BC ABAP Programming PHẦN 2 potx

Type ConversionsType Conversions Every time you assign a data object to a variable, the data types involved must either be compatible [Page 134] , that is, their technical attributes dat

Functions for all Numeric Data Types

The following built-in functions work with all three numeric data types (F, I, and P) as arguments

Functions for all numeric data types

Function Result

ABS Absolute value of argument

SIGN Sign of argument: 1 X > 0

SIGN( X) = 0 if X = 0

-1 X < 0

CEIL Smallest integer value not smaller than the argument

FLOOR Largest integer value not larger than the argument

TRUNC Integer part of argument

FRAC Fraction part of argument

The argument of these functions does not have to be a numeric data type If you choose anothertype, it is converted to a numeric type For performance reasons, however, you should use thecorrect type whenever possible The functions itself do not have a data type of their own They donot change the numerical precision of a numerical operation

DATA N TYPE P DECIMALS 2

DATA M TYPE P DECIMALS 2 VALUE '-5.55'

N = ABS( M ) WRITE: 'ABS: ', N

N = SIGN( M ) WRITE: / 'SIGN: ', N

N = CEIL( M ) WRITE: / 'CEIL: ', N

N = FLOOR( M ) WRITE: / 'FLOOR:', N

N = TRUNC( M ) WRITE: / 'TRUNC:', N

N = FRAC( M ) WRITE: / 'FRAC: ', N

The output appears as follows:

to type P Then the system processes the ABS function using the results of theconversion Then, during the assignment to the type C field T1, the result of thefunction is converted back to type C.

Floating-Point Functions

The following built-in functions work with floating point numbers (data type F) as an argument

Functions for floating point data types

ACOS, ASIN, ATAN; COS, SIN, TAN Trigonometric functions

COSH, SINH, TANH Hyperbolic functions

EXP Exponential function with base e (e=2.7182818285)

DATA: RESULT TYPE F,

Business Calculations

Business Calculations

For calculations in business applications, use packed numbers The program attribute [Page 75]

Fixed point arithmetic affects calculations using packed numbers

If the program attribute Fixed point arithmetic is not set, type P fields are interpreted as integers

without decimal places The decimal places that you specify in the DECIMALS addition of theTYPES or DATA statement only affect how the field is formatted in the WRITE statement

DATA: PACK TYPE P DECIMALS 2.

decimal point) You should therefore always set the Fixed point arithmetic attribute when you use

type P fields

DATA: PACK TYPE P

PACK = 1 / 3 * 3

WRITE PACK

If you have not set the Fixed point arithmetic attribute, the result is 0, since the

calculation is performed using integer accuracy, and the result is therefore roundedinternally to 0

If the program attribute Fixed point arithmetic is set, the result is 1 because the result

of the division is stored internally as 0.333333333333333333333333333333 with anaccuracy of up to 31 digits

Date and Time Calculations

Date and Time Calculations

Date and time fields have character types, not numeric ones However, you can still use dateand time fields in numeric operations To allow you to do so, the system performs automatic typeconversions [Page 187] You may also find it useful to use offset and length [Page 197]

specifications when using date and time fields in calculations

Example of a date calculation:

DATA: ULTIMO TYPE D

ULTIMO = SY-DATUM

ULTIMO+6(2) = '01' " = first day of this month

ULTIMO = ULTIMO - 1 " = last day of last month

Here, the last day of the previous month is assigned to the date field ULTIMO

1 ULTIMO is filled with the present date

2 Using an offset specification, the day is changed to the first day of the current month

3 1 is subtracted from ULTIMO Its contents are changed to the last day of the

previous month Before performing the subtraction, the system converts ULTIMO tothe number of days since 01.01.0001 and converts the result back to a date

Example of a time calculation:

DATA: DIFF TYPE I,

The number of hours between 02:00:00 and 20:00:00 is calculated

1 First, the difference between the time fields is calculated This is -64800, since T1and T2 are converted internally into 72000 and 7200 respectively

2 Second, with the operation MOD, this negative difference is converted to the totalnumber of seconds A positive difference would be unaffected by this calculation

3 Third, the number of hours is calculated by dividing the number of seconds by 3600

Date and Time Calculations

CONVERT DATE <d1> INTO INVERTED-DATE <d2>

Afterwards, you can convert the inverted data back into a normal date using the statementCONVERT INVERTED-DATE <d1> INTO DATE <d2>

These statements convert the field <d1> from the formats DATE or INVERTED-DATE to theformats INVERTED-DATE or DATE and assign it to the field <d2>

For the conversion, ABAP forms the nine's complement

DATA: ODATE TYPE D VALUE '19955011',

IDATE LIKE ODATE

DATA FIELD(8)

FIELD = ODATE WRITE / FIELD

CONVERT DATE ODATE INTO INVERTED-DATE IDATE

FIELD = IDATE WRITE / FIELD

CONVERT INVERTED-DATE IDATE INTO DATE ODATE

FIELD = ODATE WRITE / FIELD

Output:

80049488

19955011

19955011

Processing Character Strings

Processing Character Strings

ABAP contains a series of statements for processing character fields (types C, D, N, and T).There is no type conversion in these operations The statements treat all fields as though theywere type C fields, regardless of their actual types The internal form of type D, N, and T fields isthe same as for type C fields For this reason, you can use these statements for all character-type fields

Shifting Field Contents [Page 163]

Replacing Field Contents [Page 166]

Converting to Upper or Lower Case or Replacing Characters [Page 168]

Converting into a Sortable Format [Page 169]

Overlaying Strings [Page 170]

Finding Strings [Page 171]

Condensing Field Contents [Page 175]

Finding out the Length of a Character String [Page 174]

Concatenating Strings [Page 176]

Splitting Strings [Page 177]

There is also a variant of the MOVE statement that only works with strings:

Assigning Parts of Character Strings [Page 178]

Shifting Field Contents

Shifting Field Contents

You can shift the contents of a field using the following variants of the SHIFT statement SHIFTmoves field contents character by character

Shifting a String by a Given Number of Positions

SHIFT <c> [BY <n> PLACES] [<mode>]

This statement shifts the field <c> by <n> positions If you omit BY <n> PLACES, <n> is

interpreted as one If <n> is 0 or negative, <c> remains unchanged If <n> exceeds the length of

<c>, <c> is filled out with blanks <n> can be variable

With the different <mode> options, you can shift the field <c> in the following ways:

• <mode> is LEFT:

Shifts the field contents <n> places to the left and adds <n> blanks at the right-hand end

of the field (default)

• <mode> is RIGHT:

Shift <n> positions to the right and adds <n> blanks at the left-hand end of the field

• <mode> is CIRCULAR:

Shift <n> positions to the left so that <n> characters on the left appear on the right

DATA: T(10) VALUE 'abcdefghij',

Shifting Field Contents

Shifting a Structure up to a Given String

SHIFT <c> UP TO <str> <mode>

This statement searches the field contents of <c> until it finds the string <str> and shifts the field

<c> up to the edge of the field The <mode> options are the same as described above <str> can

Shifting a Structure According to the First or Last Character

SHIFT <c> LEFT DELETING LEADING <str>

SHIFT <c> RIGHT DELETING TRAILING <str>

This statement shifts the field <c> to the left or to the right, provided the first character on the left

Shifting Field Contents

Replacing Field Contents

Replacing Field Contents

To replace a string in a field with a different string, use the REPLACE statement

REPLACE <str1> WITH <str2> INTO <c> [LENGTH <l>]

The statement searches the field <c> for the first occurrence of the first <l> positions of thepattern <str1> If no length is specified, it searches for the pattern <str1> in its full length

Then, the statement replaces the first occurrence of the pattern <str1> in field <c> with the string

<str2> If a length <l> was specified, only the relevant part of the pattern is replaced

If the return code value of the system field SY-SUBRC is set to 0, this indicates that <str1> wasfound in <c> and replaced by <str2> A return code value other than 0 means that nothing wasreplaced <str1>, <str2>, and <len> can be variables

DATA: T(10) VALUE 'abcdefghij',

Replacing Field Contents

Converting to Upper or Lower Case or Replacing Characters

Converting to Upper or Lower Case or Replacing

Characters

The TRANSLATE statement converts characters into upper or lower case, or uses substitutionrules to convert all occurrences of one character to another character

Converting to Upper or Lower Case

TRANSLATE <c> TO UPPER CASE

TRANSLATE <c> TO LOWER CASE

These statements convert all lower case letters in the field <c> to upper case or vice versa

Substituting Characters

TRANSLATE <c> USING <r>

This statement replaces all characters in field <c> according to the substitution rule stored in field

<r> <r> contains pairs of letters, where the first letter of each pair is replaced by the secondletter <r> can be a variable

For more variants of the TRANSLATE statement with more complex substitution rules, see thekeyword documentation in the ABAP Editor

DATA: T(10) VALUE 'AbCdEfGhIj',

Converting into a Sortable Format

Converting into a Sortable Format

The CONVERT TEXT statement converts strings into a format that can be sorted alphabetically.CONVERT TEXT <c> INTO SORTABLE CODE <sc>

This statement writes a string <c> to a sortable target field <sc> The field <c> must be of type Cand the field <sc> must be of type X with a minimum size of 16 times the size of <c>

The field <sc> can serve as an alphabetic sort key for <c> These sorts are applied to internaltables [Page 252] and extract datasets [Page 332] If you sort unconverted character fields, thesystem creates an order that corresponds to the platform-specific internal coding of the individualletters The conversion CONVERT TEXT creates target fields in such a way that, after sorting thetarget fields, the order of the corresponding character fields is alphabetical

The method of conversion depends on the text environment of the current ABAP program Thetext environment is fixed in the user master record, but can also be set from a program using thefollowing ABAP statement:

SET LOCALE LANGUAGE <lg> [COUNTRY <cy>] [MODIFIER <m>]

This statement sets the text environment according to the language <lg> With the option

COUNTRY, you can specify the country additionally to the language, provided there are specific differences for languages With the option MODIFIER, you can specify another identifier,provided there are differences in the language within one country For example, in Germany, theorder for sorting umlauts is different in a dictionary from the order used in a telephone book.The fields <lg>, <cy>, and <m> must be of type C and must have the same lengths as the keyfields of table TCP0C Table TCP0C is a table, in which the text environment is maintainedplatform-dependent During the statement SET LOCALE, the system sets the text environmentaccording to the entries in TCP0C With the exception of the platform identity, which is

country-transferred internally, the table key is specified with the SET statement The platform identifier ispassed implicitly If <lg> equals SPACE, the system sets the text environment according to theuser’s master record If there is no entry in the table for the key specified, the system reacts with

a runtime error

The text environment influences all operations in ABAP that depend on the character set

For more information about this topic, see the keyword documentation for CONVERT TEXT andSET LOCALE LANGUAGE

For an example of alphabetical sorting, see Sorting Internal Tables [Page 272]

Overlaying Character Fields

Overlaying Character Fields

The OVERLAY statement overlays one string with another:

OVERLAY <c1> WITH <c2> [ONLY <str>]

This statement overlays all positions in field <c1> containing letters which occur in <str> with thecontents of <c2> <c2> remains unchanged If you omit ONLY <str>, all positions of <c1>

containing spaces are overwritten

If at least one character in <c1> was replaced, SUBRC is set to 0 In all other cases, SUBRC is set to 4 If <c1> is longer than <c2>, it is overlaid only in the length of <c2>

Finding Character Strings

Finding Character Strings

To search a character field for a particular pattern, use the SEARCH statement as follows:SEARCH <c> FOR <str> <options>

The statement searches the field <c> for <str> starting at position <n1> If successful, the returncode value of SY-SUBRC is set to 0 and SY-FDPOS is set to the offset of the string in the field

<c> Otherwise, SY-SUBRC is set to 4

The search string <str> can have one of the following forms

<str> Function

<pattern> Searches for <pattern> (any sequence of characters) Trailing blanks are ignored <pattern> Searches for <pattern> Trailing blanks are not ignored

*<pattern> A word ending with <pattern> is sought

<pattern>* Searches for a word starting with <pattern>

Words are separated by blanks, commas, periods, semicolons, colons, question marks,

exclamation marks, parentheses, slashes, plus signs, and equal signs

<option> in the SEARCH FOR statement can be any of the following:

• ABBREVIATED

Searches the field <c> for a word containing the string in <str> The characters can beseparated by other characters The first letter of the word and the string <str> must bethe same

If the search string is found, all the characters in the search string (and all the characters

in between when using ABBREVIATED) are converted to upper case

DATA STRING(30) VALUE 'This is a little sentence.'

WRITE: / 'Searched', 'SY-SUBRC', 'SY-FDPOS'

ULINE /1(26)

SEARCH STRING FOR 'X'

WRITE: / 'X', SY-SUBRC UNDER 'SY-SUBRC',

SY-FDPOS UNDER 'SY-FDPOS'

SEARCH STRING FOR 'itt '

WRITE: / 'itt ', SY-SUBRC UNDER 'SY-SUBRC',

SY-FDPOS UNDER 'SY-FDPOS'

Finding Character Strings

SEARCH STRING FOR '.e '

WRITE: / '.e ', SY-SUBRC UNDER 'SY-SUBRC',

SY-FDPOS UNDER 'SY-FDPOS'

SEARCH STRING FOR '*e'

WRITE: / '*e ', SY-SUBRC UNDER 'SY-SUBRC',

SY-FDPOS UNDER 'SY-FDPOS'

SEARCH STRING FOR 's*'

WRITE: / 's* ', SY-SUBRC UNDER 'SY-SUBRC',

SY-FDPOS UNDER 'SY-FDPOS'

SEARCH STRING FOR 'ft' ABBREVIATED

WRITE: / 'SY-FDPOS:', SY-FDPOS

POS = SY-FDPOS + 2

SEARCH STRING FOR 'ft' ABBREVIATED STARTING AT POS AND MARK

WRITE / STRING

WRITE: / 'SY-FDPOS:', SY-FDPOS

OFF = POS + SY-FDPOS -1

WRITE: / 'Off:', OFF

Finding Character Strings

Finding the Length of a Character String

Finding the Length of a Character String

The ABAP function STRLEN returns the length of a string up to the last character that is not aspace

[COMPUTE] <n> = STRLEN( <c> )

STRLEN processes any operand <c> as a character data type, regardless of its real type There

is no type conversion

As with mathematical functions [Page 157], the keyword COMPUTE is optional

DATA: INT TYPE I,

WORD1(20) VALUE '12345'

WORD2(20)

WORD3(20) VALUE ' 4 '

INT = STRLEN( WORD1 ) WRITE INT

INT = STRLEN( WORD2 ) WRITE / INT

INT = STRLEN( WORD3 ) WRITE / INT

The results are 5, 0, and 4 respectively

Condensing Field Contents

Condensing Field Contents

The CONDENSE statement deletes redundant spaces from a string:

CONDENSE <c> [NO-GAPS]

This statement removes any leading blanks in the field <c> and replaces other sequences ofblanks by exactly one blank The result is a left-justified sequence of words, each separated byone blank If the addition NO-GAPS is specified, all blanks are removed

DATA: STRING(25) VALUE ' one two three four',

WRITE: / 'Length: ', LEN

CONDENSE STRING NO-GAPS

LEN = STRLEN( STRING )

Concatenating Character Strings

Concatenating Character Strings

The CONCATENATE statement combines two or more separate strings into one

CONCATENATE <c1> <cn> INTO <c> [SEPARATED BY <s>]

This statement concatenates the character fields <c1> to <cn> and assigns the result to <c> Thesystem ignores spaces at the end of the individual source strings

The addition SEPARATED BY <s> allows you to specify a character field <s> which is placed inits defined length between the individual fields

If the result fits into <c>, SUBRC is set to 0 However, if the result has to be truncated, SUBRC is set to 4

SY-DATA: C1(10) VALUE 'Sum',

Sum - mer - holi - day

In C1 to C5, the trailing blanks are ignored The separator SEP retains them

Splitting Character Strings

Splitting Character Strings

To split a character string into two or more smaller strings, use the SPLIT statement as follows:SPLIT <c> AT <del> INTO <c1> <cn>

The system searches the field <c> for the separator <del> The parts before and after the

separator are placed in the target fields <c1> <cn>

To place all fragments in different target fields, you must specify enough target fields Otherwise,the last target field is filled with the rest of the field <c> and still contains delimiters

If all target fields are long enough and no fragment has to be truncated, SY-SUBRC is set to 0.Otherwise it is set to 4

The output appears as follows:

Part 1 *** Part 2 *** Part 3 *** Part 4 *** Part 5

You can also split a string into the individual lines of an internal table as follows:

SPLIT <c> AT <del> INTO TABLE <itab>

The system adds a new line to the internal table <itab> for each part of the string

Splitting Character Strings

Assigning Parts of Character Strings

The following variant of the MOVE statement works only with type C fields:

MOVE <c1> TO <c2> PERCENTAGE <p> [RIGHT]

Copies the percentage <p> percent of the character field <c1> left-justified (or right-justified ifspecified with the RIGHT option) to <c2>

The value of <p> can be a number between 0 and 100 The length to be copied from <f1> isrounded up or down to the next whole number

If one of the arguments in the statement is not type C, the parameter PERCENTAGE is ignored

DATA C1(10) VALUE 'ABCDEFGHIJ',

Single Bit Processing in Hexadecimal Fields

Single Bit Processing in Hexadecimal Fields

In a hexadecimal field (type X), you can process the individual bits ABAP interprets the contents

of hex fields byte by byte A hexadecimal field with length n is n bytes long, and has a displaylength in ABAP of 2xn The decimal values 0 - 255 are represented in hexadecimal by the

Decimal value in byte sequence

Binary representation of one byte

Hexadecimal representation

The illustration shows how a byte in a sequence with the decimal value 181 is represented inhex The first four bits have the decimal value 11, the second four bits have the decimal value 5.This leads to the hexadecimal value ‘B5’ - 11x16+5 = 181

ABAP contains statements that allow you to read and set the individual bits in a type X field.There are also special logical operators that you can use to compare bit sequences [Page 234].Bit sequence processing allows you to process complex conditions and set operations moreefficiently

Setting and Reading Bits [Page 180]

Bit Operations [Page 182]

Set Operations Using Bit Sequences [Page 184]

Setting and Reading Bits

Setting and Reading Bits

In a hexadecimal field (type X), you can set or read individual bits

Setting Bits

To set an individual bit, use the statement

SET BIT <n> OF <f> [TO <b>]

This statement sets the bit at position <n> of field <f> to 1 (or to the value of field <b>) Thesystem must be able to interpret field <n> as a positive integer The field <f> must have datatype X The field <b> must contain the value 0 or 1 If the bit is set, SY-SUBRC is set to 0 If <n>

is greater than the length of <f>, SY-SUBRC is unequal to zero If <n> or <b> contain invalidvalues, a runtime error occurs

DATA HEX(3) TYPE X.

SET BIT: 09 OF HEX TO 1,

The bits of the second byte in the three-character hexadecimal field HEX are set to

‘10110101’, and the list output is as follows:

00B500

The decimal value of the second byte is 181

Reading Bits

To read an individual bit, use the statement

GET BIT <n> OF <f> INTO <b>

This statement reads the bit at position <n> of field <f> into field <b> The system must be able

to interpret field <n> as a positive integer The field <f> must have data type X If the bit is read,SY-SUBRC is set to 0 If <n> is greater than the length of <f>, SY-SUBRC is unequal to zero and

<b> is set to zero If <n> contains an invalid value, a runtime error occurs

Setting and Reading Bits

Here, the eight buts of the single-character hexadecimal field HEX (value ‘B5’) areread and displayed as follows:

10110101

<bitexp> can be one of the following bit expressions:

Bit expression Meaning

BIT-NOT <y> Negation

<y> BIT-AND <z> And

<y> BIT-XOR <z> Exclusive or

<y> BIT-OR <z> Or

The operands <y> and <z> are linked bit by bit in the above operators, and the result is placed inthe result field <x> The result field <x>, and the operands <y> and <z> must all be of type X Ifthe field lengths are different, all operands are converted [Page 188] to the field length of theresult field <x>

The following rules apply to bit operations:

<y> <z> BIT-NOT <y> <y> BIT-AND <z> <y> BIT-XOR <z> <y> BIT-XOR <z>

As with mathematical expressions, you can use parentheses with bit expressions

DATA: HEX1(1) TYPE X VALUE 'B5',

HEX2(1) TYPE X VALUE '5B', HEX3(1) TYPE X.

HEX3 = BIT-NOT ( HEX1 BIT-XOR HEX2 ).

WRITE HEX3.

The output is:

Bit Operations

1 0 1 1 0 1 0 1

B 5 HEX1

Hexadecimal

5 B HEX2

0 1 0 1 1 0 1 1

1 1 1 0 1 1 1 0 0 0 0 1 0 0 0 1 HEX1

Set Operations Using Bit Sequences

Set Operations Using Bit Sequences

If you have a set of m elements, you can represent any subset as a sequence of bits If the nthelement of the set is present in a subset, the nth bit is set to 1, otherwise, it is set to 0 The

universal set therefore contains no zeros

In ABAP, you can represent a set of m elements using a field with type X and length of at leastm/8 To include a member of the universal set in a set, use the SET BIT statement To checkwhich members of the universal set are included in a particular set, use the GET BIT statement.You can use bit operations for the following set operations:

Set operation Bit operation

Intersection BIT-AND

Symmetrical difference BIT-XOR

You can also use the O operator when comparing bit sequences [Page 234] to determine

whether a particular set is a subset of another

DATA: FRANKFURT(4) TYPE X,

FRISCO(4) TYPE X, INTERSECT(4) TYPE X, UNION(4) TYPE X,

DATA: CARRID TYPE SPFLI-CARRID,

CARRIER LIKE SORTED TABLE OF CARRID

WITH UNIQUE KEY TABLE LINE.

DATA WA TYPE SPFLI.

SELECT CARRID FROM SCARR INTO TABLE CARRIER.

SELECT CARRID CITYFROM FROM SPFLI

INTO CORRESPONDING FIELDS OF WA.

WRITE: / WA-CARRID, WA-CITYFROM.

READ TABLE CARRIER FROM WA-CARRID TRANSPORTING NO FIELDS.

CASE WA-CITYFROM.

WHEN 'FRANKFURT'.

SET BIT SY-TABIX OF FRANKFURT.

WHEN 'SAN FRANCISCO'.

SET BIT SY-TABIX OF FRISCO.

Set Operations Using Bit Sequences

WRITE 'Airlines flying from Frankfurt and San Francisco:'.

This produces the following output list:

The program uses four hexadecimal fields with length 4 - FRANKFURT, FRISCO,INTERSECT, and UNION Each of these fields can represent a set of up to 32elements The basic set is the set of all airlines from database table SCARR Eachbit of the corresponding bit sequences representes one airline To provide an index,the external index table CARRIER is created and filled with the airline codes from

Set Operations Using Bit Sequences

table SCARR It is then possible to identify an airline using the internal index of tableCARRIER

In the SELECT loop for database table SPFLI, the corresponding bit for the airline isset either in the FRANKFURT field or the FRISCO field, depending on the departurecity The line number SY-TABIX is determined using a READ statement in which nofields are transported

The intersection and union of FRANKFURT and FRISCO are constructed using thebit operations BIT-AND and BIT-OR

The bits in INTERSECT and UNION are read one by one and evaluated in two DOloops For each position in the fields with the value 1, a READ statement retrievesthe airline code from the table CARRIER

Type Conversions

Type Conversions

Every time you assign a data object to a variable, the data types involved must either be

compatible [Page 134] , that is, their technical attributes (data type, field length, number ofdecimal places) must be identical, or the data type of the source field must be convertible into thedata type of the target field

In ABAP, two non-compatible data types can be converted to each other if a correspondingconversion rule exists If data types are compatible, no conversion rule is necessary

If you use the MOVE statement to transfer values between non-compatible objects, the value ofthe source object is always converted into the data type of the target object With all ABAPoperations that perform value assignments between data objects (for example, arithmetic

operations or filling internal tables), the system handles all the necessary type conversions as forthe MOVE statement If you try to assign values between two data types for which no conversionrule exists, a syntax error or runtime error occurs

The following sections contain the conversion rules for incompatible ABAP data types:

Conversion Rules for Elementary Data Types [Page 188]

Conversion Rules for References [Page 192]

Conversion Rules for Structures [Page 193]

Conversion Rules for Internal Tables [Page 195]

With some ABAP statements that pass data between different objects, the alignment of the dataobjects is also important

Alignment of Data Objects [Page 196]

Conversion Rules for Elementary Data Types

Conversion Rules for Elementary Data Types

There are eight predefined ABAP data types [Page 97] There are 64 possible type combinationsbetween these elementary data types ABAP supports automatic type conversion and lengthadjustment for all of them except type D (date) and type T (time) fields which cannot be

converted into each other

The following conversion tables define the rules for converting elementary data types for allpossible combinations of source and target fields

Source Type Character

Conversion table for source type C

Target Conversion

C The target field is filled from left to right If it is too long, it is filled with blanks from theright If it is too short, the contents are truncated at the right-hand end

D The character field should contain an 8-character date in the format YYYYMMDD

F The contents of the source field must be a valid representation of a type F field asdescribed in Literals [Page 120]

N Only the digits in the source field are copied The field is right-justified and filled withtrailing zeros

I

P The source field must contain the representation of a decimal number, that is, asequence of digits with an optional sign and no more than one decimal point Thesource field can contain blanks If the target field is too short, an overflow may occur.This may cause the system to terminate the program

T The character field should contain a 6-character time in HHMMSS format

X Since the character field should contain a hexadecimal-character string, the only validcharacters are 0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F This character string is packed as ahexadecimal number, transported left-justified, and padded with zeros or truncated onthe right

Source Type Date

Conversion table for source type D

Target Conversion

C The date is transported left-justified without conversion

D Transport without conversion

F The date is converted into a packed number The packed number is then convertedinto a floating point number (see corresponding table)

N The date is converted into a character field The character field is then converted into

a numeric text field (see corresponding table)

Conversion Rules for Elementary Data Types

Source Type Floating Point Number

Conversion table for source type F

Target Conversion

C The floating point number is converted to the <mantissa>E<exponent> format andtransported to the character field The value of the mantissa lies between 1 and 10unless the number is zero The exponent is always signed If the target field is tooshort, the mantissa is rounded The length of the character field should be at least 6bytes

D The source field is converted into a packed number The packed number is thenconverted into a date field (see corresponding table)

F Transport without conversion

N The source field is converted into a packed number The packed number is thenconverted into a numeric text field (see corresponding table)

I

P The floating point number is converted to an integer or fixed point value and, ifnecessary, rounded

T The source field is converted into a packed number The packed number is thenconverted into a time field (see corresponding table)

X The source field is converted into a packed number The packed number is thenconverted into a hexadecimal number (see corresponding table)

Source Type Integer

Type I is always treated in the same way as type P without decimal places Wherever type P ismentioned, the same applies to type I fields

Source Type Numeric Text

Conversion table for source type N

Target Conversion

C The numeric field is treated like a character field Leading zeros are retained

D The numeric field is converted into a character field The character field is thenconverted into a date field (see corresponding table)

F The numeric field is converted into a packed number The packed number is thenconverted into a floating point number (see corresponding table)

N The numeric field is transported right-justified and padded with zeros or truncated onthe left

Conversion Rules for Elementary Data Types

Source Type Packed Number

If the program attribute Fixed point arithmetic is set, the system rounds type P fields according to

the number of decimal places or fills them out with zeros

Conversion table for source type P

Target Conversion

C The packed field is transported right-justified to the character field, if required with adecimal point The first position is reserved for the sign Leading zeros appear asblanks If the target field is too short, the sign is omitted for positive numbers If this isstill not sufficient, the field is truncated on the left ABAP indicates the truncation with

an asterisk (*) If you want the leading zeros to appear in the character field, useUNPACK instead of MOVE

D The packed field value represents the number of days since 01.01.0001 and isconverted to a date in YYYYMMDD format

F The packed field is accepted and transported as a floating point number

N The packed field is rounded if necessary, unpacked, and then transported justified The sign is omitted If required, the target field is padded with zeros on theleft

right-I

P The packed field is transported right-justified If the target field is too short, anoverflow occurs

T The packed field value represents the number of seconds since midnight and isconverted to a time in HHMMSS format

X The packed field is rounded if necessary and then converted to a hexadecimal

number Negative numbers are represented by the two's complement If the targetfield is too short, the number is truncated on the left

Source Type Time

Conversion table for source type T

Target Conversion

C The source field is transported left-justified without conversion

D Not supported Results in an error message during the syntax check or in a runtimeerror

F The source field is converted into a packed number The packed number is thenconverted into a floating point number (see corresponding table)

N The date is converted into a character field The character field is then converted into

a numeric text field (see corresponding table)

I

P The date is converted to the number of seconds since midnight.

T Transport without conversion

X The date is converted to the number of seconds since midnight in hexadecimal

Conversion Rules for Elementary Data Types

Target Conversion

C The value in the hexadecimal field is converted to a hexadecimal character string,transported left-justified to the target field, and padded with zeros

D The source field value represents the number of days since 01.01.0001 and is

converted to a date in YYYYMMDD format

F The source field is converted into a packed number The packed number is thenconverted into a floating point number (see corresponding table)

N The source field is converted into a packed number The packed number is thenconverted into a numeric text field (see corresponding table)

I

P

The value of the source field is interpreted as a hexadecimal number It is converted

to a packed decimal number and transported right-justified to the target field If thehexadecimal field is longer than 4 bytes, only the last four bytes are converted If it istoo short, a runtime error may occur

T The source field value represents the number of seconds since midnight and isconverted to a time in HHMMSS format

X The value is transported left-justified and filled with X'00' on the right, if necessary

Conversion Rules for References

Conversion Rules for References

ABAP currently uses class and interface variables within ABAP Objects [Page 1344] Both arepointers to objects You can assign values to them in the following combinations:

• If the two class references are incompatible, the class of the target field must be thepredefined empty class OBJECT

• When you assign a class reference to an interface reference, the class of the source fieldmust implement the interface of the target field

• If two interface references are incompatible, the interface of the target field must containthe interface of the source field as a component

• When you assign an interface reference to a class reference, the class of the source fieldmust be the predefined empty class OBJECT

Conversion Rules for Structures

Conversion Rules for Structures

ABAP has one rule for converting structures that do not contain internal tables as components.There are no conversion rules for structures that contain internal tables You can only makeassignments between structures that are compatible

You can combine convertible structures in the following combinations:

• Converting a structure into a non-compatible structure

• Converting elementary fields into structures

• Converting structures into elementary fields

In each case, the system first converts all the structures concerned to type C fields and thenperforms the conversion between the two resulting elementary fields The length of the type Cfields is the sum of the lengths of the structure components This rule applies to all operationsusing structures that do not contain internal tables

If a structure is aligned [Page 196], the filler fields are also added to the length of the type C field

A non-aligned structure without filler fields:

C 75

If you convert a structure into a shorter structure, the original structure is truncated If you

convert a structure into a longer one, the parts at the end are not initialized according to theirtype, but filled with blanks

It can make sense to assign a structure to another, incompatible, structure if, for example, thetarget structure is shorter than the source, and both structures have the same construction overthe length of the shorter structure However, numeric components of structures that are filled inincompatible assignments may contain nonsensical or invalid values that may cause runtimeerrors

DATA: BEGIN OF FS1,

INT TYPE I VALUE 5,

PACK TYPE P DECIMALS 2 VALUE ‘2.26’,

TEXT(10) TYPE C VALUE ‘Fine text’,

FLOAT TYPE F VALUE ‘1.234e+05’,

DATA TYPE D VALUE ‘19950916’,

END OF FS1

DATA: BEGIN OF FS2,

INT TYPE I VALUE 3,

PACK TYPE P DECIMALS 2 VALUE ‘72.34’,

Conversion Rules for Structures

TEXT(5) TYPE C VALUE ‘Hello’,

END OF FS2

WRITE: / FS1-INT, FS1-PACK; FS1-TEXT, FS1-FLOAT, FS1-DATE

WRITE: / FS2-INT, FS2-PACK, FS2-TEXT

Conversion Rules for Internal Tables

Conversion Rules for Internal Tables

Internal tables can only be converted into other internal tables You cannot convert them intostructures or elementary fields

Internal tables are convertible if their line types are convertible The convertibility of internaltables does not depend on the number of lines

Conversion rules for internal tables:

• Internal tables which have internal tables as their line type are convertible if the internaltables which define the line types are convertible

• Internal tables which have line types that are structures with internal tables as

components are convertible according to the conversion rules for structures [Page 193] ifthe structures are compatible