Exactly One Not Null Check

Exactly One Not Null Check#

Check: exactly-one-not-null-check

Purpose: Checks whether exactly one of the specified columns is non-null in each row. A row fails the check if none or more than one of the columns are populated. This is useful for enforcing mutual exclusivity — for example, when exactly one of email, phone, or user_id must be present.

Python Configuration#

from sparkdq.checks import ExactlyOneNotNullCheckConfig
from sparkdq.core import Severity

ExactlyOneNotNullCheckConfig(
    check_id="must_provide_exactly_one_id",
    columns=["email", "phone", "user_id"],
    severity=Severity.CRITICAL
)

Declarative Configuration#

- check: exactly-one-not-null-check
  check-id: must_provide_exactly_one_id
  columns:
    - email
    - phone
    - user_id
  severity: critical

Typical Use Cases#

  • ✅ Enforce mutual exclusivity across optional identifiers (e.g. email or phone or user_id).

  • ✅ Prevent ambiguous data entries where multiple conflicting fields are filled.

  • ✅ Detect rows missing essential identification if all relevant fields are null.

Contents