Loops

Most real records contain lists — an order has line items, an invoice has rows. A loop maps each item in a source list to an item in a target list.

How a loop is defined

Setting	What it does
From	The source list to iterate over (e.g. line_items).
Target	The target list to produce (e.g. items[]).
Filter	Optional conditions to decide which items to include.
Fields	The field mappings applied to each item in the list.

The field mappings inside a loop work exactly like top-level field mappings — same modes, transforms, and type casting — but they run once per item.

Filtering items

A filter lets you include only some items. You add one or more conditions and join them with AND or OR:

• AND — an item is included only if all conditions match.
• OR — an item is included if any condition matches.

For example, include only line items whose quantity is greater than zero.

Example

Source:

{
  "line_items": [
    { "sku": "ABC-1", "quantity": 2, "price": "19.99" },
    { "sku": "XYZ-9", "quantity": 0, "price": "5.00" }
  ]
}

Loop: line_items → items[], filter quantity > 0, mapping sku → itemId and casting price to a number.

Output:

{
  "items": [
    { "itemId": "ABC-1", "quantity": 2, "price": 19.99 }
  ]
}

The second line item is dropped because it failed the filter.

tip

Build the loop's per-item fields the same way you'd build a single record, then let the loop apply them across the whole list.