Transform Data step

The Transform Data step reshapes a list without writing a script. You build a pipeline of operations (filter, map, sort, group and so on), and each one works on the result of the one before.

Set it up

In the app	`data` field	What it does
Source	`source`	The variable or dot path holding the list, like `orders.items` or `order.line_items`. A variable holding JSON text is parsed for you.
Operations	`operations`	The steps of the pipeline, in order. Select Add operation, and drag to reorder.
Response variable	`response_variable`	Saves the result.

Operations

Operation	`op`	Settings	Result
Filter	`filter`	Field, Operator, Value	Keeps the items that match.
Map	`map`	Fields: one `name: path` per line	New objects with just those fields, renamed.
Pick	`pick`	Keep: one path per line	Each item with only those paths. A nested path like `customer.email` becomes a flat key `customer.email`.
Sort	`sort`	Field, Descending	The list in order.
Group	`group_by`	Field	An object of lists, keyed by the field's value.
Sum	`sum`	Field	A number.
Count	`count`		A number.
Limit	`limit`	Limit (`n`)	The first `n` items. Always set `n`.
Unique	`unique`	Field	The first item for each value of the field. Leave Field blank to compare whole items.
First	`first`		The first item, or `null` if the list is empty.

Filter operators are Equals, Not equals, Greater than, Greater than or equal, Less than, Less than or equal, Contains, Is in, Exists and Does not exist.

Value can use Liquid, like {{ store_id }}.
Is in takes a comma-separated list (open, pending) or a JSON array.
Contains matches text without caring about case, or checks whether a list contains the value.
Numbers are compared as numbers, so "5" equals 5.
If some items may not have the field at all, add an Exists filter first, so you know exactly which items you're comparing.

Fields and paths use dots for nesting, like customer.email or lines.0.sku.

Example 1: Keep a webhook order's line items

Trigger: Webhook.

Script: Read the order: $.order = JSON.parse($.default);
Transform Data: Keep the line items, Source order.line_items, Response variable lines, with operations:
1. Filter: Field quantity, Greater than, 0
2. Map: Fields
```
sku: sku
qty: quantity
price: price
```

$.lines is now a tidy list like [{"sku": "TEE-RED-M", "qty": 2, "price": "25.00"}], ready for a Loop or a request body.

Example 2: Export open order lines

Trigger: Scheduled.

Shopify GraphQL: Fetch every order page, with Fetch all pages on, Response variable orders.
Transform Data: Keep the open lines, Source orders.items, Response variable open_orders, with operations:
1. Filter: Field displayFulfillmentStatus, Is in, UNFULFILLED, PARTIALLY_FULFILLED
2. Sort: Field createdAt
3. Pick: Keep name, createdAt, displayFulfillmentStatus
File: Write the export, Source open_orders. See File.

Tips

Reach for a Script when it gets clever. Transform Data is great for filter-sort-pick. Calculations across fields are easier in a Script.
Count and Sum end the list. They return a number, so put them last.
Test with Run. The response variable shows in the job's variables, so you can check the result before wiring up the next step.

← All Glue step guides