Loop
The Loop step repeats a set of steps once per item. On the canvas the loop is drawn as a
group containing the steps that repeat, so its scope is visible at a glance.
Configuration
| Field | Notes |
|---|---|
| Input Type | How the loop is driven — For Each over a collection, While or Do While against a condition, Previous Step Value, or Multi Record File to iterate a file's records |
| Loop On | The collection or value being iterated |
| Type | The data type of each item, so later steps know its shape |
| Intely Storage File Path | The file to read, for a multi-record file loop |
| Field Delimiter | The delimiter of that file |
| Entry Offset | Rows to skip before the first record — used to step past a header |
| On Error | What happens when an iteration fails: Continue With Next Iteration, or Exit Loop |
| Run iterations in parallel | Process items concurrently rather than one at a time |
Controlling the loop from inside
| Step | Effect |
|---|---|
| Continue Loop | Skip the rest of this iteration, carry on with the next |
| Stop Loop | Leave the loop entirely |
Notes
- On Error is the setting that matters most. Left to exit, one malformed record ends the
batch; set to continue, the run completes and the failures are in the log. - Run iterations in parallel is faster but gives up ordering, and multiplies the load on
whatever each iteration calls. Leave it off when the downstream system is rate-limited or the
order of writes matters. - For very large files, splitting first with Split File is often better than
one long loop.
Updated 16 days ago
Did this page help you?
