Control / subflow · All operations

Declare a workflow's inputs and outputs

A variant of Call a reusable workflow

Mark the points where a reusable workflow receives rows, files or variables from its caller and hands results back.

Status

In arcform

Not built yet

arcform does not do this yet; it would live in how a Protocol runs. Its place on the roadmap

What other tools call it

Meridian's classification; an independent blind rating agreed on 95.7% of 324 rows (variant, version 3).

Alteryx Designer (4)
  • Macro Inputread 2026-09-27

    Declare an input anchor and its template data for a macro

  • Macro Input In-DBread 2026-09-27

    Declare an in-DB input connection on a macro, with placeholder values

  • Macro Outputread 2026-09-27

    Declare an output anchor for a macro

  • Macro Output In-DBread 2026-09-27

    Declare an in-DB output connection on a macro

Apache Hop (3)
  • Injectorread 2026-09-27

    Entry point for rows injected through the Java API or a streaming consumer

  • Mapping inputread 2026-09-27

    Declare the input fields a mapping sub-pipeline receives

  • Mapping outputread 2026-09-27

    Declare the output fields a mapping sub-pipeline returns

Apache NiFi (2)
  • Input Portread 2026-09-27

    Entry point that transfers data into a Process Group

  • Output Portread 2026-09-27

    Exit point that transfers data out of a Process Group

EasyMorph (1)
  • Inputread 2026-09-27

    Receive the dataset passed in by the module or project that called this one

KNIME Analytics Platform (12)
Pentaho Data Integration (5)
  • Injectorread 2026-09-27

    Entry point for rows injected into the transformation through the Java API

  • MapReduce Inputread 2026-09-27

    Receive key-value pairs from Hadoop MapReduce into a mapper or reducer transformation

  • MapReduce Outputread 2026-09-27

    Return key-value pairs from a mapper or reducer transformation to Hadoop MapReduce

  • Mapping Input Specificationread 2026-09-27

    Declare the input interface of a mapping sub-transformation

  • Mapping Output Specificationread 2026-09-27

    Declare the output interface of a mapping sub-transformation

Background

A module interface expresses the elements that are provided and required by the module, and other modules can detect those elements.[1] In Apache Hop, the Mapping input transform describes the input of a mapping pipeline, whose data the calling parent pipeline provides.[2]

When to use it

In Apache Hop, a child pipeline called from the Simple Mapping transform must contain exactly one Mapping input, while one called from Multi Mapping may contain zero or more.[2] In Apache NiFi, an Input Port transfers data into a Process Group, a set of grouped components, and an Output Port transfers data out of it to destinations outside.[3] In Apache NiFi, all Ports within a Process Group must have unique names.[3]

In KNIME, the Container Input (Table) node outputs a configured template table when no external input is given.[4] In KNIME, the template table allows downstream nodes to be configured when no external input is present.[4] In Alteryx Designer, a Macro Input tool takes a template input, typed in or read from a file, for testing the macro as a standard workflow.[5] In EasyMorph, the Input action returns an empty dataset when the current module was not called from another module or project.[6] In EasyMorph, test data pasted into the Input action is not saved and does not survive reopening the project.[6] Give every declared input a small example table, so the reusable piece can be opened, run and checked without its caller. (Meridian's view)

The method

A parameter, or formal argument, is a variable that represents an argument to a function call, and a function's signature defines its parameters.[7] A function call evaluates each argument expression and associates the result with the corresponding parameter.[7]

In Apache Hop, the fields named in a Mapping input are those the mapping refers to, and a caller maps its own fields to those names.[2] In Apache Hop, the data entering a Mapping output transform is streamed to the parent pipeline that called the mapping.[8] In Alteryx Designer, the Show Field Map option of a Macro Input tool prompts the macro's user to map their data.[5]

In KNIME, a Container Input (Table) node's parameter name is exposed to callers, including the Call Workflow (Table Based) node.[4] In KNIME, when several Workflow Input nodes define the same parameter name, KNIME appends each node's ID to make the names unique.[9]

Sources

  1. [1]Modular programmingsecondary source
  2. [2]Mapping input | Apache Hopdocumentation
  3. [3]Apache NiFi User Guidedocumentation
  4. [4]Container Input (Table)documentation
  5. [5]Macro Input Tooldocumentation
  6. [6]INPUT [EasyMorph Help]documentation
  7. [7]Parameter (computer programming)secondary source
  8. [8]Mapping output | Apache Hopdocumentation
  9. [9]Workflow Inputdocumentation