SQLALCHEMY query on a many to many relationship with map table - mysql

Question

I have a database model (see below). I'm working on a project that should filter query results based on user assigned categories - there can be several categories associated to a user. In other words, I want to show the User with Articles that matches his/her assigned categories. Articles can also be associated with several categories. How can I achieve this using sqlalchemy query?
categories_users = db.Table(
'categories_users',
db.Column('user_id', db.Integer(), db.ForeignKey('user.id')),
db.Column('category_id', db.Integer(), db.ForeignKey('category.id'))
)
categories_articles = db.Table(
'categories_articles',
db.Column('article_id', db.Integer(), db.ForeignKey('article.id')),
db.Column('category_id', db.Integer(), db.ForeignKey('category.id'))
)
class Category(db.Model):
id = db.Column(db.Integer(), primary_key=True)
name = db.Column(db.String(80), unique=True)
articles = db.relationship('Article', secondary=categories_articles,
backref=db.backref('categories', lazy='dynamic'))
def __str__(self):
return self.name
class Article(db.Model):
id = db.Column(db.Integer, primary_key=True)
name = db.Column(db.String(255))
def __str__(self):
return self.name
class User(db.Model):
id = db.Column(db.Integer, primary_key=True)
email = db.Column(db.String(255), unique=True, nullable=False)
categories = db.relationship('Category', secondary=categories_users,
backref=db.backref('users', lazy='dynamic'))
def __str__(self):
return self.email

Answer 1

In a normal m2m you have an association table with join path between the two entities that is clearly defined by foreign keys. As such, you can just tell sqlalchemy which table you want to use to associate two other tables, and it does the rest (As you've done here: secondary=categories_articles). Joining User's to Articles is slightly more complicated.
You want a many to many relationship between User and Article, so you can get the articles that users are interested in based on common categories. Or simply:
User <===> some join path involving categories <===> Article.
Our user table has a foreign key to the categories_users table and our categories_articles table has a foreign key to our articles table. Most importantly, both the categories_users table and the categories_articles table have the category_id field and so we can join those two tables together.
So, you can add an articles relationship to your User model with the following code:
articles = db.relationship(
'Article',
secondary=\
'join(categories_users, categories_articles, '
'categories_users.c.category_id==categories_articles.c.category_id)'
)
Accessing the User.articles attribute then emits this SQL:
SELECT article.id AS article_id, article.name AS article_name
FROM article, categories_users
INNER JOIN categories_articles
ON categories_users.category_id = categories_articles.category_id
WHERE %(param_1)s = categories_users.user_id
AND article.id = categories_articles.article_id
where it's clear to see what is happening. categories_users gets joined with categories_articles effectively making a new table where user rows are combined with article rows on their common categories. SQLAlchemy can then treat this as a "normal" many to many relationship with a single association table and leverage the existing foreign keys to piece together the rest of the relationship and construct the necessary query.