BoolOr¶

Overview¶

BoolOr is an aggregate function that returns true if at least one of the input boolean values is true, otherwise returns false. It is implemented as a runtime-replaceable aggregate that delegates to the Max function internally, treating true as the maximum boolean value.

Syntax¶

bool_or(column)
SELECT bool_or(col) FROM table_name;

// DataFrame API
df.agg(expr("bool_or(column_name)"))

Arguments¶

Argument	Type	Description
child	Expression	The boolean expression or column to aggregate

Return Type¶

Returns BooleanType - either true, false, or null.

Supported Data Types¶

BooleanType only
Input values are implicitly cast to boolean if needed due to ImplicitCastInputTypes trait

Algorithm¶

Internally replaced with Max(child) aggregate function at runtime
Leverages the fact that true > false in boolean ordering
Max function handles the aggregation logic for finding the maximum boolean value
Null values are ignored during aggregation (standard Max behavior)
Returns null only if all input values are null

Partitioning Behavior¶

How this expression affects partitioning:

Does not preserve partitioning as it's an aggregate function
Requires shuffle for global aggregation across partitions
Partial aggregation can be performed within partitions before shuffle

Edge Cases¶

Null handling: Null values are ignored; function returns false if all non-null values are false
All nulls: Returns null if all input values are null
Empty input: Returns null for empty input sets
Mixed nulls: bool_or(true, null, false) returns true; bool_or(false, null, false) returns false

Code Generation¶

This expression supports Tungsten code generation through its replacement with Max aggregate function, which has codegen support for boolean types.

Examples¶

-- Basic usage
SELECT bool_or(col) FROM VALUES (false), (false), (NULL) AS tab(col);
-- Result: false

-- Mixed true/false
SELECT bool_or(col) FROM VALUES (true), (false), (false) AS tab(col);
-- Result: true

-- All nulls
SELECT bool_or(col) FROM VALUES (NULL), (NULL) AS tab(col);
-- Result: NULL

// DataFrame API usage
import org.apache.spark.sql.functions._
df.agg(expr("bool_or(is_active)")).show()

// Groupby usage
df.groupBy("category").agg(expr("bool_or(has_discount)")).show()