C$REPLACE_ALL

The C$REPLACE_ALL library routine allows to replace parts of text in an alphanumeric variable according to a regular expression.

Syntax

CALL "C$REPLACE_ALL" USING variable                             regExp                             replacement                            [options]                            [errorDesc]                      GIVING returnCode

Parameters

variable	PIC X(n)	Variable subjected to replacement
regExp	PIC X(n)	Regular expression
replacement	PIC X(n)	Text to use as replacement. This text is not trimmed, so trailing spaces will be used as replacement as well.
options	PIC 9(n)	Optional parameter. It’s the sum between two or more of the following values defined in iscobol.def: CREP_CASE_INSENSITIVE CREP_LEFT_TRIMMED CREP_RIGHT_TRIMMED   These options affect only variable.
errorDesc	PIC X(n)	Optional parameter. It holds the error description in case of illegal regular expression.

Return code:

returnCode can be any signed numeric data item and provides additional information:

0	operation successful
1	invalid regular expression
2	insufficient parameters
3	malformed replacement

Examples:

Example - Replace the beginning of a string with a prefix

*> copy "iscobol.def" in working-storage *> define wrk-str, search-reg-exp, new-text and w-error as pic x(n)   move " second third fourth" to wrk-str move "^" to search-reg-exp | Search for the beginning of the string move "first" to new-text call "c$replace-all" using wrk-str                             search-reg-exp                            new-text                            crep-right-trimmed                             W-error *> wrk-str new value will be: first second third fourth