源出处：点击传送门
filter_by 与filter的区别：
1. filter_by只能取值= filter可以==，!=,>=,<=等多种取值方式
2. filter带参数必须表名.字段名 User.username
#带条件查询
# print(session.query(User).filter_by(username='111').all())
# row = session.query(User).filter(User.username=='dandan').all()
# print(row)
结果
[<User(id='14' ,username='111',password='111',createtime='2018-03-09 20:29:18',_locked ='False',)>]
[<User(id='1' ,username='dandan',password='123asd',createtime='2018-03-06 22:52:51',_locked ='False',)>]
如何查看sqlalchemey的sql语句
去掉all()就可以看到
row = session.query(User).filter(User.username!='aaa')
print(row)
结果
SELECT user.id AS user_id, user.username AS user_username, user.password AS user_password, user.createtime AS user_createtime, user._locked AS user__locked
FROM user
WHERE user.username != %(username_1)s
1
表中单个字段查询
print(session.query(User.username).filter(User.username!='dandan').all())
结果
[('tobee',), ('tree',), ('aaa',), ('coding',), ('choupi',), ('111',), ('choupidan',), ('youku',)]
first()
仅查询显示第一个
print(session.query(User.username).filter(User.username!='dandan').first())
只查询出来第一个，有且只有一个
如果有两个符合条件的，会报错
print(session.query(User).filter(User.username=='choupi').one())
结果
<User(id='12' ,username='choupi',password='q1',createtime='2018-03-09 20:27:54',_locked ='False',)>
get 根据主键查询
主键在表中只有一个
例如ID为主键，查询id=3的元素
print(session.query(User).get(3))<User(id='3' ,username='tree',password='zzz111',createtime='2018-03-09 14:44:21',_locked ='False',)>
limit 限制查询结果
limit(3) 仅查出3条结果
print(session.query(User).filter(User.username!='dandan').limit(3).all())[<User(id='2' ,username='tobee',password='234qwe',createtime='2018-03-07 16:16:05',_locked ='False',)>, <User(id='3' ,username='tree',password='zzz111',createtime='2018-03-09 14:44:21',_locked ='False',)>, <User(id='4' ,username='aaa',password='111',createtime='2018-03-09 17:57:15',_locked ='False',)>]
offset() 限制前面n个，显示后面n+1个
显示第三个以后
print(session.query(User.username).filter(User.username!='dandan').all())
print(session.query(User.username).filter(User.username!='dandan').limit(3).all())
print(session.query(User.username).filter(User.username!='dandan').offset(3).all())
结果：
[('tobee',), ('tree',), ('aaa',), ('coding',), ('choupi',), ('111',), ('choupidan',), ('youku',)]
[('tobee',), ('tree',), ('aaa',)]
[('coding',), ('choupi',), ('111',), ('choupidan',), ('youku',)]
slice() 切片
slice(1,3) 与python的slice一致，从0开始 左闭右开，显示1,2两个元素
print(session.query(User.username).filter(User.username!='dandan').slice(1,3).all())[('tree',), ('aaa',)]
order_by() 元素排序 顺序
print(session.query(User.username).filter(User.username!='dandan').order_by(User.username).all())
按数字字符顺序排序
[('111',), ('aaa',), ('choupi',), ('choupidan',), ('coding',), ('tobee',), ('tree',), ('youku',)]
desc() 逆序排序
from sqlalchemy import desc
print(session.query(User.username).filter(User.username!='dandan').order_by(desc(User.username)).all())
逆序排序
[('youku',), ('tree',), ('tobee',), ('coding',), ('choupidan',), ('choupi',), ('aaa',), ('111',)]
like()
结合%使用 与原生sql一致
print(session.query(User.username).filter(User.username.like('%e')).all())[('tobee',), ('tree',)]
notlike()
print(session.query(User.username).filter(User.username.notlike('%e')).all())[('dandan',), ('aaa',), ('coding',), ('choupi',), ('111',), ('choupidan',), ('youku',)]
in_()
print(session.query(User.username).filter(User.username.in_(['dandan','aaa'])).all())print(session.query(User.username).filter(User.username.in_(['dandan','aaa'])).all())
notin_
print(session.query(User.username).filter(User.username.notin_(['dandan','aaa'])).all())[('tobee',), ('tree',), ('coding',), ('choupi',), ('111',), ('choupidan',), ('youku',)]
is_
两种表达方式 None
print(session.query(User.username).filter(User.username==None).all())
print(session.query(User.username).filter(User.username.is_(None)).all())
isnot
filter支持多条件查询
print(session.query(User.username).filter(User.username.isnot(None),User.password=='111').all())[('aaa',), ('111',)]
or_
from sqlalchemy import or_
print(session.query(User.username).filter(or_(User.username.isnot(None),User.password=='111')).all())
[('dandan',), ('tobee',), ('tree',), ('aaa',), ('coding',), ('choupi',), ('111',), ('choupidan',), ('youku',)]
聚合函数
count group_by
from sqlalchemy import func
print(session.query(User.password,func.count(User.id)).group_by(User.password).all())
查询原生sql
SELECT user.password AS user_password, count(user.id) AS count_1
FROM user GROUP BY user.password[('111', 2), ('123asd', 1), ('234qwe', 1), ('333', 1), ('choupidan', 1), ('q1', 1), ('qwer', 1), ('zzz111', 1)]
having的用法
having字句可以让我们筛选成组后的各种数据，where字句在聚合前先筛选记录，也就是说作用在group by和having字句前。
而having子句在聚合后对组记录进行筛选。真实表中没有此数据，这些数据是通过一些函数生存。
print(session.query(User.password,func.count(User.id)).group_by(User.password).\
having(func.count(User.id)>1).all())[('111', 2)]
sum
print(session.query(User.password,func.sum(User.id)).group_by(User.password).all())
[(‘111’, Decimal(‘18’)), (‘123asd’, Decimal(‘1’)), (‘234qwe’, Decimal(‘2’)), (‘333’, Decimal(‘16’)), (‘choupidan’, Decimal(‘15’)), (‘q1’, Decimal(‘12’)), (‘qwer’, Decimal(‘5’)), (‘zzz111’, Decimal(‘3’))]
max
print(session.query(User.password,func.max(User.id)).group_by(User.password).all())
[(‘111’, 14), (‘123asd’, 1), (‘234qwe’, 2), (‘333’, 16), (‘choupidan’, 15), (‘q1’, 12), (‘qwer’, 5), (‘zzz111’, 3)]
min
print(session.query(User.password,func.min(User.id)).group_by(User.password).all())
[(‘111’, 4), (‘123asd’, 1), (‘234qwe’, 2), (‘333’, 16), (‘choupidan’, 15), (‘q1’, 12), (‘qwer’, 5), (‘zzz111’, 3)]
lable 别名
lable别名不能用在having中
extract
提取时间元素
from sqlalchemy import extract
print(session.query(extract('minute',User.createtime).label('minute'),func.count(User.id)).group_by('minute').all())
分钟排序
[(16, 1), (27, 1), (29, 3), (44, 1), (52, 1), (57, 2)]
天数排序
print(session.query(extract('day',User.createtime).label('day'),func.count(User.id)).group_by('day').all())
[(6, 1), (7, 1), (9, 7)]