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”).

Examples

Given the operand dataset DS_1:

Input DS_1 (see structure)

Id_1	Id_2	Me_1
1	A	hello world
2	A	say hello
3	A	he
4	A	hello!

Example 1

DS_r := replace (DS_1,"ello","i");

results in (see structure):

DS_r

Id_1	Id_2	Me_1
1	A	hi world
2	A	say hi
3	A	he
4	A	hi!

Example 2

DS_r := DS_1[ calc Me_2:= replace (Me_1,"ello","i")];

results in (see structure):

DS_r

Id_1	Id_2	Me_1	Me_2
1	A	hello world	hi world
2	A	say hello	say hi
3	A	he	he
4	A	hello!	hi!