String pattern replacement: replace

Syntax

replace ( op , pattern1, pattern2 )

Input parameters

op	the operand
pattern1	the pattern to be replaced
pattern2	the replacing pattern

Examples of valid syntaxes

replace(DS_1, "Hello", "Hi")
replace(DS_1, "Hello")

Semantics for scalar operations

Replaces all the occurrences of a specified string-pattern (pattern1) with another one (pattern2). If pattern2 is omitted then all occurrences of pattern1 are removed. For example:

replace("Hello world", "Hello", "Hi") gives "Hi world"
replace("Hello world", "Hello") gives " world"
replace ("Hello", "ello", "i") gives "Hi"

Input parameters type

op

dataset { measure<string> _+ }
| component<string>
| string

pattern1, pattern2

component<string>
| string

Result type

result

dataset { measure<string> _+ }
| component<string>
| string

Additional Constraints

The second parameter (pattern1) cannot be omitted.

Behaviour

As for the invocations at Data Set level, the operator has the behaviour of the “Operators applicable on one Scalar Value or Data Set or Data Set Component”. As for the invocations at Component or Scalar level, the operator has the behaviour of the “Operators applicable on more than two Scalar Values or Data Set Components” (see the section “Typical behaviours of the ML Operators”).