FlowmageDocs

How data moves between nodes

Each node receives what the node before it gave back. See it in the node's dialog, put it in a field with an expression, and check the result.

Why it matters

Most fields take their values from earlier nodes. What a node receives tells you what its fields can use, and why a value comes out empty.

How it works

A node's dialog has three panes.

The dialog of the Condition Is it down?, with Input on the left, its rules in the middle and Output on the right with Test step
PaneWhat it holds
InputEverything this node can read: Suggested, the value the field most likely wants; From earlier steps, each earlier node and Start, for what started the run; Build, with Functions, Dates, Secrets and Files. Search data and functions finds any of them.
SettingsThe node's fields, with values from Input drawn as chips. Result, under a field, shows what it gives with the data the editor has.
OutputWhat the node gave back the last time it ran in the editor, as Table, JSON or Schema. Test step runs it for real, on Test.

To use a value, click in a field, then click the value under Input. Flowmage writes the expression, such as {{ json.orderId }}.

No data yet under Input means nothing before this node has run: Test step runs those nodes with the sample on Start.

What json holds

Where the node standsWhat json is
One node leads into itThat node's output.
Several nodes lead into itEmpty (null). Read each one by name, as node["Check the site"], from From earlier steps.
First node inside a loopEmpty. Read the current item as loop.item.
First node after StartThe trigger payload. In the editor, that is the Sample data on Start.

What started the run is also trigger, from any node.

An example

Open Start in a flow with no trigger yet, switch Sample data to JSON, and enter:

{
  "orderId": "A-1001",
  "status": "paid",
  "total": 250
}

Add a Log node after Start with the Message Order {{ json.orderId }}. Result under Message reads Order A-1001, and after Test run the run's log shows that line on the Log node.

A table node that adds a row gives back that row's ID and version:

  • rowId"row_1"
  • rowVersion1

Limits

The Output you see in the editor stays with the draft: a published version never includes it, and real runs keep their own data in Runs.