python - 加入条件以急切地加载 sqlalchemy orm

Question

from sqlalchemy.orm import subqueryload, joinedload, eagerload
from sqlalchemy import Column, DateTime, String, Integer, ForeignKey, func,Float, sql
from sqlalchemy.orm import relation
from sqlalchemy.ext.declarative import declarative_base
from sqlalchemy.orm import sessionmaker
from sqlalchemy import create_engine

engine = create_engine('sqlite:///testdb.sqlite')
engine.echo = True
Base = declarative_base()
session = sessionmaker()
session.configure(bind=engine)
Base.metadata.create_all(engine)
s= session()

class Stock(Base):
__tablename__ = 'stock'
stock_id = Column(Integer, primary_key=True)
name = Column(String)
prices = relation("StockPrice")

class StockPrice(Base):
__tablename__ = 'stock_price'
stock_id = Column(Integer, ForeignKey('stock.stock_id'), primary_key=True)
date = Column(String, primary_key=True)
price = Column(Float)
source = Column(String, primary_key=True)
user = Column(String)

Base.metadata.create_all(engine)

stockprice1 = StockPrice(stock_id = 1, date="2014-10-29", price="170.0", source="X Firm", user="U1")
stockprice2 = StockPrice(stock_id = 1, date="2014-10-30", price="175.0", source="X Firm", user="U2")
stock1 = Stock(stock_id = 1, name = "GOOGLE", prices=[stockprice1, stockprice2])

stockprice1 = StockPrice(stock_id = 2, date="2014-10-29", price="150.0", source="X Firm", user="U1")
stockprice2 = StockPrice(stock_id = 2, date="2014-10-30", price="155.0", source="X Firm", user="U2")
stock2 = Stock(stock_id = 2, name = "YAHOO", prices=[stockprice1, stockprice2])

s.add_all([stock1, stock2])
s.commit()

预先加载股票价格:

stock = s.query(Stock).options(joinedload(Stock.prices)).filter(Stock.stock_id == 1).one()

一种快速加载给定日期股票价格的方法:

stock = s.query(Stock).options(joinedload(Stock.prices)).filter(Stock.stock_id == 1).filter(StockPrice.date == "2014-10-30") .one()

但该方法的问题是，如果您有可能与 Stock 相关的 StockPrice 表，并且如果您想加载给定日期的所有关系，那么在加入所有关系后结果集会变得非常庞大。过滤器在 WHERE 子句中添加条件，而我需要一种方法来指定连接条件以预先加载。

stock = s.query(Stock).options(joinedload(Stock.prices, #condition 在这里不起作用)).filter(Stock.stock_id == 1).one()

Answer 1

代替 joinedload(Stock.prices) 执行以下操作:

stock = (s.query(Stock)

         # @note: this replaces `joinedload(Stock.prices)`
         .join(StockPrice,
               and_(StockPrice.stock_id == Stock.stock_id,
                    StockPrice.date == "2014-10-30")
               )

         # effectively *trick* SQLAlchemy into thinking that above we loaded all
         # items for the relationship *Stock.prices*
         .options(contains_eager(Stock.prices))

         ).get(1) # will retrieve the instance for Stock.stock_id = 1