Skip to main content

databricks.labs.dqx.anomaly.segment_utils

Segment naming and filtering for row anomaly detection.

canonicalize_segment_values​

def canonicalize_segment_values(
        segment_values: Mapping[str, Any] | None) -> dict[str, str]

Canonicalize segment values for deterministic naming and filtering.

build_segment_name​

def build_segment_name(segment_values: Mapping[str, Any] | None) -> str

Build deterministic segment name from segment values.

build_segment_filter​

def build_segment_filter(
        segment_values: dict[str, str] | None) -> Column | None

Build Spark filter expression for a segment's values.

Arguments:

  • segment_values - Dictionary mapping segment column names to values

Returns:

Spark Column expression combining all segment filters with AND None if segment_values is None or empty

Example:

>>> build_segment_filter(dict(region="US", product="A")) Column<'((region = US) AND (product = A))'> >>> build_segment_filter(None) None