链表查询
join方法允许你指定两个或多个表之间的连接条件，并返回一个新的查询对象，该对象包含了连接后的结果。
内连接
from sqlalchemy import join
使用join函数
query = db.session.query(User, Order).join(Order, User.id == Order.user_id)
results = query.all()
或者使用Query对象的join方法
query = db.session.query(User, Order).join(Order, User.id == Order.user_id)
results = query.all()
遍历结果
for user, order in results:
print(user.username, order.order_id) # 假设Order表有一个order_id字段
左连接
query = db.session.query(User, Order).outerjoin(Order, User.id == Order.user_id)
results = query.all()
遍历结果，注意这里可能有些User没有对应的Order
for user, order in results:
if order is not None:
print(user.username, order.order_id)
else:
print(user.username, "No orders")
通过左连接，查询dialogue_detail和detail_eval表；
conditions = [DialogueDetail.dialog_id == dialog_id,
DialogueDetail.create_by == get_user_id()]
dialog_detail_fields = ["id", "dialog_id", "content", "dialog_role",
"user_source", "create_by", "create_time", "is_delete"]
detail_eval_fields = ["comment_type", "feedback_type", "feedback_content"]
query = (db.session.query(*[getattr(DialogueDetail, f1) for f1 in dialog_detail_fields],
*[getattr(DialogueDetailEvaluation, f2) for f2 in detail_eval_fields]).
join(DialogueDetailEvaluation,
DialogueDetailEvaluation.dialog_detail_id == DialogueDetail.id,
isouter=True).
filter(*conditions).order_by(DialogueDetail.create_time))
tmp_all = query.all()多表连接
假设还有第三个表Product，Order表有一个product_id字段指向Product表的id字段
query = db.session.query(User, Order, Product).join(Order, User.id ==Order.user_id).join(Product, Order.product_id == Product.id)
results = query.all()
遍历结果
for user, order, product in results:
print(user.username, order.order_id, product.name) # 假设Product表有一个name字段
参考：