根据你的问题，我猜你想要：
所有客户都在一个表中，交易在一个单独的表中。为了满足这个问题，我们可以使用客户和交易表之间的内连接（只返回两个表中的那些记录）和过滤方法来过滤作为客户的天数。该方法如下面的示例所示。
from sqlalchemy import Column, Integer, String, DateTime
from sqlalchemy.ext.declarative import declarative_base
from sqlalchemy import create_engine, ForeignKey
from sqlalchemy.orm import sessionmaker, relationship
from random import randint
Base = declarative_base()
## Define sample models
class Customer(Base):
__tablename__ = 'customer'
customer_id = Column('id', Integer, primary_key=True)
nr_days_customer = Column('nr_days_customer', Integer)
name = Column('name', String(250))
transaction = relationship('Transaction', back_populates='customer')
def __init__(self, nr_days_customer, name):
self.nr_days_customer = nr_days_customer
self.name = name
def __repr__(self):
return '[{customer_id}] - {name} - {nr_days_customer} days'.format(customer_id=self.customer_id,
name=self.name,
nr_days_customer=self.nr_days_customer)
class Transaction(Base):
__tablename__ = 'transaction'
transaction_id = Column('id', Integer, primary_key=True)
customer_id = Column('customer_id', Integer, ForeignKey('customer.id'))
name = Column(String(250))
customer = relationship('Customer', back_populates='transaction')
def __init__(self, customer_id, name):
self.customer_id = customer_id
self.name = name
def __repr__(self):
return '[{transaction_id}] {name} - customer: {customer_id}'.format(transaction_id=self.transaction_id,
name=self.name,
customer_id=self.customer_id)
engine = create_engine('sqlite:///')
session = sessionmaker()
session.configure(bind=engine)
ex_ses = session()
Base.metadata.create_all(engine)
## Create sample data
# http://listofrandomnames.com/ for something else then A,B,...
names = ['Virgil', 'Ina', 'Oleta', 'Suzette', 'Usha', 'Ilda', 'Lorean', 'Cinthia', 'Sheba', 'Waneta', 'Donnette',
'Minerva', 'Owen', 'Georgia', 'Gertrudis', 'Sherika', 'Terry', 'Deja', 'Regena', 'Carolin']
# Create customers > 30 days with transaction
for customer_nr in range(5):
ex_ses.add(Customer(name=names[customer_nr].format(customer_nr),
nr_days_customer=randint(30, 120)))
# Create customers < 30 days with transaction
for customer_nr in range(5):
ex_ses.add(Customer(name=names[customer_nr + 5].format(customer_nr),
nr_days_customer=randint(0, 29)))
# Create customers > 30 days without transaction
for customer_nr in range(5):
ex_ses.add(Customer(name=names[customer_nr + 10].format(customer_nr),
nr_days_customer=randint(30, 120)))
# Create customers < 30 days without transaction
for customer_nr in range(5):
ex_ses.add(Customer(name=names[customer_nr + 15].format(customer_nr),
nr_days_customer=randint(0, 29)))
# Create transactions
for transaction_nr in range(10):
ex_ses.add(Transaction(name='Transaction {}'.format(transaction_nr), customer_id=transaction_nr + 1))
ex_ses.commit()
# Queries
print('customers:')
for customer in ex_ses.query(Customer).all():
print('* ', customer)
'''
customers:
* [1] - Virgil - 113 days
* [2] - Ina - 112 days
* [3] - Oleta - 44 days
* [4] - Suzette - 61 days
* [5] - Usha - 82 days
* [6] - Ilda - 10 days
* [7] - Lorean - 14 days
* [8] - Cinthia - 24 days
* [9] - Sheba - 25 days
* [10] - Waneta - 11 days
* [11] - Donnette - 38 days
* [12] - Minerva - 110 days
* [13] - Owen - 98 days
* [14] - Georgia - 68 days
* [15] - Gertrudis - 36 days
* [16] - Sherika - 18 days
* [17] - Terry - 6 days
* [18] - Deja - 7 days
* [19] - Regena - 21 days
* [20] - Carolin - 14 days
'''
print('Transactions:')
for transaction in ex_ses.query(Transaction).all():
print('* ', transaction)
'''
Transactions:
* [1] Transaction 0 - customer: 1
* [2] Transaction 1 - customer: 2
* [3] Transaction 2 - customer: 3
* [4] Transaction 3 - customer: 4
* [5] Transaction 4 - customer: 5
* [6] Transaction 5 - customer: 6
* [7] Transaction 6 - customer: 7
* [8] Transaction 7 - customer: 8
* [9] Transaction 8 - customer: 9
* [10] Transaction 9 - customer: 10
'''
'''
-- using inner join only records in both tables are being returned
-- https://www.w3schools.com/sql/sql_join.asp
-- condition on more then 30 days
-- untested SQL:
SELECT customer.customer_id, transaction.transaction_id, customer.days_a_customer from customer
INNER JOIN transaction ON customer.customer_id = transaction.customer_id
WHERE customer.days_a_customer > 30 AND customer.customer_id.
'''
print('Customer - transactions:')
for cust_trans in ex_ses.query(Customer.name, Customer.nr_days_customer,Transaction.transaction_id).join(Transaction).filter(
Customer.nr_days_customer > 30).all():
print('* Customer {customer_name} - {nr_days_customer} --> transaction {transaction_id}'.format(
customer_name = cust_trans[0],nr_days_customer=cust_trans[1],transaction_id=cust_trans[2]))
'''
Customer - transactions:
* Customer Virgil - 113 --> transaction 1
* Customer Ina - 112 --> transaction 2
* Customer Oleta - 44 --> transaction 3
* Customer Suzette - 61 --> transaction 4
* Customer Usha - 82 --> transaction 5
'''
正如您在注释文本中看到的那样，此查询仅返回 5 个客户超过 30 天且在交易表中的客户。
在 cmets 中提供了指向更多资源的 fez 链接。除此之外，我想强调以下其他资源：
我在上面的例子中没有使用 mysql。我相信提供的示例可以轻松应用于 mysql 设置。
如果有其他问题，请发表评论。