
    # 条件查询
    # 条件可以是:参数,字典,Q

def exclude(self, *args, **kwargs)
    # 条件查询
    # 条件可以是:参数,字典,Q

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。
     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')

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。
            # 获取所有用户表
            # 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')



            from django.db.models import Count, Case, When, IntegerField
            Article.objects.annotate(
                numviews=Count(Case(
                    When(readership__what_time__lt=treshold, then=1),
                    output_field=CharField(),
                ))
            )

            students = Student.objects.all().annotate(num_excused_absences=models.Sum(
                models.Case(
                    models.When(absence__type='Excused', then=1),
                default=0,
                output_field=models.IntegerField()
            )))

def 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  先distinct再count

def distinct(self, *field_names)
    # 用于distinct去重
    models.UserInfo.objects.values('nid').distinct()
    # select distinct nid from userinfo

    注:只有在PostgreSQL中才能使用distinct进行去重

def order_by(self, *field_names)
    # 用于排序
    models.UserInfo.objects.all().order_by('-id','age')

def 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'])
应对的情况是select nid 1 from tb;  select nid, (select name from tb2 where tb2.id=tb.id) from tb where ... order by ...
按平均成绩从低到高 显示所有学生的“语文”、“数学”、“英语”三门的课程成绩,按如下形式显示: 学生ID,语文,数学,英语,有效课程数,有效平均分;
    select sc.student_id,
        (select num from score left join course on score.course_id = course.cid where course.cname = "生物" and score.student_id=sc.student_id) as sy,
        (select num from score left join course on score.course_id = course.cid where course.cname = "物理" and score.student_id=sc.student_id) as wl,
        (select num from score left join course on score.course_id = course.cid where course.cname = "体育" and score.student_id=sc.student_id) as ty,
        count(sc.course_id),
        avg(sc.num)
    from score as sc
    group by student_id desc   

 def 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


 def 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

 def 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')

 def using(self, alias):
     指定使用的数据库,参数为别名(setting中的设置)


##################################################
# PUBLIC METHODS THAT RETURN A QUERYSET SUBCLASS #
##################################################

def raw(self, raw_query, params=None, translations=None, using=None):
    # 执行原生SQL
    models.UserInfo.objects.raw('select * from userinfo')

    # 如果SQL是其他表时,必须将名字设置为当前UserInfo对象的主键列名
    models.UserInfo.objects.raw('select id as nid from 其他表')

    # 为原生SQL设置参数
    models.UserInfo.objects.raw('select id as nid from userinfo where nid>%s', params=[12,])

    # 将获取的到列名转换为指定列名
    name_map = {'first': 'first_name', 'last': 'last_name', 'bd': 'birth_date', 'pk': 'id'}
    Person.objects.raw('SELECT * FROM some_other_table', translations=name_map)

    # 指定数据库
    models.UserInfo.objects.raw('select * from userinfo', using="default")

    ################### 原生SQL ###################
    from django.db import connection, connections
    cursor = connection.cursor()  # cursor = connections['default'].cursor()
    cursor.execute("""SELECT * from auth_user where id = %s""", [1])
    row = cursor.fetchone() # fetchall()/fetchmany(..)


def values(self, *fields):
    # 获取每行数据为字典格式

def values_list(self, *fields, **kwargs):
    # 获取每行数据为元祖

def dates(self, field_name, kind, order='ASC'):
    # 根据时间进行某一部分进行去重查找并截取指定内容
    # kind只能是:"year"(年), "month"(年-月), "day"(年-月-日)
    # order只能是:"ASC"  "DESC"
    # 并获取转换后的时间
        - year : 年-01-01
        - month: 年-月-01
        - day  : 年-月-日

    models.DatePlus.objects.dates('ctime','day','DESC')

def datetimes(self, field_name, kind, order='ASC', tzinfo=None):
    # 根据时间进行某一部分进行去重查找并截取指定内容,将时间转换为指定时区时间
    # kind只能是 "year", "month", "day", "hour", "minute", "second"
    # order只能是:"ASC"  "DESC"
    # tzinfo时区对象
    models.DDD.objects.datetimes('ctime','hour',tzinfo=pytz.UTC)
    models.DDD.objects.datetimes('ctime','hour',tzinfo=pytz.timezone('Asia/Shanghai'))

    """
    pip3 install pytz
    import pytz
    pytz.all_timezones
    pytz.timezone(‘Asia/Shanghai’)
    """

def none(self):
    # 空QuerySet对象


####################################
# METHODS THAT DO DATABASE QUERIES #
####################################

def aggregate(self, *args, **kwargs):
   # 聚合函数,获取字典类型聚合结果
   from django.db.models import Count, Avg, Max, Min, Sum
   result = models.UserInfo.objects.aggregate(k=Count('u_id', distinct=True), n=Count('nid'))
   ===> {'k': 3, 'n': 4}

def count(self):
   # 获取个数

def get(self, *args, **kwargs):
   # 获取单个对象

def create(self, **kwargs):
   # 创建对象

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)

def get_or_create(self, defaults=None, **kwargs):
    # 如果存在,则获取,否则,创建
    # defaults 指定创建时,其他字段的值, created是True或者False,create只能create一条
    obj, created = models.UserInfo.objects.get_or_create(username='root1',password='123456', defaults={'email': '1111111','u_id': 2, 't_id': 2})

def update_or_create(self, defaults=None, **kwargs):
    # 如果存在,则更新,否则,创建
    # defaults 指定创建时或更新时的其他字段
    obj, created = models.UserInfo.objects.update_or_create(username='root1', username='root1',password='123456',defaults={'email': '1111111','u_id': 2, 't_id': 1})

def first(self):
   # 获取第一个

def last(self):
   # 获取最后一个

def in_bulk(self, id_list=None):
   # 根据主键ID进行查找
   id_list = [11,21,31]
   models.DDD.objects.in_bulk(id_list)

def delete(self):
   # 删除

def update(self, **kwargs):
    # 更新

def exists(self):
   # 是否有结果

select_related 和 prefetch_related 的用法

数据库


index.html

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <title>Title</title>
</head>
<body>
    {% for item in v %}
        {{ item.name }}{{ item.email }}{{ item.user_type.name }}
    {% endfor %}
</body>
</html>

view.py

情况一

v = models.User.objects.all()
return render(request,'index.html',{'v': v})

发现执行了四次sql语句查询

SELECT "app01_user"."id", "app01_user"."name", "app01_user"."email", "app01_user"."user_type_id" FROM "app01_user";
SELECT "app01_usertype"."id", "app01_usertype"."name" FROM "app01_usertype" WHERE "app01_usertype"."id" = '1'
SELECT "app01_usertype"."id", "app01_usertype"."name" FROM "app01_usertype" WHERE "app01_usertype"."id" = '1'
SELECT "app01_usertype"."id", "app01_usertype"."name" FROM "app01_usertype" WHERE "app01_usertype"."id" = '1'

情况二

v = models.User.objects.all().select_related('user_type')
return render(request,'index.html',{'v': v})

发现只执行了一次sql查询

SELECT "app01_user"."id", "app01_user"."name", "app01_user"."email", "app01_user"."user_type_id",
"app01_usertype"."id", "app01_usertype"."name" FROM "app01_user" 
INNER JOIN "app01_usertype" ON ("app01_user"."user_type_id" = "app01_usertype"."id")

情况三

v = models.User.objects.all().prefetch_related('user_type')
return render(request,'index.html',{'v': v})

发现执行了两次sql语句

SELECT "app01_user"."id", "app01_user"."name", "app01_user"."email", "app01_user"."user_type_id" FROM "app01_user"
SELECT "app01_usertype"."id", "app01_usertype"."name" FROM "app01_usertype" WHERE "app01_usertype"."id" IN ('1')

django会帮我们做整合

ORM函数操作

# ########### 基础函数 ###########
    from django.db.models.functions import *
    # 1. Concat,用于做类型转换
    # v = models.UserInfo.objects.annotate(c=Cast('pwd', FloatField()))

    # 2. Coalesce,从前向后,查询第一个不为空的值
    # v = models.UserInfo.objects.annotate(c=Coalesce('name', 'pwd'))
    # v = models.UserInfo.objects.annotate(c=Coalesce(Value('666'),'name', 'pwd'))

    # 3. Concat,拼接
    # models.UserInfo.objects.update(name=Concat('name', 'pwd'))
    # models.UserInfo.objects.update(name=Concat('name', Value('666')))
    # models.UserInfo.objects.update(name=Concat('name', Value('666'),Value('999')))

    # 4.ConcatPair,拼接(仅两个参数)
    # v = models.UserInfo.objects.annotate(c=ConcatPair('name', 'pwd'))
    # v = models.UserInfo.objects.annotate(c=ConcatPair('name', Value('666')))

    # 5.Greatest,获取比较大的值;least 获取比较小的值;
    # v = models.UserInfo.objects.annotate(c=Greatest('id', 'pwd',output_field=FloatField()))

    # 6.Length,获取长度
    # v = models.UserInfo.objects.annotate(c=Length('name'))

    # 7. Lower,Upper,变大小写
    # v = models.UserInfo.objects.annotate(c=Lower('name'))
    # v = models.UserInfo.objects.annotate(c=Upper('name'))

    # 8. Now,获取当前时间
    # v = models.UserInfo.objects.annotate(c=Now())

    # 9. substr,子序列, 序号从1开始不是从0开始,2表示取几个字符串
    # v = models.UserInfo.objects.annotate(c=Substr('name',1,2))

    # ########### 时间类函数 ###########
    # 1. 时间截取,不保留其他:Extract, ExtractDay, ExtractHour, ExtractMinute, ExtractMonth,ExtractSecond, ExtractWeekDay, ExtractYear,
    # v = models.UserInfo.objects.annotate(c=functions.ExtractYear('ctime'))
    # v = models.UserInfo.objects.annotate(c=functions.ExtractMonth('ctime'))
    # v = models.UserInfo.objects.annotate(c=functions.ExtractDay('ctime'))
    #
    # v = models.UserInfo.objects.annotate(c=functions.Extract('ctime', 'year'))
    # v = models.UserInfo.objects.annotate(c=functions.Extract('ctime', 'month'))
    # v = models.UserInfo.objects.annotate(c=functions.Extract('ctime', 'year_month'))
    """
    MICROSECOND
    SECOND
    MINUTE
    HOUR
    DAY
    WEEK
    MONTH
    QUARTER
    YEAR
    SECOND_MICROSECOND
    MINUTE_MICROSECOND
    MINUTE_SECOND
    HOUR_MICROSECOND
    HOUR_SECOND
    HOUR_MINUTE
    DAY_MICROSECOND
    DAY_SECOND
    DAY_MINUTE
    DAY_HOUR
    YEAR_MONTH
    """

    # 2. 时间截图,保留其他:Trunc, TruncDate, TruncDay,TruncHour, TruncMinute, TruncMonth, TruncSecond, TruncYear
    # v = models.UserInfo.objects.annotate(c=functions.TruncHour('ctime'))
    # v = models.UserInfo.objects.annotate(c=functions.TruncDate('ctime'))
    # v = models.UserInfo.objects.annotate(c=functions.Trunc('ctime','year'))
posted @ 2018-07-09 10:31  龙云飞谷  阅读(154)  评论(0编辑  收藏  举报