Subworkflows

You can use subworkflows to define a piece of logic or a set of steps you want to call multiple times, simplifying the workflow definition. Subworkflows are similar to a function or routine in a programming language. They can accept parameters and return values, allowing you to create more complex workflows with a broader range of applications.

For information on basic workflow tasks, see Create and update an existing workflow and Execute a workflow.

Main block

If a workflow has a subworkflow, the main workflow must be placed in a main block. Subworkflows are always defined after the main body of the workflow definition:

YAML

  main:
      steps:
          - STEP_NAME:
              ...
          ...

  SUBWORKFLOW_NAME:
      params: [PARAMETER_1,PARAMETER_2...]
      steps:
          - SUBWORKFLOW_STEP_NAME:
          ...

JSON

  {
    "main": {
      "steps": [
        {
          "STEP_NAME":
            ...
        }
        ...
      ]
    },
    "SUBWORKFLOW_NAME": {
      "params": "[PARAMETER_1,PARAMETER_2...]",
      "steps": [
        {
          "SUBWORKFLOW_STEP_NAME":
            ...
        }
      ]
    }
  }

If you don't need to pass parameters to the subworkflow, delete the params block. Note that any parameters must be enclosed in square brackets. For example:

params: [Street, ZipCode, Country]

Note that the steps block is required inside a SUBWORKFLOW_NAME block. It can contain the following:

Default parameters in subworkflows

Subworkflows support default values for parameters. The default parameter value is used only if a parameter isn't provided as part of a subworkflow call.

YAML

  SUBWORKFLOW_NAME:
      params: [PARAMETER_1, PARAMETER_2: DEFAULT_VALUE2...]
      steps:
          - SUBWORKFLOW_STEP_NAME:
          ...

JSON

  {
    "SUBWORKFLOW_NAME": {
      "params": "[PARAMETER_1,PARAMETER_2: DEFAULT_VALUE2...]",
      "steps": [
        {
          "SUBWORKFLOW_STEP_NAME":
            ...
        }
      ]
    }
  }

For example, the following workflow can be called with or without a Country parameter, and if the Country is not specified, it defaults to "United States".

YAML

  build_address:
      params: [Street, ZipCode, Country: "United States"]
      steps:
          - concatenate:
              return: ${Street + ", " + ZipCode + ", " + Country}

JSON

  {
    "build_address": {
      "params": [
        "Street",
        "ZipCode",
        {
          "Country": "United States"
        }
      ],
      "steps": [
        {
          "concatenate": {
            "return": "${Street + \", \" + ZipCode + \", \" + Country}"
          }
        }
      ]
    }
  }

Call a subworkflow

You call a subworkflow using the call field within a workflow step in the main workflow, optionally supplying arguments to pass to the subworkflow:

YAML

   main:
       steps:
           - STEP_NAME:
               call: SUBWORKFLOW_NAME
               args:
                   ARG_1: VALUE_1
                   ARG_2: VALUE_2
                   ...
               result: OUTPUT_VARIABLE
   SUBWORKFLOW_NAME:
        params: [PARAMETER_1,PARAMETER_2...]
        steps:
            - step_1:
            ...

JSON

  {
    "main": {
      "steps": [
        {
          "STEP_NAME": {
            "call": "SUBWORKFLOW_NAME",
            "args": {
              "ARG_1": "VALUE_1",
              "ARG_2": "VALUE_2"
            },
            "result": "OUTPUT_VARIABLE"
          }
        }
      ]
    },
    "SUBWORKFLOW_NAME": {
      "params": ["PARAMETER_1,PARAMETER_2"...],
      "steps": [
        {
          "step_1":
          ...
        }
      ]
    }
  }

If the subworkflow accepts parameters, you can pass arguments (ARG_1, ARG_2) and their values (VALUE_1, VALUE_2) to the workflow. If the subworkflow returns anything, you can store the data in the OUTPUT_VARIABLE.

This example defines a subworkflow named name_message and calls it from the main workflow:

YAML

  main:
      steps:
          - call_subworkflow:
              call: name_message
              args:
                  first_name: "Ada"
                  last_name: "Lovelace"
              result: output
          - call_subworkflow2:
              call: name_message
              args:
                  first_name: "Sherlock"
                  last_name: "Holmes"
              result: output2
          - return_message:
              return: ${output + " " + output2}

  name_message:
      params: [first_name, last_name, country: "England"]
      steps:
          - prepareMessage:
              return: ${"Hello " + first_name + " " + last_name + " from " + country + "."}

JSON

  {
    "main": {
      "steps": [
        {
          "call_subworkflow": {
            "call": "name_message",
            "args": {
              "first_name": "Ada",
              "last_name": "Lovelace"
            },
            "result": "output"
          }
        },
        {
          "call_subworkflow2": {
            "call": "name_message",
            "args": {
              "first_name": "Sherlock",
              "last_name": "Holmes"
            },
            "result": "output2"
          }
        },
        {
          "return_message": {
            "return": "${output + \" \" + output2}"
          }
        }
      ]
    },
    "name_message": {
      "params": [
        "first_name",
        "last_name",
        {
          "country": "England"
        }
      ],
      "steps": [
        {
          "prepareMessage": {
            "return": "${\"Hello \" + first_name + \" \" + last_name + \" from \" + country + \".\"}"
          }
        }
      ]
    }
  }

This workflow definition does the following:

  • The main workflow calls the name_message subworkflow twice, from the steps call_subworkflow and call_subworkflow2.
  • The two steps supply different inputs for the arguments first_name and last_name, which are then passed to the subworkflow.
  • The subworkflow name_message takes the arguments passed to it and constructs a simple message, supplying the default value of England for the country variable.
  • The subworkflow returns the message it constructs to the main workflow.
  • The steps in the main workflow store the results as output and output2, respectively.
  • The main workflow combines output and output2 into a single message and returns the result: Hello Ada Lovelace from England. Hello Sherlock Holmes from England.

For more information, see Calls.

Call a subworkflow in an expression

This example is identical to the preceding example except that the workflow calls the subworkflow using an expression. Note that expressions cannot contain calls to subworkflows with blocking steps.

YAML

  main:
      steps:
          - call_subworkflow:
              assign:
                - output: ${name_message("Ada","Lovelace")}
          - call_subworkflow2:
              assign:
                - output2: ${name_message("Sherlock","Holmes")}
          - return_message:
              return: ${output + " " + output2}

  name_message:
      params: [first_name, last_name, country: "England"]
      steps:
          - prepareMessage:
              return: ${"Hello " + first_name + " " + last_name + " from " + country + "."}

JSON

  {
    "main": {
      "steps": [
        {
          "call_subworkflow": {
            "assign": [
              {
                "output": "${name_message(\"Ada\",\"Lovelace\")}"
              }
            ]
          }
        },
        {
          "call_subworkflow2": {
            "assign": [
              {
                "output2": "${name_message(\"Sherlock\",\"Holmes\")}"
              }
            ]
          }
        },
        {
          "return_message": {
            "return": "${output + \" \" + output2}"
          }
        }
      ]
    },
    "name_message": {
      "params": [
        "first_name",
        "last_name",
        {
          "country": "England"
        }
      ],
      "steps": [
        {
          "prepareMessage": {
            "return": "${\"Hello \" + first_name + \" \" + last_name + \" from \" + country + \".\"}"
          }
        }
      ]
    }
  }

Simple subworkflow

This sample implements a simple subworkflow. When a workflow has a subworkflow, the main workflow must be placed in a main block. Subworkflows are always defined after the main body of the workflow definition.

YAML

main:
  steps:
    - first:
        call: hello
        args:
          input: "Kristof"
        result: some_output
    - second:
        return: ${some_output}

hello:
  params: [input]
  steps:
    - first:
        return: ${"Hello "+input}

JSON

{
  "main": {
    "steps": [
      {
        "first": {
          "call": "hello",
          "args": {
            "input": "Kristof"
          },
          "result": "some_output"
        }
      },
      {
        "second": {
          "return": "${some_output}"
        }
      }
    ]
  },
  "hello": {
    "params": [
      "input"
    ],
    "steps": [
      {
        "first": {
          "return": "${\"Hello \"+input}"
        }
      }
    ]
  }
}