Has One

In a one-to-one relationship, two feature sets have at most one feature set on both sides of the relation. One-to-one relationships are specified via the function has_one, where the first argument specifies a function returning how to join the tables.

​

Example

In the example below, users have an associated profile:

from chalk.features import has_one, ...

@features
class Profile:
    ...
    user_id: str
    email_age_years: float

@features
class User:
    ...
    uid: str
    profile: Profile = has_one(lambda: Profile.user_id == User.uid)

Now, you can reference features on Profile through User. For example:

user_email_age = User.profile.email_age_years

​

Back-references

​

One-to-one

You can also add the back-reference to User from Profile. You don’t have to explicitly use has_one again on Profile, however. Instead, the join condition is assumed to be symmetric and copied over. Building on the above example, all you need to do to complete the one-to-one relationship is to add a User to the Profile class:

@features
class Profile:
  ...
  user_id: str
  email_age_years: float
  user: "User"

@features
class User:
  ...
  uid: str
  profile: Profile = has_one(lambda: Profile.user_id == User.uid)

​

Optional relationships

You can define optional relationships simply by using the typing.Optional[...] keyword:

from typing import Optional

@features
class User:
  ...
  uid: str
  profile: Profile = has_one(lambda: Profile.user_id == User.uid)
  profile: Optional[Profile] = has_one(lambda: Profile.user_id == User.uid)