Built-in Validators Reference¶

Complete parameter reference for all 264 built-in validators organized by category.

Table of Contents¶

Schema Validators (14)
Completeness Validators (12)
Uniqueness Validators (17)
Distribution Validators (15)
String Validators (20)
Datetime Validators (10)
Aggregate Validators (8)
Cross-Table Validators (4)
Query Validators (19)
Multi-Column Validators (20)
Table Validators (17)
Geospatial Validators (9)
Drift Validators (14)
Anomaly Validators (15)
Referential Validators (13)
Time Series Validators (14)
Business Rule Validators (8)
Profiling Validators (7)
Localization Validators (9)
ML Feature Validators (5)
Privacy Validators (15)

1. Schema Validators¶

Import: from truthound.validators.schema import *

ColumnExistsValidator¶

Validates the presence of specified columns.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Column names that must exist

validator = ColumnExistsValidator(columns=["id", "name", "email"])

ColumnNotExistsValidator¶

Ensures specified columns are absent.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Column names that must not exist

ColumnCountValidator¶

Validates column count.

Parameter	Type	Required	Default	Description
`exact_count`	`int \\| None`	No	`None`	Exact expected count
`min_count`	`int \\| None`	No	`None`	Minimum count
`max_count`	`int \\| None`	No	`None`	Maximum count

validator = ColumnCountValidator(min_count=5, max_count=20)

RowCountValidator¶

Validates row count.

Parameter	Type	Required	Default	Description
`exact_count`	`int \\| None`	No	`None`	Exact expected count
`min_count`	`int \\| None`	No	`None`	Minimum count
`max_count`	`int \\| None`	No	`None`	Maximum count

ColumnTypeValidator¶

Validates column data types.

Parameter	Type	Required	Default	Description
`expected_types`	`dict[str, str \\| type[pl.DataType]]`	Yes	-	Column to type mapping

Supported type aliases: "int", "uint", "float", "numeric", "string", "bool", "date", "datetime", "time".

# Using Polars types directly
validator = ColumnTypeValidator(expected_types={
    "id": pl.Int64,
    "name": pl.String,
    "price": pl.Float64,
})

# Using type aliases
validator = ColumnTypeValidator(expected_types={
    "id": "int",           # Matches Int8, Int16, Int32, Int64
    "amount": "numeric",   # Matches any int, uint, or float
    "email": "string",     # Matches String, Utf8
})

ColumnOrderValidator¶

Validates column ordering.

Parameter	Type	Required	Default	Description
`expected_order`	`list[str]`	Yes	-	Expected column order
`strict`	`bool`	No	`False`	No extra columns allowed

TableSchemaValidator¶

Validates complete schema.

Parameter	Type	Required	Default	Description
`schema`	`dict[str, type]`	Yes	-	Column to type mapping
`strict`	`bool`	No	`False`	Reject extra columns

ColumnPairValidator¶

Validates column pair relationships.

Parameter	Type	Required	Default	Description
`column_a`	`str`	Yes	-	First column
`column_b`	`str`	Yes	-	Second column
`relationship`	`str`	Yes	-	Relationship type

MultiColumnUniqueValidator¶

Validates composite key uniqueness.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Columns forming composite key

ReferentialIntegrityValidator¶

Validates foreign key references.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Foreign key column
`reference_data`	`pl.DataFrame`	Yes	-	Reference table
`reference_column`	`str`	Yes	-	Primary key in reference

validator = ReferentialIntegrityValidator(
    column="department_id",
    reference_data=departments_df,
    reference_column="id"
)

MultiColumnSumValidator¶

Validates column sum equals target.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Columns to sum
`expected_sum`	`float`	Yes	-	Expected sum
`tolerance`	`float`	No	`0.0`	Acceptable tolerance

MultiColumnCalculationValidator¶

Validates arithmetic relationships.

Parameter	Type	Required	Default	Description
`expression`	`str`	Yes	-	Math expression
`tolerance`	`float`	No	`0.0`	Acceptable tolerance

ColumnPairInSetValidator¶

Validates column pairs in allowed set.

Parameter	Type	Required	Default	Description
`column_a`	`str`	Yes	-	First column
`column_b`	`str`	Yes	-	Second column
`valid_pairs`	`set[tuple]`	Yes	-	Valid (a, b) pairs

ColumnPairNotInSetValidator¶

Validates column pairs not in forbidden set.

Parameter	Type	Required	Default	Description
`column_a`	`str`	Yes	-	First column
`column_b`	`str`	Yes	-	Second column
`invalid_pairs`	`set[tuple]`	Yes	-	Forbidden (a, b) pairs

2. Completeness Validators¶

Import: from truthound.validators.completeness import *

NullValidator¶

Detects null values.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target columns (all if None)

NotNullValidator¶

Ensures no null values.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target columns (all if None)

# Single column
validator = NotNullValidator(columns=["email"])

# Multiple columns
validator = NotNullValidator(columns=["user_id", "username", "email"])

# All columns (default)
validator = NotNullValidator()

CompletenessRatioValidator¶

Validates minimum completeness.

Parameter	Type	Required	Default	Description
`min_ratio`	`float`	No	`0.95`	Min completeness (0.0-1.0)
`columns`	`list[str] \\| None`	No	`None`	Target columns (all if None)

# Single column with custom ratio
validator = CompletenessRatioValidator(min_ratio=0.95, columns=["phone"])

# Multiple columns
validator = CompletenessRatioValidator(min_ratio=0.90, columns=["email", "phone"])

EmptyStringValidator¶

Detects empty strings.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target columns

WhitespaceOnlyValidator¶

Detects whitespace-only values.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target columns

ConditionalNullValidator¶

Validates nulls based on condition.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`condition`	`pl.Expr`	Yes	-	Polars condition expression

# If status is "active", email must not be null
validator = ConditionalNullValidator(
    column="email",
    condition=pl.col("status") == "active"
)

DefaultValueValidator¶

Detects default/placeholder values.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`default_values`	`list`	Yes	-	Values considered defaults

NaNValidator¶

Detects NaN values in float columns.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target float columns
`mostly`	`float \\| None`	No	`None`	Min non-NaN ratio

NotNaNValidator¶

Ensures no NaN values.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

NaNRatioValidator¶

Validates maximum NaN ratio.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`max_ratio`	`float`	Yes	-	Max NaN ratio (0.0-1.0)

InfinityValidator¶

Detects infinite values.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target numeric columns

FiniteValidator¶

Ensures all values are finite.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

3. Uniqueness Validators¶

Import: from truthound.validators.uniqueness import *

UniqueValidator¶

Ensures column values are unique.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target columns (None = all)

UniqueRatioValidator¶

Validates unique value ratio.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_ratio`	`float`	No	`0.0`	Min unique ratio
`max_ratio`	`float`	No	`1.0`	Max unique ratio

DistinctCountValidator¶

Validates distinct value count.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_count`	`int \\| None`	No	`None`	Min distinct count
`max_count`	`int \\| None`	No	`None`	Max distinct count

DuplicateValidator¶

Detects duplicates.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Columns to check

DuplicateWithinGroupValidator¶

Detects duplicates within groups.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Column to check
`group_by`	`list[str]`	Yes	-	Grouping columns

PrimaryKeyValidator¶

Validates primary key (unique + non-null).

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Primary key column

CompoundKeyValidator¶

Validates composite primary key.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Columns forming compound key

DistinctValuesInSetValidator¶

All distinct values must be in set.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`value_set`	`set`	Yes	-	Allowed values

DistinctValuesEqualSetValidator¶

Distinct values must exactly match set.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`value_set`	`set`	Yes	-	Expected values

DistinctValuesContainSetValidator¶

Distinct values must contain set.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`value_set`	`set`	Yes	-	Required values

DistinctCountBetweenValidator¶

Distinct count within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_count`	`int`	Yes	-	Min count
`max_count`	`int`	Yes	-	Max count

UniqueWithinRecordValidator¶

Values unique within each row.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Columns to compare

AllColumnsUniqueWithinRecordValidator¶

All values unique within each row.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Columns to check (all if None)

ColumnPairUniqueValidator¶

Column pair uniqueness.

Parameter	Type	Required	Default	Description
`column_a`	`str`	Yes	-	First column
`column_b`	`str`	Yes	-	Second column

ApproximateDistinctCountValidator¶

HyperLogLog-based distinct count.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_count`	`int \\| None`	No	`None`	Min distinct count
`max_count`	`int \\| None`	No	`None`	Max distinct count
`precision`	`int`	No	`14`	HLL precision bits

ApproximateUniqueRatioValidator¶

Approximate unique ratio.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_ratio`	`float`	No	`0.0`	Min unique ratio
`max_ratio`	`float`	No	`1.0`	Max unique ratio

4. Distribution Validators¶

Import: from truthound.validators.distribution import *

BetweenValidator¶

Values within specified range.

Parameter	Type	Required	Default	Description
`min_value`	`float \\| None`	No	`None`	Minimum value
`max_value`	`float \\| None`	No	`None`	Maximum value
`inclusive`	`bool`	No	`True`	Include boundaries

RangeValidator¶

Auto-detects expected ranges based on column names.

Parameter	Type	Required	Default	Description
(no parameters)	-	-	-	Uses built-in KNOWN_RANGES mapping

Auto-detected ranges: - age: [0, 150] - percentage/percent/pct/rate/score: [0, 100] - rating: [0, 5] - year: [1900, 2100] - month: [1, 12], day: [1, 31] - hour: [0, 23], minute/second: [0, 59] - price/quantity/count/amount: [0, None]

PositiveValidator¶

All values > 0. Automatically applies to all numeric columns.

Parameter	Type	Required	Default	Description
(no parameters)	-	-	-	Uses `NumericValidatorMixin` to auto-detect numeric columns

# Validates all numeric columns have positive values
validator = PositiveValidator()
issues = validator.validate(lf)

NonNegativeValidator¶

All values >= 0. Automatically applies to all numeric columns.

Parameter	Type	Required	Default	Description
(no parameters)	-	-	-	Uses `NumericValidatorMixin` to auto-detect numeric columns

# Validates all numeric columns have non-negative values
validator = NonNegativeValidator()
issues = validator.validate(lf)

InSetValidator¶

Values must be in allowed set. Applies to all columns (or specified via columns kwarg).

Parameter	Type	Required	Default	Description
`allowed_values`	`list[Any]`	Yes	-	List of allowed values

validator = InSetValidator(allowed_values=["active", "inactive", "pending"])
issues = validator.validate(lf)

NotInSetValidator¶

Values must not be in forbidden set. Applies to all columns (or specified via columns kwarg).

Parameter	Type	Required	Default	Description
`forbidden_values`	`list[Any]`	Yes	-	List of forbidden values

validator = NotInSetValidator(forbidden_values=["N/A", "unknown", ""])
issues = validator.validate(lf)

IncreasingValidator¶

Values monotonically increasing. Automatically applies to all numeric columns.

Parameter	Type	Required	Default	Description
`strict`	`bool`	No	`False`	Strictly increasing (no equal values)

# Check for monotonic increase across all numeric columns
validator = IncreasingValidator(strict=True)
issues = validator.validate(lf)

DecreasingValidator¶

Values monotonically decreasing. Automatically applies to all numeric columns.

Parameter	Type	Required	Default	Description
`strict`	`bool`	No	`False`	Strictly decreasing (no equal values)

validator = DecreasingValidator()
issues = validator.validate(lf)

OutlierValidator¶

IQR-based outlier detection. Automatically applies to all numeric columns.

Parameter	Type	Required	Default	Description
`iqr_multiplier`	`float`	No	`1.5`	IQR multiplier for bounds

validator = OutlierValidator(iqr_multiplier=2.0)
issues = validator.validate(lf)

Formula: Bounds = [Q1 - k×IQR, Q3 + k×IQR]

ZScoreOutlierValidator¶

Z-score outlier detection. Automatically applies to all numeric columns.

Parameter	Type	Required	Default	Description
`threshold`	`float`	No	`3.0`	Z-score threshold for outlier detection

validator = ZScoreOutlierValidator(threshold=2.5)
issues = validator.validate(lf)

QuantileValidator¶

Quantile bounds validation.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`quantile`	`float`	Yes	-	Quantile (0.0-1.0)
`max_value`	`float`	Yes	-	Max value at quantile

DistributionValidator¶

Distribution shape validation.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`distribution`	`str`	Yes	-	Expected distribution

KLDivergenceValidator¶

KL divergence threshold.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`reference_distribution`	`dict`	Yes	-	Reference distribution
`threshold`	`float`	Yes	-	Max KL divergence

Formula: D_KL(P||Q) = Σ P(x) × log(P(x) / Q(x))

ChiSquareValidator¶

Chi-square goodness of fit.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`expected_frequencies`	`dict`	Yes	-	Expected frequencies
`alpha`	`float`	No	`0.05`	Significance level

MostCommonValueValidator¶

Most common value validation.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`expected_value`	`Any`	Yes	-	Expected most common value
`min_frequency`	`float`	No	`0.0`	Min frequency ratio

5. String Validators¶

Import: from truthound.validators.string import *

RegexValidator¶

Regex pattern matching.

Parameter	Type	Required	Default	Description
`pattern`	`str`	Yes	-	Regex pattern
`columns`	`list[str] \\| None`	No	`None`	Target columns
`match_full`	`bool`	No	`True`	Match entire string
`case_insensitive`	`bool`	No	`False`	Ignore case
`mostly`	`float \\| None`	No	`None`	Acceptable match ratio

Note: RegexValidator requires a pattern parameter and is not in BUILTIN_VALIDATORS.

validator = RegexValidator(pattern=r"^[A-Z]{3}-\d{4}$")

RegexListValidator¶

Multiple patterns (any match passes).

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`patterns`	`list[str]`	Yes	-	List of patterns

NotMatchRegexValidator¶

Must not match pattern.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`pattern`	`str`	Yes	-	Forbidden pattern

NotMatchRegexListValidator¶

Must not match any pattern.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`patterns`	`list[str]`	Yes	-	Forbidden patterns

LengthValidator¶

String length constraints.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_length`	`int \\| None`	No	`None`	Min length
`max_length`	`int \\| None`	No	`None`	Max length

EmailValidator¶

RFC 5322 email format.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

UrlValidator¶

URL format validation.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

PhoneValidator¶

Phone number format.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`country_code`	`str \\| None`	No	`None`	Expected country code

UuidValidator¶

UUID format (v1-v5).

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

IpAddressValidator¶

IPv4 address format.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`version`	`int \\| None`	No	`None`	IP version (4 or 6)

Ipv6AddressValidator¶

IPv6 address format.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

FormatValidator¶

Auto-detect format by column name.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target columns

Auto-detected formats: - email - email, e-mail, mail - phone - phone, tel, mobile, cell - url - url, link, website - uuid - uuid, guid - ip - ip, ip_address - date - date, dob, birth - code - product_code, sku, barcode

JsonParseableValidator¶

Valid JSON strings.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

JsonSchemaValidator¶

JSON Schema validation.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`schema`	`dict`	Yes	-	JSON Schema spec

AlphanumericValidator¶

Alphanumeric characters only.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

ConsistentCasingValidator¶

Consistent case style.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`casing`	`str`	Yes	-	Expected style (snake_case, camelCase, etc.)

LikePatternValidator¶

SQL LIKE pattern match.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`pattern`	`str`	Yes	-	LIKE pattern

validator = LikePatternValidator(column="code", pattern="PRD-%")

NotLikePatternValidator¶

SQL LIKE pattern exclusion.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`pattern`	`str`	Yes	-	Forbidden LIKE pattern

6. Datetime Validators¶

Import: from truthound.validators.datetime import *

DateFormatValidator¶

Date format validation.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`format`	`str`	Yes	-	strptime format

DateBetweenValidator¶

Date within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_date`	`date \\| datetime`	Yes	-	Min date
`max_date`	`date \\| datetime`	Yes	-	Max date

FutureDateValidator¶

Date must be in future.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

PastDateValidator¶

Date must be in past.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

DateOrderValidator¶

Start date < end date.

Parameter	Type	Required	Default	Description
`start_column`	`str`	Yes	-	Start date column
`end_column`	`str`	Yes	-	End date column

validator = DateOrderValidator(start_column="start_date", end_column="end_date")

TimezoneValidator¶

Timezone-aware validation.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`expected_timezone`	`str \\| None`	No	`None`	Expected timezone

RecentDataValidator¶

Data within max age.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Datetime column
`max_age_days`	`int`	Yes	-	Max age in days

DatePartCoverageValidator¶

Coverage across date parts.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Datetime column
`date_part`	`str`	Yes	-	Part to check (day, month, hour)
`min_coverage`	`float`	Yes	-	Min coverage ratio

GroupedRecentDataValidator¶

Recency within groups.

Parameter	Type	Required	Default	Description
`datetime_column`	`str`	Yes	-	Datetime column
`group_column`	`str`	Yes	-	Grouping column
`max_age_days`	`int`	Yes	-	Max age per group

DateutilParseableValidator¶

Parseable by dateutil.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column

7. Aggregate Validators¶

Import: from truthound.validators.aggregate import *

MeanBetweenValidator¶

Column mean within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_value`	`float`	Yes	-	Min mean
`max_value`	`float`	Yes	-	Max mean

MedianBetweenValidator¶

Column median within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_value`	`float`	Yes	-	Min median
`max_value`	`float`	Yes	-	Max median

StdBetweenValidator¶

Standard deviation within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_value`	`float`	Yes	-	Min std
`max_value`	`float`	Yes	-	Max std

VarianceBetweenValidator¶

Variance within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_value`	`float`	Yes	-	Min variance
`max_value`	`float`	Yes	-	Max variance

MinBetweenValidator¶

Column minimum within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_value`	`float`	Yes	-	Min of min
`max_value`	`float`	Yes	-	Max of min

MaxBetweenValidator¶

Column maximum within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_value`	`float`	Yes	-	Min of max
`max_value`	`float`	Yes	-	Max of max

SumBetweenValidator¶

Column sum within range.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`min_value`	`float`	Yes	-	Min sum
`max_value`	`float`	Yes	-	Max sum

TypeValidator¶

Aggregate-level type validation.

Parameter	Type	Required	Default	Description
`columns`	`list[str] \\| None`	No	`None`	Target columns

8. Cross-Table Validators¶

Import: from truthound.validators.cross_table import *

CrossTableRowCountValidator¶

Compare row counts between tables.

Parameter	Type	Required	Default	Description
`reference_data`	`pl.DataFrame`	Yes	-	Reference table
`operator`	`str`	Yes	-	Comparison operator
`tolerance`	`float`	No	`0.0`	Acceptable tolerance

CrossTableRowCountFactorValidator¶

Row count ratio validation.

Parameter	Type	Required	Default	Description
`reference_data`	`pl.DataFrame`	Yes	-	Reference table
`expected_factor`	`float`	Yes	-	Expected ratio
`tolerance`	`float`	No	`0.0`	Acceptable tolerance

CrossTableAggregateValidator¶

Compare aggregates between tables.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`reference_data`	`pl.DataFrame`	Yes	-	Reference table
`aggregate`	`str`	Yes	-	Aggregate function
`operator`	`str`	Yes	-	Comparison operator

CrossTableDistinctCountValidator¶

Compare distinct counts between tables.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Target column
`reference_data`	`pl.DataFrame`	Yes	-	Reference table

9. Query Validators¶

Import: from truthound.validators.query import *

ExpressionValidator¶

Polars expression validation.

Parameter	Type	Required	Default	Description
`expression`	`pl.Expr`	Yes	-	Polars expression
`description`	`str`	No	`""`	Description

validator = ExpressionValidator(
    expression=pl.col("price") > 0,
    description="Price must be positive"
)

QueryReturnsSingleValueValidator¶

Query returns exactly one value.

Parameter	Type	Required	Default	Description
`query`	`pl.Expr`	Yes	-	Query expression

QueryReturnsNoRowsValidator¶

Query returns no rows.

Parameter	Type	Required	Default	Description
`query`	`pl.Expr`	Yes	-	Filter expression

QueryReturnsRowsValidator¶

Query returns at least one row.

Parameter	Type	Required	Default	Description
`query`	`pl.Expr`	Yes	-	Filter expression

QueryResultMatchesValidator¶

Query result matches expected.

Parameter	Type	Required	Default	Description
`query`	`pl.Expr`	Yes	-	Query expression
`expected`	`Any`	Yes	-	Expected value

QueryRowCountValidator¶

Query result row count.

Parameter	Type	Required	Default	Description
`filter_expr`	`pl.Expr`	Yes	-	Filter expression
`expected_count`	`int \\| None`	No	`None`	Expected count
`min_count`	`int \\| None`	No	`None`	Min count
`max_count`	`int \\| None`	No	`None`	Max count

QueryRowCountRatioValidator¶

Ratio of matching rows.

Parameter	Type	Required	Default	Description
`filter_expr`	`pl.Expr`	Yes	-	Filter expression
`min_ratio`	`float`	No	`0.0`	Min ratio
`max_ratio`	`float`	No	`1.0`	Max ratio

QueryRowCountCompareValidator¶

Compare row counts between queries.

Parameter	Type	Required	Default	Description
`query_a`	`pl.Expr`	Yes	-	First query
`query_b`	`pl.Expr`	Yes	-	Second query
`operator`	`str`	Yes	-	Comparison operator

QueryColumnValuesValidator¶

Validate column values in query results.

Parameter	Type	Required	Default	Description
`query`	`pl.Expr`	Yes	-	Query expression
`column`	`str`	Yes	-	Column to validate
`expected_values`	`set`	Yes	-	Expected values

QueryColumnUniqueValidator¶

Query column uniqueness.

Parameter	Type	Required	Default	Description
`query`	`pl.Expr`	Yes	-	Query expression
`column`	`str`	Yes	-	Column to check

QueryColumnNotNullValidator¶

Query column non-null.

Parameter	Type	Required	Default	Description
`query`	`pl.Expr`	Yes	-	Query expression
`column`	`str`	Yes	-	Column to check

QueryAggregateValidator¶

Query aggregate validation.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Column
`aggregate`	`str`	Yes	-	Aggregate function
`expected_value`	`float`	Yes	-	Expected value
`tolerance`	`float`	No	`0.0`	Tolerance

QueryGroupAggregateValidator¶

Group aggregate validation.

Parameter	Type	Required	Default	Description
`group_by`	`list[str]`	Yes	-	Group columns
`column`	`str`	Yes	-	Aggregate column
`aggregate`	`str`	Yes	-	Aggregate function

QueryAggregateCompareValidator¶

Compare aggregates between queries.

Parameter	Type	Required	Default	Description
`column`	`str`	Yes	-	Column
`aggregate`	`str`	Yes	-	Aggregate function
`reference_data`	`pl.DataFrame`	Yes	-	Reference data

CustomExpressionValidator¶

Custom expression strings.

Parameter	Type	Required	Default	Description
`expression_str`	`str`	Yes	-	Expression string

ConditionalExpressionValidator¶

Conditional expressions.

Parameter	Type	Required	Default	Description
`condition`	`pl.Expr`	Yes	-	Condition
`then_expr`	`pl.Expr`	Yes	-	Expression if true

MultiConditionValidator¶

Multiple conditions.

Parameter	Type	Required	Default	Description
`conditions`	`list[pl.Expr]`	Yes	-	List of conditions
`mode`	`str`	No	`"all"`	all/any

RowLevelValidator¶

Row-level validation.

Parameter	Type	Required	Default	Description
`check_expr`	`pl.Expr`	Yes	-	Check expression

10. Multi-Column Validators¶

Import: from truthound.validators.multi_column import *

ColumnSumValidator¶

Columns sum to target.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Columns to sum
`target_column`	`str`	Yes	-	Expected sum column
`tolerance`	`float`	No	`0.0`	Tolerance

validator = ColumnSumValidator(
    columns=["subtotal", "tax", "shipping"],
    target_column="total",
    tolerance=0.01
)

ColumnProductValidator¶

Column multiplication.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Columns to multiply
`target_column`	`str`	Yes	-	Expected product column
`tolerance`	`float`	No	`0.0`	Tolerance

ColumnDifferenceValidator¶

Column difference validation.

Parameter	Type	Required	Default	Description
`column_a`	`str`	Yes	-	First column
`column_b`	`str`	Yes	-	Second column
`expected_difference`	`float`	Yes	-	Expected difference

ColumnRatioValidator¶

Column ratio validation.

Parameter	Type	Required	Default	Description
`numerator_column`	`str`	Yes	-	Numerator
`denominator_column`	`str`	Yes	-	Denominator
`expected_ratio`	`float`	Yes	-	Expected ratio
`tolerance`	`float`	No	`0.0`	Tolerance

ColumnPercentageValidator¶

Percentage validation.

Parameter	Type	Required	Default	Description
`part_column`	`str`	Yes	-	Part column
`whole_column`	`str`	Yes	-	Whole column
`min_percentage`	`float`	No	`0.0`	Min percentage
`max_percentage`	`float`	No	`100.0`	Max percentage

ColumnComparisonValidator¶

Column comparison.

Parameter	Type	Required	Default	Description
`column_a`	`str`	Yes	-	First column
`column_b`	`str`	Yes	-	Second column
`operator`	`str`	Yes	-	Operator (>, <, >=, <=, ==, !=)

validator = ColumnComparisonValidator(
    column_a="end_date",
    column_b="start_date",
    operator=">"
)

ColumnChainComparisonValidator¶

Ordered column chain.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Columns in order
`operator`	`str`	Yes	-	Comparison operator

ColumnMutualExclusivityValidator¶

At most one non-null per row.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Mutually exclusive columns

ColumnCoexistenceValidator¶

All null or all non-null.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Coexisting columns

ColumnDependencyValidator¶

Functional dependency.

Parameter	Type	Required	Default	Description
`determinant_column`	`str`	Yes	-	Determinant
`dependent_column`	`str`	Yes	-	Dependent

ColumnImplicationValidator¶

If A then B.

Parameter	Type	Required	Default	Description
`antecedent`	`pl.Expr`	Yes	-	Condition
`consequent`	`pl.Expr`	Yes	-	Result

ColumnCorrelationValidator¶

Correlation validation.

Parameter	Type	Required	Default	Description
`column_a`	`str`	Yes	-	First column
`column_b`	`str`	Yes	-	Second column
`min_correlation`	`float`	No	`-1.0`	Min correlation
`max_correlation`	`float`	No	`1.0`	Max correlation

ColumnCovarianceValidator¶

Covariance validation.

Parameter	Type	Required	Default	Description
`column_a`	`str`	Yes	-	First column
`column_b`	`str`	Yes	-	Second column
`expected_covariance`	`float`	Yes	-	Expected covariance
`tolerance`	`float`	No	`0.0`	Tolerance

MultiColumnVarianceValidator¶

Variance across columns.

Parameter	Type	Required	Default	Description
`columns`	`list[str]`	Yes	-	Columns to analyze
`max_variance`	`float`	Yes	-	Max variance

11-21. Remaining Categories¶

For detailed parameter references for:

Table Validators - See Categories Reference
Geospatial Validators - See Categories Reference
Drift Validators - See Categories Reference
Anomaly Validators - See Categories Reference
Referential Validators - See Categories Reference
Time Series Validators - See Categories Reference
Business Rule Validators - See Categories Reference
Profiling Validators - See Categories Reference
Localization Validators - See Categories Reference
ML Feature Validators - See Categories Reference
Privacy Validators - See Categories Reference

Common Parameters¶

These parameters are shared across many validators:

Parameter	Type	Default	Description
`column`	`str`	-	Target column name (legacy, prefer `columns`)
`columns`	`list[str]`	`None`	Multiple columns (None = all)
`mostly`	`float`	`None`	Min pass ratio (0.0-1.0)
`max_anomaly_ratio`	`float`	`0.1`	Max anomaly ratio
`reference_data`	`pl.DataFrame`	-	Baseline dataset for drift
`tolerance`	`float`	`0.0`	Acceptable numeric tolerance

Note: Most validators support both column (single) and columns (multiple) parameters via **kwargs. The columns parameter is preferred for consistency and flexibility. When columns=None, validators typically apply to all applicable columns.

Severity Levels¶

Severity	Description	Typical Condition
`LOW`	Minor issue	< 1% affected
`MEDIUM`	Moderate concern	1-5% affected
`HIGH`	Significant issue	5-10% affected
`CRITICAL`	Severe problem	> 10% affected

Installation Options¶

# Core validators only
pip install truthound

# With drift detection (scipy)
pip install truthound[drift]

# With anomaly detection (scipy, scikit-learn)
pip install truthound[anomaly]

# All dependencies
pip install truthound[all]

Next Steps¶

Categories Reference - Category descriptions and submodules
Custom Validators - Build your own validators
Security Features - ReDoS and SQL injection protection