[Feature] Support Spark expression: make_dt_interval

[andygrove]

What is the problem the feature request solves?

Note: This issue was generated with AI assistance. The specification details have been extracted from Spark documentation and may need verification.

Comet does not currently support the Spark make_dt_interval function, causing queries using this function to fall back to Spark's JVM execution instead of running natively on DataFusion.

The MakeDTInterval expression creates a day-time interval value from separate day, hour, minute, and second components. This expression is used to construct DayTimeIntervalType values programmatically by combining individual time unit values into a single interval representation.

Supporting this expression would allow more Spark workloads to benefit from Comet's native acceleration.

Describe the potential solution

Spark Specification

Syntax:

make_dt_interval(days, hours, minutes, seconds)
make_dt_interval(days, hours, minutes)
make_dt_interval(days, hours)
make_dt_interval(days)
make_dt_interval()

Arguments:

Argument	Type	Description
days	IntegerType	Number of days in the interval (optional, defaults to 0)
hours	IntegerType	Number of hours in the interval (optional, defaults to 0)
minutes	IntegerType	Number of minutes in the interval (optional, defaults to 0)
seconds	DecimalType(MAX_LONG_DIGITS, 6)	Number of seconds including microsecond precision (optional, defaults to 0)

Return Type: Returns a DayTimeIntervalType() representing the constructed interval.

Supported Data Types:

• days: Integer values
• hours: Integer values
• minutes: Integer values
• seconds: Decimal values with up to 6 decimal places for microsecond precision

Edge Cases:

• Null handling: Expression is null-intolerant (nullIntolerant = true), meaning if any input is null, the result is null
• Default values: Missing parameters default to 0 (literal values)
• Precision handling: Seconds parameter uses DecimalType with 6 decimal places to preserve microsecond precision
• Overflow behavior: Delegates to IntervalUtils.makeDayTimeInterval() for overflow validation and error handling
• Error context: Includes query context information for meaningful error messages when interval construction fails

Examples:

-- Create a 5-day, 3-hour, 30-minute, 45.5-second interval
SELECT make_dt_interval(5, 3, 30, 45.5);

-- Create a 2-day interval
SELECT make_dt_interval(2);

-- Create a 1-day, 12-hour interval  
SELECT make_dt_interval(1, 12);

-- Create an empty interval
SELECT make_dt_interval();

// DataFrame API usage
import org.apache.spark.sql.functions._

// Create interval from literal values
df.select(expr("make_dt_interval(5, 3, 30, 45.5)"))

// Create interval from column values
df.select(expr("make_dt_interval(day_col, hour_col, min_col, sec_col)"))

Implementation Approach

See the Comet guide on adding new expressions for detailed instructions.

1. Scala Serde: Add expression handler in spark/src/main/scala/org/apache/comet/serde/
2. Register: Add to appropriate map in QueryPlanSerde.scala
3. Protobuf: Add message type in native/proto/src/proto/expr.proto if needed
4. Rust: Implement in native/spark-expr/src/ (check if DataFusion has built-in support first)

Additional context

Difficulty: Large
Spark Expression Class: org.apache.spark.sql.catalyst.expressions.MakeDTInterval

Related:

• MakeYMInterval - Creates year-month intervals
• IntervalUtils - Utility class for interval operations
• DayTimeIntervalType - The data type returned by this expression
• Extract - Extracts components from interval values

---

This issue was auto-generated from Spark reference documentation.

[YutaLin]

take

[peterxcli]

This should be closed by #4541
cc @andygrove

[peterxcli]

oh, seems like #4338 is the native path for make_dt_interval with datafusion_spark::SparkMakeDtInterval. Suggest to rename the title to "[Feature] Support native make_dt_interval" and update the description to reflect the existing codegen path for make_dt_interval.