Contact Us 1-800-596-4880
  1. Homepage
  2. DataWeave (2.3)
  3. Language Guide

  4. Define Functions

Define DataWeave Functions

You can define your own DataWeave functions using the fun declaration in the header of a DataWeave script. Before you begin, note that DataWeave version 2 is for Mule 4 apps. For a Mule 3 app, refer to the DataWeave version 1 documentation set in the Mule 3.9 documentation. For other Mule versions, you can use the version selector for the Mule Runtime table of contents.

DataWeave Function Definition Syntax

To define a function in DataWeave use the following syntax:

fun myFunction(param1, param2, ...) = <code to execute>

  • The fun keyword starts the definition of a function.

  • myFunction is the name you define for the function.

    Function names must be valid identifiers. For additional details about valid identifiers, see Rules for Declaring Valid Identifiers.

  • (param1, param2, …​ , paramn) represents the parameters that your function accepts.

    You can specify from zero to any number of parameters, separated by commas (,) and enclosed in parentheses.

  • The = sign marks the beginning of the code block to execute when the function is called.

  • <code to execute> represents the actual code that you define for your function.

Example Function Definitions

The following example defines a DataWeave function that accepts a single string argument, calls the upper function using the received string argument, and then outputs the result:

Example: Simple Function Definition
%dw 2.0
output application/json
fun toUpper(aString) = upper(aString)
---
toUpper("hello")
Output
"HELLO"

The next example shows how the argument to a DataWeave function can be any DataWeave expression:

Example: Using Expressions as Arguments
%dw 2.0
output application/json
fun toUpper(aString) = upper(aString)
---
toUpper("h" ++ "el" ++ lower("LO") )
Output
"HELLO"

The following example tests the type of the argument passed to the function by performing pattern matching with the built-in match operation:

Example: Function that Uses Pattern Matching
%dw 2.0
output application/json
fun toUpper(aString)
= aString match {
  case is String -> upper(aString)
  else -> null
}
---
toUpper("h" ++ "el" ++ lower("LO") )
Output
"HELLO"

This example creates a function that reformats a numeric string into a common phone number format:

Example: toPhoneFormat() Function
%dw 2.0
output application/json
fun toPhoneFormat(str: String) =
    "(" ++ str[0 to 2] ++ ") " ++ str[3 to 5] ++ "-" ++ str[6 to 9]
---
toPhoneFormat("1234567890")
Output
"(123) 456-7890"

Type Constraint Definition

You can define type constraint expressions when you declare functions to ensure that the arguments for a function call respect the defined type constraint.

To define type constraints for functions, use the following syntax:

fun myFunction(param1: Type, param2: Type): ResultType = <code to execute>
Example: Function with Type Constraints
%dw 2.0
output application/json
fun toUser(id: Number, userName: String): String = "you called the function toUser!"

In this example, the type constraints define that the function toUser accepts only a first argument of type Number and a second argument of type String.

Also, the function specifies a constraint for the result to be of type String. This return type constraint does not affect the function call; rather, this constraint applies to the function code to validate that the function generates a result of the defined type.

You can find more details about the DataWeave type system in https://docs.mulesoft.com/mule-runtime/4.3/dataweave-type-system.

Optional Parameters Definition

You can define optional parameters by assigning a default value to them. The following example shows a function that defines one parameter that is mandatory (name) and a second parameter that is optional (countryCode) and has a default value:

Example: Function with Optional Parameters
%dw 2.0
output application/json
fun createUserData(name, countryCode = "US") = { "User name" : name, "Location": countryCode}
---
[createUserData("Cristian", "AR"), createUserData("Max The Mule")]
Output
[
  {
    "User name": "Cristian",
    "Location": "AR"
  },
  {
    "User name": "Max The Mule",
    "Location": "US"
  }
]

Optional Parameter Order

DataWeave enables you to define optional parameters at the beginning or at the end of the parameter definition:

Example: Functions with Optional Parameters
%dw 2.0
output application/json
fun optionalParamsLast(a, b = 2, c = 3)
fun optionalParamsFirst(a = 1, b = 2, c)

When you call a function, the arguments are assigned from left to right. However, if you define optional parameters first, then the arguments are assigned from right to left. In addition, if all the parameters in a function are optional, the assignment is also from right to left:

Example: Optional Parameter Assignment
%dw 2.0
output application/json
fun optionalParamsLast(param1, param2 = 2, param3 = 3) =
    { "param1": param1, "param2": param2, "param3": param3}
fun optionalParamsFirst(param1 = 1, param2 = 2, param3) =
    { "param1": param1, "param2": param2, "param3": param3}
fun allParametersOptional(param1 = 1, param2 = 2, param3 = 3) =
    { "param1": param1, "param2": param2, "param3": param3}
---
{
  "optionalParamsLast(A)":optionalParamsLast('A'),
  "optionalParamsLast(A, B)":optionalParamsLast('A', 'B'),
  "optionalParamsLast(A, B, C)":optionalParamsLast('A', 'B', 'C'),
  "optionalParamsFirst(A)":optionalParamsFirst('A'),
  "optionalParamsFirst(A, B)":optionalParamsFirst('A', 'B'),
  "optionalParamsFirst(A, B, C)":optionalParamsFirst('A', 'B', 'C'),
  "allParametersOptional(A)":allParametersOptional('A'),
  "allParametersOptional(A, B)":allParametersOptional('A', 'B'),
  "allParametersOptional(A, B, C)":allParametersOptional('A', 'B', 'C')
}
Output
{
  "optionalParamsLast(A)": {
    "param1": "A",
    "param2": 2,
    "param3": 3
  },
  "optionalParamsLast(A, B)": {
    "param1": "A",
    "param2": "B",
    "param3": 3
  },
  "optionalParamsLast(A, B, C)": {
    "param1": "A",
    "param2": "B",
    "param3": "C"
  },
  "optionalParamsFirst(A)": {
    "param1": 1,
    "param2": 2,
    "param3": "A"
  },
  "optionalParamsFirst(A, B)": {
    "param1": 1,
    "param2": "A",
    "param3": "B"
  },
  "optionalParamsFirst(A, B, C)": {
    "param1": "A",
    "param2": "B",
    "param3": "C"
  },
  "allParametersOptional(A)": {
    "param1": 1,
    "param2": 2,
    "param3": "A"
  },
  "allParametersOptional(A, B)": {
    "param1": 1,
    "param2": "A",
    "param3": "B"
  },
  "allParametersOptional(A, B, C)": {
    "param1": "A",
    "param2": "B",
    "param3": "C"
  }
}

Function Overloading

DataWeave enables you to create multiple functions with the same name but different parameters. This feature is useful for defining different behaviors based on the arguments of a function call.

Parameters in overloaded functions differ in number or type. To understand more about defining type constraints on function parameters, see Type System.

DataWeave uses the first function it finds that accepts the arguments of the function call, based on the order in which the functions are declared in the script.

Example: Function that Uses Function Overloading
%dw 2.0
output application/json

fun toUpper(a: String) = upper(a)
fun toUpper(a: Any) = null
fun toUpper(a: String, b: Number) = upper(a) ++ b as String
---
toUpper("hi!")

The argument of the function call toUpper("hi!") matches the types String and Any, so it is possible to call the function with the first two definitions. However, DataWeave executes the function fun toUpper(a: String) = upper(a) because that function is defined before the one that uses the type Any.

Output of toUpper("hi!")
"HI!"
Output of toUpper(true)
null
Output of toUpper("age: ", 26)
"AGE: 26"

See Also

Create Custom Modules and Mappings

DataWeave Reference