ABAP Keyword Documentation →  ABAP - Reference →  Processing Internal Data →  Character String and Byte String Processing →  Expressions and Functions for String Processing →  Regular Expressions →  Syntax of Regular Expressions 

Single Character Patterns

Single characters are represented by literal characters or operators. If preceded by a backslash \, a special character of an operator is interpreted as a literal character. This applies in particular for the backslash \ itself, so that the regular expression \ is the same as the single character \. If the backslash is followed by a literal character, the backslash is ignored as if it does not exist.

Other versions: 7.31 | 7.40 | 7.54

Literal Characters

A literal character is a character that is not a special character, or a special character preceded by a backslash \, or enclosed between \Q ... \E. As a search string, a literal character matches the same single character exactly.

Note

You can control case-sensitivity using the appropriate commands or methods.

Examples

The following table shows some results of a test that is case-sensitive.

Pattern	text	match
A	A	X
A	a	-
.	.	X
A	AB	-
AB	AB	X

The regular expression AB is a concatenation of two expressions for single characters.

Operators for Single Characters

These operators are constructed from the special characters ., [, ], ^, and - where the last two only work as special characters in certain positions between [ ]. The special characters can be made into literal characters using the prefix \.

Placeholders for Single Characters

The special character . is a placeholder for any single character. The operator \C has the same effect as the special character .. A regular expression . or \C matches exactly one single character.

Examples

The following table shows some results of a test that is not case-sensitive.

Pattern	Text	match
.	A	X
\C	a	X
.	AB	-
..	AB	X

The regular expression .. is a concatenation of two expressions for single characters.

Self-Defined Sets for Single Characters

The special characters [ ] can be placed around any number of literal characters or names for character classes (see below), and thus define a set of literal characters. A regular expression [...] matches exactly one single character listed as a literal character within the brackets, or which is contained in a specified character class. At least one literal character or one name for a character class (see below) must be contained within the brackets. A character [ or ], which is positioned directly after the opening bracket, is interpreted as a literal character. Some of the special characters that start with a backslash, such as \A or \Q, lose their special function within sets, and are interpreted as the simple literal character A or Q.

Examples

The following table shows some results from the test.

Pattern	Text	match
[ABC]	B	X
[ABC]	ABC	-
[AB][CD]	AD	X
[\d]	9	X

The regular expression [AB][CD] is a concatenation of two expressions for single characters.

Negation of a Self-Defined Set for Single Characters

If the character ^ is the first character in a self-defined set for single characters and is listed directly after [, it acts as a special character and negates the rest of the set of literal characters or character classes. A regular expression [^...] matches exactly one single character not listed as a literal character within the brackets, or which is not contained in a specified character class. A character ^ that is not listed directly after [ acts as a literal character.

Examples

The following table shows some results from the test.

Pattern	Text	match
[^ABC]	B	-
[^ABC]	Y	X
[^A][^B]	BA	X
[A^B]	^	X

The regular expression [^A][^B] is a concatenation of two expressions for single characters.

Ranges in a Self-Defined Set for Single Characters

If the character - is between two literal characters, it acts as a special character and defines a range between the literal characters. The range is the set of characters that is enclosed by literal characters in the code page of the current operating system. A regular expression [...-...] matches exactly one single character that is within the defined range. A character -, which is not between two literal characters, acts as a literal character. A literal character cannot be part of two ranges, for example, 'a-z-Z' is not a regular expression.

Examples

The following table shows some results from the test.

Pattern	Text	match
[A-Za-z0-9]	B	X
[A-Za-z0-9]	5	X
[A-Za-z0-9]	#	-
[A-Za-z0-9]	-	-
[A-Za-z0-9-]	-	X

In the last expression, the closing - does not act as a special character.

Character Classes

Predefined platform-independent and language-independent character classes can be specified within sets for single characters defined using [ ]:

• [:alnum:]
  Set of all alphanumeric characters (union of [:alpha:] and [:digit:])

• [:alpha:]
  Set of all uppercase and lowercase letters including language-specific special characters (umlauts, accents, diphthongs) without numbers.

• [:blank:]
  Blank characters and horizontal tabs

• [:cntrl:]
  Set of all control characters

• [:digit:]
  Set of all digits 0 to 9

• [:graph:]
  Set of all displayable characters except for blanks and horizontal tabs

• [:lower:]
  Set of all lowercase letters including language-dependent special characters (umlauts, accents, diphthongs)

• [:print:]
  Set of all displayable characters (union of [:graph:] and [:blank:])

• [:punct:]
  Set of all punctuation characters

• [:space:]
  Set of all blank characters, tabs, and carriage feeds

• [:unicode:]
  Set of all characters with a character representation larger than 255 (only in Unicode systems)

• [:upper:]
  Set of all uppercase letters including language-dependent special characters (umlauts, accents, diphthongs)

• [:word:]
  Set of all alphanumeric characters including underscore _

• [:xdigit:]
  Set of all hexadecimal digits ("0"-"9", "A"-"F", and "a"-"f")

Note

Character classes only act within [ ] as specified. A regular expression [:digit:] does not define the set of all digits, but instead defines a character set consisting of ":", "d", "g", "i", and "t". To specify the set of all digits, use the regular expression [[:digit:]].

Examples

The following table shows some results of a test that is case-sensitive.

Pattern	Text	match
[[:alnum:]]	a	X
[[:alnum:]]	;	-
[[:alpha:]]	1	-
[[:digit:][:punct:]]	X4	X
[[:digit:][:punct:]]	.	X
[[:lower:]]	â	X
[[:upper:]]	Ä	X

Abbreviations for Character Sets

For frequently used character sets, specific operators are available as abbreviations:

Character Set	Abbr.	Meaning
[[:digit:]]	\d	Placeholder for a digit
[^[:digit:]]	\D	Placeholder for a non-digit
[[:lower:]]	\l	Placeholder for a lowercase letter
[^[:lower:]]	\L	Placeholder for a character that is not a lowercase letter
[[:space:]]	\s	Placeholder for a blank character
[^[:space:]]	\S	Placeholder for a non-blank character
[[:upper:]]	\u	Placeholder for an uppercase letter
[^[:upper:]]	\U	Placeholder for a character that is not an uppercase letter
[[:word:]]	\w	Placeholder for an alphanumeric character plus underscore _
[^[:word:]]	\W	Placeholder for an non-alphanumeric character without underscore _

Note

If case is ignored by the ABAP statements FIND and REPLACE and when generating an object of the class CL_ABAP_REGEX, then \l and \u are equivalent to [[:alpha:]] or \L and \U is equivalent to [^[:alpha:]]. The special characters \w, \u, \l, \d, and \s can also be listed within sets [...]. Use of the special characters \W, \U, \L, \D, and \S within sets is not permitted and raises the exception CX_SY_INVALID_REGEX.

Examples

The following table shows some results of a test that is case-sensitive.

Pattern	Text	match
\d	4	X
\D	;	X
\l	u	X
\l	U	-
\L	S	X
\s		X
\S	#	X
\u	U	X
\U	.	X
\w	A	X
\w	8	X
\W	:	X
\W	_	-

Equivalence Classes

The operators [..] and [==] are reserved for future language enhancements and currently raise the exception CX_SY_INVALID_REGEX if used in sets.