Skip to content

ABAP Keyword Documentation →  ABAP - Reference →  Predefined Types, Data Objects, Functions, and Constructors →  Predefined Data Types →  Predefined ABAP Types 

Predefined Numeric Types

The data objects of the numeric data types are used to handle number values.

Other versions: 7.31 | 7.40 | 7.54

Properties

Type Length Standard Length Name
b 1 byte   1-byte integer (internal)
s 2 byte   2-byte integer (internal)
i 4 byte   4-byte integer
p 1 to 16 bytes 8 byte Packed number
decfloat16 8 byte   Decimal floating point number with 16 decimal places
decfloat34 16 byte   Decimal floating point number with 34 decimal places
f 8 byte   Binary floating point number with 17 decimal places

Value Ranges and Initial Values

Type Value Range Initial Value
b 0 to 255 0
s -32,768 to +32,767 0
i -2,147,483,648 to +2,147,483,647 0
p The valid length for packed numbers is between 1 and 16 bytes; two decimal places are packed intoone byte, where the last byte only contains one place and the plus/minus sign (the number of decimal places or digits is calculated from 2 * len-1); after the decimal separator, up to 14decimal places arepermitted. Depending on the field length len and the number of decimal places dec, the following appliesto the value range: (-10^(2len-1) +1) / (10^(+dec)) to (+10^(2len-1)-1) /(10^(+dec)) in increments of 10^(-dec); any intermediate values are rounded (decimal); invalid content produces undefined behavior. 0
decfloat16 Decimal floating point numbers of this type are represented internally with 16 decimal places inaccordance with the IEEE-754-2008 standard; valid values are numbers between 1E385(1E-16 - 1) and -1E-383for the negative range, 0 and +1E-383 to 1E385(1 - 1E-16) for the positive range. Values lying betweenthe ranges form the subnormal range and are rounded; outside of the subnormal range, each 16-digit decimal number can be represented precisely with such a decimal floating point number 0
decfloat34 Decimal floating point numbers of this type are represented internally with 34 decimal places inaccordance with the IEEE-754-2008 standard; valid values are numbers between 1E6145(1E-34 - 1) and -1E-6143for the negative range, 0 and +1E-6143 and 1E6145(1 - 1E-34) for the positive range. Values lying betweenthe ranges form the subnormal range and are rounded; outside of the subnormal range, each 34-digit decimal number can be represented precisely with such a decimal floating point number 0
f Binary floating point numbers are represented internally in accordance with the IEEE-754 standard(double precision); in ABAP, 17 decimal places are represented (one place before the decimal point and16 places in the fractional part). Valid values are numbers between -1.7976931348623157E+308 and -2.2250738585072014E-308for the negative range and between +2.2250738585072014E-308 and +1.7976931348623157E+308 for the positiverange, plus 0. Both validity intervals are extended in the direction of zero using subnormal numbers in accordance with the IEEE-754 standard. 0

Programming Guideline

Selecting the Numeric Type


Notes

  • The numeric data types are used for numeric calculations. Here, the data type f for binary floating point numbers is replaced largely by the types decfloat16 and decfloat34 for decimal floating point numbers.

  • The types b and s are internal types and cannot be specified either statically or dynamically in ABAP statements. Self-defined data types and data objects in ABAP programs have the data types b or s if they have been defined with reference to data elements in ABAP Dictionary that have the external data types INT1 or INT2.

  • The system class CL_ABAP_MATH contains constants for the minimum and maximum values of most numeric types.

  • The system class CL_ABAP_ELEMDESCR contains constants TYPE_P_MAX_LENGTH and TYPE_P_MAX_DECIMALS for the maximum length and the maximum number of decimal places p.

  • Since the decimal places of a floating point number of type f are represented internally as dual fractions, there is not an exact equivalent for every number that can be represented in the decimal system. This can lead to rounding errors in conversions and intermediate results of calculations. These errors can be avoided by using a two-step rounding procedure.

  • The type p, for which a length interval is specified in the second column in the first table, is generic, which means that the length is not part of the type description. Also, the fractional portion is undefined as well as the length. The entry in the Standard Length column specifies the length used in declarations of data objects when using types with generic lengths, if no explicit length is specified in the relevant statement.

  • The data type p is used to implement fixed point numbers. The number of decimal places in a packed number with the type p is a type attribute defined using the addition DECIMALS and is not saved together with the number. Technically, the number value is determined by dividing the saved sequence of digits in the packed number by 10 to the power of the number decimal places (10^(+dec)). In the definition of a packed number, the number of decimal places cannot be greater than the number of decimal places calculated from 2 *  len - 1. Otherwise, the decimal separator is outside the sequence of digits and not all decimal places can be given values. For example, if a packed number with length 1 and two decimal places has a value range of -0.09 to +0.09 in increments of 0.01, there is no possible value for which the first decimal place is filled, for example 0.14.

  • For data objects of data type p, the program attribute Fixed Point Arithmetic must be set so that the decimal separator is respected. Otherwise, in all operations, the content is handled as if there is no decimal separator. The sequence of digits in the variables of type p is interpreted as a whole number. Exceptions are:

  • Representation on dynpros

  • Assignments to character-like objects with the types c and string


Example

In accordance with the formula in the table, the value range of a packed number with length 2 and two decimal places is (-10^(2x2 -1) +1) / (10^2) to (+10^(2x2 -1) -1) / (10^2) and therefore =-9.99 to +9.99 in increments of 0.01. A value within this range, for example 1.428, is rounded up to 1.43.

DATA: pack   TYPE p LENGTH 2 DECIMALS 2, 
      result TYPE REF TO data. 

FIELD-SYMBOLS <result> TYPE ANY. 

result = cl_abap_exceptional_values=>get_min_value( pack ). 

IF result IS NOT INITIAL. 
  ASSIGN result->* TO <result>. 
  cl_demo_output=>write_data( <result> ). 
ENDIF. 

result = cl_abap_exceptional_values=>get_max_value( pack ). 

IF result IS NOT INITIAL. 
  ASSIGN result->* TO <result>. 
  cl_demo_output=>write_data( <result> ). 
ENDIF. 

cl_demo_output=>display( ).