Comparison API Reference

qualink.comparison — Low-level cross-table comparison utilities.

These classes are used internally by the cross-table constraints but can also be used standalone.

from qualink.comparison.referential_integrity import ReferentialIntegrity, ReferentialIntegrityResult
from qualink.comparison.row_count_match import RowCountMatch, RowCountMatchResult
from qualink.comparison.schema_match import SchemaMatch, SchemaMatchResult

`qualink.comparison.referential_integrity`

@dataclass(frozen=True)

class ReferentialIntegrityResult

Field	Type	Default
`match_ratio`	`float`	required
`unmatched_count`	`int`	required
`total_count`	`int`	required

Properties:

is_valid: bool

class ReferentialIntegrity(LoggingMixin)

Checks that all values in *child_table.child_column* exist in *parent_table.parent_column*.

Executes a LEFT ANTI JOIN via DataFusion SQL.

ReferentialIntegrity(child_table: str, child_column: str, parent_table: str, parent_column: str)

Param	Type	Default
`child_table`	`str`	—
`child_column`	`str`	—
`parent_table`	`str`	—
`parent_column`	`str`	—

async run(ctx: SessionContext) → ReferentialIntegrityResult

`qualink.comparison.row_count_match`

Row count match check between two DataFusion tables.

@dataclass(frozen=True)

class RowCountMatchResult

Field	Type	Default
`count_a`	`int`	required
`count_b`	`int`	required
`ratio`	`float`	required

Properties:

is_match: bool

class RowCountMatch(LoggingMixin)

Compares row counts of two tables via DataFusion.

RowCountMatch(table_a: str, table_b: str)

Param	Type	Default
`table_a`	`str`	—
`table_b`	`str`	—

async run(ctx: SessionContext) → RowCountMatchResult

`qualink.comparison.schema_match`

Schema match check between two DataFusion tables.

@dataclass(frozen=True)

class SchemaMatchResult

Field	Type	Default
`matching_columns`	`list[str]`	`field(default_factory=list)`
`only_in_a`	`list[str]`	`field(default_factory=list)`
`only_in_b`	`list[str]`	`field(default_factory=list)`
`type_mismatches`	`dict[str, tuple[str, str]]`	`field(default_factory=dict)`

Properties:

is_match: bool

class SchemaMatch(LoggingMixin)

Compares schemas of two tables via DataFusion.

SchemaMatch(table_a: str, table_b: str)

Param	Type	Default
`table_a`	`str`	—
`table_b`	`str`	—

async run(ctx: SessionContext) → SchemaMatchResult