Data types

Applies to: Databricks SQL Databricks Runtime

For rules governing how conflicts between data types are resolved, see SQL data type rules.

Supported data types

Databricks supports the following data types:

Data Type	Description
BIGINT	Represents 8-byte signed integer numbers.
BINARY	Represents byte sequence values.
BOOLEAN	Represents Boolean values.
DATE	Represents values comprising values of fields year, month and day, without a time-zone.
DECIMAL(p,s)	Represents numbers with maximum precision p and fixed scale s.
DOUBLE	Represents 8-byte double-precision floating point numbers.
FLOAT	Represents 4-byte single-precision floating point numbers.
INT	Represents 4-byte signed integer numbers.
INTERVAL intervalQualifier	Represents intervals of time either on a scale of seconds or months.
VOID	Represents the untyped NULL.
SMALLINT	Represents 2-byte signed integer numbers.
STRING	Represents character string values.
TIMESTAMP	Represents values comprising values of fields year, month, day, hour, minute, and second, with the session local timezone.
TIMESTAMP_NTZ	Represents values comprising values of fields year, month, day, hour, minute, and second. All operations are performed without taking any time zone into account.
TINYINT	Represents 1-byte signed integer numbers.
ARRAY < elementType >	Represents values comprising a sequence of elements with the type of elementType.
MAP < keyType,valueType >	Represents values comprising a set of key-value pairs.
STRUCT < [fieldName : fieldType [NOT NULL][COMMENT str][, …]] >	Represents values with the structure described by a sequence of fields.
VARIANT	Represents semi-structured data.
OBJECT	Represents values in a VARIANT with the structure described by a set of fields.

Important

Delta Lake does not support the VOID type.

Data type classification

Data types are grouped into the following classes:

• Integral numeric types represent whole numbers:

  • TINYINT

  • SMALLINT

  • INT

  • BIGINT

• Exact numeric types represent base-10 numbers:

  • Integral numeric

  • DECIMAL

• Binary floating point types use exponents and a binary representation to cover a large range of numbers:

  • FLOAT

  • DOUBLE

• Numeric types represents all numeric data types:

  • Exact numeric

  • Binary floating point

• Date-time types represent date and time components:

  • DATE

  • TIMESTAMP

  • TIMESTAMP_NTZ

• Simple types are types defined by holding singleton values:

  • Numeric

  • Date-time

  • BINARY

  • BOOLEAN

  • INTERVAL

  • STRING

• Complex types are composed of multiple components of complex or simple types:

  • ARRAY

  • MAP

  • STRUCT

  • VARIANT

  • OBJECT

Language mappings

Applies to: Databricks Runtime

Spark SQL data types are defined in the package org.apache.spark.sql.types. You access them by importing the package:

import org.apache.spark.sql.types._

SQL type	Data type	Value type	API to access or create data type
TINYINT	ByteType	Byte	ByteType
SMALLINT	ShortType	Short	ShortType
INT	IntegerType	Int	IntegerType
BIGINT	LongType	Long	LongType
FLOAT	FloatType	Float	FloatType
DOUBLE	DoubleType	Double	DoubleType
DECIMAL(p,s)	DecimalType	java.math.BigDecimal	DecimalType
STRING	StringType	String	StringType
BINARY	BinaryType	Array[Byte]	BinaryType
BOOLEAN	BooleanType	Boolean	BooleanType
TIMESTAMP	TimestampType	java.sql.Timestamp	TimestampType
TIMESTAMP_NTZ	TimestampNTZType	java.time.LocalDateTime	TimestampNTZType
DATE	DateType	java.sql.Date	DateType
year-month interval	YearMonthIntervalType	java.time.Period	YearMonthIntervalType (3)
day-time interval	DayTimeIntervalType	java.time.Duration	DayTimeIntervalType (3)
ARRAY	ArrayType	scala.collection.Seq	ArrayType(elementType [, containsNull]). (2)
MAP	MapType	scala.collection.Map	MapType(keyType, valueType [, valueContainsNull]). (2)
STRUCT	StructType	org.apache.spark.sql.Row	StructType(fields). fields is a Seq of StructField. [4](#4).
	StructField	The value type of the data type of this field(For example, Int for a StructField with the data type IntegerType)	StructField(name, dataType [, nullable]). [4](#4)
VARIANT	VariantType	org.apache.spark.unsafe.type.VariantVal	VariantType
OBJECT	Not Supported	Not supported	Not supported

Spark SQL data types are defined in the package org.apache.spark.sql.types. To access or create a data type, use factory methods provided in org.apache.spark.sql.types.DataTypes.

SQL type	Data Type	Value type	API to access or create data type
TINYINT	ByteType	byte or Byte	DataTypes.ByteType
SMALLINT	ShortType	short or Short	DataTypes.ShortType
INT	IntegerType	int or Integer	DataTypes.IntegerType
BIGINT	LongType	long or Long	DataTypes.LongType
FLOAT	FloatType	float or Float	DataTypes.FloatType
DOUBLE	DoubleType	double or Double	DataTypes.DoubleType
DECIMAL(p,s)	DecimalType	java.math.BigDecimal	DataTypes.createDecimalType() DataTypes.createDecimalType(precision, scale).
STRING	StringType	String	DataTypes.StringType
BINARY	BinaryType	byte[]	DataTypes.BinaryType
BOOLEAN	BooleanType	boolean or Boolean	DataTypes.BooleanType
TIMESTAMP	TimestampType	java.sql.Timestamp	DataTypes.TimestampType
TIMESTAMP_NTZ	TimestampNTZType	java.time.LocalDateTime	DataTypes.TimestampNTZType
DATE	DateType	java.sql.Date	DataTypes.DateType
year-month interval	YearMonthIntervalType	java.time.Period	YearMonthIntervalType (3)
day-time interval	DayTimeIntervalType	java.time.Duration	DayTimeIntervalType (3)
ARRAY	ArrayType	ava.util.List	DataTypes.createArrayType(elementType [, containsNull]).(2)
MAP	MapType	java.util.Map	DataTypes.createMapType(keyType, valueType [, valueContainsNull]).(2)
STRUCT	StructType	org.apache.spark.sql.Row	DataTypes.createStructType(fields). fields is a List or array of StructField. [4](#4)
	StructField	The value type of the data type of this field (For example, int for a StructField with the data type IntegerType)	DataTypes.createStructField(name, dataType, nullable) [4](#4)
VARIANT	VariantType	org.apache.spark.unsafe.type.VariantVal	VariantType
OBJECT	Not Supported	Not supported	Not supported

Spark SQL data types are defined in the package pyspark.sql.types. You access them by importing the package:

from pyspark.sql.types import *

SQL type	Data type	Value type	API to access or create data type
TINYINT	ByteType	int or long. (1)	ByteType()
SMALLINT	ShortType	int or long. (1)	ShortType()
INT	IntegerType	int or long	IntegerType()
BIGINT	LongType	long (1)	LongType()
FLOAT	FloatType	float (1)	FloatType()
DOUBLE	DoubleType	float	DoubleType()
DECIMAL(p,s)	DecimalType	decimal.Decimal	DecimalType()
STRING	StringType	string	StringType()
BINARY	BinaryType	bytearray	BinaryType()
BOOLEAN	BooleanType	bool	BooleanType()
TIMESTAMP	TimestampType	datetime.datetime	TimestampType()
TIMESTAMP_NTZ	TimestampNTZType	datetime.datetime	TimestampNTZType()
DATE	DateType	datetime.date	DateType()
year-month interval	YearMonthIntervalType	Not supported	Not supported
day-time interval	DayTimeIntervalType	datetime.timedelta	DayTimeIntervalType (3)
ARRAY	ArrayType	list, tuple, or array	ArrayType(elementType, [containsNull]).(2)
MAP	MapType	dict	MapType(keyType, valueType, [valueContainsNull]).(2)
STRUCT	StructType	list or tuple	StructType(fields). field is a Seq of StructField. (4)
	StructField	The value type of the data type of this field (For example, Int for a StructField with the data type IntegerType)	StructField(name, dataType, [nullable]).(4)
VARIANT	VariantType	VariantVal	VariantType()
OBJECT	Not Supported	Not supported	Not supported

SQL type	Data type	Value type	API to access or create data type
TINYINT	ByteType	integer (1)	‘byte’
SMALLINT	ShortType	integer (1)	‘short’
INT	IntegerType	integer	‘integer’
BIGINT	LongType	integer (1)	‘long’
FLOAT	FloatType	numeric (1)	‘float’
DOUBLE	DoubleType	numeric	‘double’
DECIMAL(p,s)	DecimalType	Not supported	Not supported
STRING	StringType	character	‘string’
BINARY	BinaryType	raw	‘binary’
BOOLEAN	BooleanType	logical	‘bool’
TIMESTAMP	TimestampType	POSIXct	‘timestamp’
TIMESTAMP_NTZ	TimestampNTZType	datetime.datetime	TimestampNTZType()
DATE	DateType	Date	‘date’
year-month interval	YearMonthIntervalType	Not supported	Not supported
day-time interval	DayTimeIntervalType	Not supported	Not supported
ARRAY	ArrayType	vector or list	list(type=’array’, elementType=elementType, containsNull=[containsNull]).(2)
MAP	MapType	environment	list(type=’map’, keyType=keyType, valueType=valueType, valueContainsNull=[valueContainsNull]).(2)
STRUCT	StructType	named list	list(type=’struct’, fields=fields). fields is a Seq of StructField. (4)
	StructField	The value type of the data type of this field (For example, integer for a StructField with the data type IntegerType)	list(name=name, type=dataType, nullable=[nullable]).(4)
VARIANT	Not Supported	Not supported	Not supported
OBJECT	Not Supported	Not supported	Not supported

(1) Numbers are converted to the domain at runtime. Make sure that numbers are within range.

(2) The optional value defaults to TRUE.

(3) Interval types

• YearMonthIntervalType([startField,] endField): Represents a year-month interval which is made up of a contiguous subset of the following fields:

  startField is the leftmost field, and endField is the rightmost field of the type. Valid values of startField and endField are 0(MONTH) and 1(YEAR).

• DayTimeIntervalType([startField,] endField): Represents a day-time interval which is made up of a contiguous subset of the following fields:

  startField is the leftmost field, and endField is the rightmost field of the type. Valid values of startField and endField are 0(DAY), 1(HOUR), 2(MINUTE), 3(SECOND).

(4) StructType

• StructType(fields) Represents values with the structure described by a sequence, list, or array of StructFields (fields). Two fields with the same name are not allowed.

• StructField(name, dataType, nullable) Represents a field in a StructType. The name of a field is indicated by name. The data type of a field is indicated by dataType. nullable indicates if values of these fields can have null values. This is the default.

Related articles

• Special floating point values