从raw sql到flask-sqlalchemy

Question

hello im trying to achieve this raw query with sqlalchemy:

SELECT m.*, SUM(case when f.monkey = m.id then 1 else 0 end) as friends
FROM monkey as  m
LEFT JOIN friendship as f ON  m.id = f.monkey
GROUP BY m.id,  m.name
order by friends desc

so far i get the result i want with this raw query but i want to be able to .paginate them so i could keep working properly

with my other queries what i did was this:

monkeys = models.Monkey.query.order_by(models.Monkey.name).paginate(page, 5, False)

fairly simple and i got what i wanted, i belive i have to do something like

monkeys = models.Monkey.query.join(models.Friendship, db.func.count(models.Monkey.id == models.Friendship.monkey))

but im not getting what i want, i know im missing the sum() part but i tried with func.c.count()but i just dont know how to get it work, is it posible to achieve this in sqlalchemy? im using postgres btw

Answer 1

Looks like this will accomplish what you need

monkeys = models.Monkey.query.outerjoin(models.Friendship, db.func.count(models.Monkey.id == models.Friendship.monkey))

#or if you really want to stick to your above query

monkeys = models.Monkey.query.outerjoin(models.Friendship, db.func.SUM(func.IF(models.Monkey.id == models.Friendship.monkey, 1, 0)))

FYI I'm used to doing this in mysql, the func.* you call my be slightly different