Welcome to OGeek Q&A Community for programmer and developer-Open, Learning and Share
Welcome To Ask or Share your Answers For Others

Categories

0 votes
275 views
in Technique[技术] by (71.8m points)

python - How to filter by joinloaded table in SqlAlchemy?

Lets say I got 2 models, Document and Person. Document got relationship to Person via "owner" property. Now:

session.query(Document)
    .options(joinedload('owner'))
    .filter(Person.is_deleted!=True)

Will double join table Person. One person table will be selected, and the doubled one will be filtered which is not exactly what I want cuz this way document rows will not be filtered.

What can I do to apply filter on joinloaded table/model ?

See Question&Answers more detail:os

与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…
Welcome To Ask or Share your Answers For Others

1 Reply

0 votes
by (71.8m points)

You are right, table Person will be used twice in the resulting SQL, but each of them serves different purpose:

  • one is to filter the the condition: filter(Person.is_deleted != True)
  • the other is to eager load the relationship: options(joinedload('owner'))

But the reason your query returns wrong results is because your filter condition is not complete. In order to make it produce the right results, you also need to JOIN the two models:

qry = (session.query(Document).
        join(Document.owner). # THIS IS IMPORTANT
        options(joinedload(Document.owner)).
        filter(Person.is_deleted != True)
        )

This will return correct rows, even though it will still have 2 references (JOINs) to Person table. The real solution to your query is that using contains_eager instead of joinedload:

qry = (session.query(Document).
        join(Document.owner). # THIS IS STILL IMPORTANT
        options(contains_eager(Document.owner)).
        filter(Person.is_deleted != True)
        )

与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…
OGeek|极客中国-欢迎来到极客的世界,一个免费开放的程序员编程交流平台!开放,进步,分享!让技术改变生活,让极客改变未来! Welcome to OGeek Q&A Community for programmer and developer-Open, Learning and Share
Click Here to Ask a Question

...