> For the complete documentation index, see [llms.txt](https://docs.brigge.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.brigge.ai/documentation/brigge-for-transformers/pipelines/pipeline-nodes/split-node.md).
# Split node
## Overview
The *Split* node extracts items from an incoming payload at a given point in the payload structure. Extracted items can be removed from the payload, or batched into separate payloads for onward processing.
{% hint style="info" %}
If you choose to split items into separate payloads, each one is passed to the next step in the pipeline and processed in sequence.
{% endhint %}
A common use case is splitting a payload into single-item batches, so records are subsequently processed one at a time. Consider the example below:
{% code overflow="wrap" lineNumbers="true" %}
```json
{
"timesheets": [
{ "id": 1, "name": "Alice Carter", "days_to_invoice": 3 },
{ "id": 2, "name": "Brian Foster", "days_to_invoice": 7 },
{ "id": 3, "name": "Diana Reyes", "days_to_invoice": 2 },
{ "id": 4, "name": "Marcus Lee", "days_to_invoice": 14 },
{ "id": 5, "name": "Priya Nair", "days_to_invoice": 5 }
]
}
```
{% endcode %}
Here, one payload contains five timesheet records. Using the `split` shape, we can extract each timesheet record into its own payload - so five payloads in all:
{% code overflow="wrap" lineNumbers="true" %}
```json
{ "id": 1, "name": "Alice Carter", "days_to_invoice": 3 }
```
{% endcode %}
{% code overflow="wrap" lineNumbers="true" %}
```json
{ "id": 2, "name": "Brian Foster", "days_to_invoice": 7 }
```
{% endcode %}
{% code overflow="wrap" lineNumbers="true" %}
```json
{ "id": 3, "name": "Diana Reyes", "days_to_invoice": 2 }
```
{% endcode %}
{% code overflow="wrap" lineNumbers="true" %}
```json
{ "id": 4, "name": "Marcus Lee", "days_to_invoice": 14 }
```
{% endcode %}
{% code overflow="wrap" lineNumbers="true" %}
```json
{ "id": 5, "name": "Priya Nair", "days_to_invoice": 5 }
```
{% endcode %}
## Configuring a split node
Configuration options for the `split` node are summarised below:
{% stepper %}
{% step %}
**Schema**
Select the schema associated with the payload structure you expect the `split` node to receive.
{% hint style="info" %}
You don't have to provide a schema, but it's useful for navigating the expected payload structure to set the `split element`.
{% endhint %}
{% endstep %}
{% step %}
**Split element**
The `split element` is the point at which you want to start splitting the payload.
Example
{
"batch_ref": "20260601",
"timesheets": [
{ "id": 1, "name": "Alice Carter", "days_to_invoice": 3 },
{ "id": 2, "name": "Brian Foster", "days_to_invoice": 7 },
{ "id": 3, "name": "Diana Reyes", "days_to_invoice": 2 },
{ "id": 4, "name": "Marcus Lee", "days_to_invoice": 14 },
{ "id": 5, "name": "Priya Nair", "days_to_invoice": 5 }
]
}
Here, choosing to split this payload at the `timesheets` array element will create payloads (of your given batch size) from the array items.
If you selected a schema in the previous step, use the dropdown data path to select the required data element to split - for example:
{% hint style="info" %}
Without a schema, you can enter a dot notation path to the required field manually.
{% endhint %}
{% endstep %}
{% step %}
**Split behaviour**
Define what action to take with extracted items:
| Option | Summary |
|---|
| Keep this | Batch the extracted items into payloads for onward processing. The number of payloads is determined by your specified batch size. |
| Remove this | Remove the extracted items before the payload moves on for processing. |
{% endstep %}
{% step %}
**Batch size**
If you have chosen to keep extracted items, use this option to determine the number of items for each ongoing payload.
Example
{
"batch_ref": "20260601",
"timesheets": [
{ "id": 1, "name": "Alice Carter", "days_to_invoice": 3 },
{ "id": 2, "name": "Brian Foster", "days_to_invoice": 7 },
{ "id": 3, "name": "Diana Reyes", "days_to_invoice": 2 },
{ "id": 4, "name": "Marcus Lee", "days_to_invoice": 14 },
{ "id": 5, "name": "Priya Nair", "days_to_invoice": 5 }
]
}
Here, choosing to split this payload at the `timesheets` array element will extract 5 items.
Setting `batch size` to 5, results in 1 payload containing 5 items:
{% code overflow="wrap" lineNumbers="true" %}
```json
[
{
"id": 1,
"name": "Alice Carter",
"days_to_invoice": 3
},
{
"id": 2,
"name": "Brian Foster",
"days_to_invoice": 7
},
{
"id": 3,
"name": "Diana Reyes",
"days_to_invoice": 2
},
{
"id": 4,
"name": "Marcus Lee",
"days_to_invoice": 14
},
{
"id": 5,
"name": "Priya Nair",
"days_to_invoice": 5
}
]
```
{% endcode %}
Setting `batch size` to 1, results in 5 payloads, each containing 1 item:
{% code overflow="wrap" lineNumbers="true" %}
```json
[
{
"id": 1,
"name": "Alice Carter",
"days_to_invoice": 3
}
]
```
{% endcode %}
{% endstep %}
{% step %}
**Unwrap single array**
When a `batch size` of `1` is set to create single-item payloads from an input array, the default output is to wrap each item in an array. For example:
[
{
"id": 1,
"name": "Alice Carter",
"days_to_invoice": 3
}
]
To remove the array, toggle the `unwrap single array` option to `on`. The resulting payload contains just the extracted element, for example:
{
"id": 1,
"name": "Alice Carter",
"days_to_invoice": 3
}
{% endstep %}
{% step %}
**Parent fields**
Use this option to inject parent field values into each split payload.
{% endstep %}
{% endstepper %}
