Flask SQLAlchemy query, specify column names

2020-01-23 06:15发布

How do I specify the column that I want in my query using a model (it selects all columns by default)? I know how to do this with the sqlalchmey session: session.query(self.col1), but how do I do it with with models? I can't do SomeModel.query(). Is there a way?

6条回答
SAY GOODBYE
2楼-- · 2020-01-23 06:53

You can use Query.values, Query.values

session.query(SomeModel).values('id', 'user')

查看更多
Anthone
3楼-- · 2020-01-23 06:59
session.query().with_entities(SomeModel.col1)

is the same as

session.query(SomeModel.col1)

for alias, we can use .label()

session.query(SomeModel.col1.label('some alias name'))
查看更多
We Are One
4楼-- · 2020-01-23 06:59

An example here:

movies = Movie.query.filter(Movie.rating != 0).order_by(desc(Movie.rating)).all()

I query the db for movies with rating <> 0, and then I order them by rating with the higest rating first.

Take a look here: Select, Insert, Delete in Flask-SQLAlchemy

查看更多
三岁会撩人
5楼-- · 2020-01-23 07:04

You can use Model.query, because the Model (or usually its base class, especially in cases where declarative extension is used) is assigned Sesssion.query_property. In this case the Model.query is equivalent to Session.query(Model).

I am not aware of the way to modify the columns returned by the query (except by adding more using add_columns()).
So your best shot is to use the Session.query(Model.col1, Model.col2, ...) (as already shown by Salil).

查看更多
爱情/是我丢掉的垃圾
6楼-- · 2020-01-23 07:06

You can use the with_entities() method to restrict which columns you'd like to return in the result. (documentation)

result = SomeModel.query.with_entities(SomeModel.col1, SomeModel.col2)

Depending on your requirements, you may also find deferreds useful. They allow you to return the full object but restrict the columns that come over the wire.

查看更多
Root(大扎)
7楼-- · 2020-01-23 07:08

You can use load_only function:

from sqlalchemy.orm import load_only

fields = ['name', 'addr', 'phone', 'url']
companies = session.query(SomeModel).options(load_only(*fields)).all()
查看更多
登录 后发表回答