Skip to content

Split Task

Overview

The Split task breaks apart string values or array data into multiple pieces. You can split a delimited string into an array, or split an array into separate named fields. It's the opposite of the Merge task.

When to Use

Use the Split task when you need to:

  • Parse delimited strings (CSV, comma-separated lists, etc.)
  • Break apart full names into first/last name fields
  • Split addresses into structured components
  • Convert comma-separated tags into an array
  • Distribute array values into individual named fields

Settings

Key / Column name

Type: Text field | Taggable: Yes

Specifies which part of your data contains the value to split.

  • Use dot notation for nested data (e.g., customer.address)
  • Leave empty for root level data

Action

Required | Type: Select dropdown | Default: Split value

Choose the split operation:

  • Split value: Split a string into an array using a separator
  • Split into column keys: Take an array and create separate fields for each item
  • Split value and split into column keys: Split string, then distribute to separate fields

Column key split method

Type: Select dropdown | Conditional: Only when Action is "Split into column keys" or "Split both"

Choose how to name the new fields:

  • Custom names: Manually specify the name for each field
  • Indexed name: Auto-generate field names using a pattern

Column key names

Type: Grid field | Taggable: Yes | Conditional: Only when using Custom names

Define field names for each split value:

Current value index/key: Position in array (0-based, optional)
New column key name: Name for the field

Indexed key

Type: Text field | Taggable: Yes | Default: {*key*}_{*index*} | Conditional: Only when using Indexed name

Template pattern for generating field names:

  • {*key*} - Original Key/Column name
  • {*index*} - Item index

Example: item_{*index*} creates "item_0", "item_1", "item_2"

Index starts with

Type: Number field | Default: 0 | Conditional: Only when using Indexed name

Starting number for the index in generated field names.

Remove original key(s)?

Type: Switch | Default: No | Conditional: Only when Action is "Split into column keys" or "Split both"

Whether to delete the original field after splitting.

Separator

Type: Select dropdown | Customizable | Conditional: Only when Action is "Split value" or "Split both"

Character(s) or pattern marking where to split:

  • Comma (,), Semicolon (;), Space, Tab ({*tab*}), New line ({*nl*})
  • Regular expression: Custom regex pattern
  • Custom: Any custom separator

Split value column definition

Type: Column type selector | Optional | Conditional: Only when Action is "Split value" or "Split both"

Apply data type transformation to each split value (text trimming, number conversion, date parsing, etc.).

How It Works

Split Value

  1. Retrieves string from Key / Column name
  2. Applies split using separator (regex or standard)
  3. Applies column type transformation (if configured)
  4. Returns array

Split into Column Keys

  1. Retrieves array from Key / Column name
  2. Removes original field (if configured)
  3. Creates new fields using custom names or indexed pattern
  4. Returns updated data

Split Both

Combines both operations: splits string, then distributes to separate fields.

Use Cases

Parse Comma-Separated Tags

Configuration: - Key: tags - Action: Split value - Separator: Comma

Result: "tag1,tag2,tag3" → ["tag1", "tag2", "tag3"]

Split Full Name

Configuration: - Key: full_name - Action: Split both - Separator: Space - Method: Custom names (first_name, last_name) - Remove original: Yes

Result: "John Doe" → {first_name: "John", last_name: "Doe"}

Parse Address

Configuration: - Key: address - Action: Split both - Separator: Comma - Method: Custom names (street, city, state, zip) - Column definition: Text (trim)

Result: Splits "123 Main St, Springfield, IL, 62701" into structured fields

Distribute Array

Configuration: - Key: items - Action: Split into column keys - Method: Indexed name (item_{*index*}) - Start: 1

Result: ["Apple", "Banana"] → {item_1: "Apple", item_2: "Banana"}

Best Practices

Choose the Right Action: Split value for parsing strings, split into keys for distributing arrays, split both for parsing and distributing in one step.

Use Column Types: Apply trimming or type conversion to clean up split values.

Custom vs Indexed Names: Use custom names for known structures (first_name, last_name), indexed names for variable-length data.

Test Regex Patterns: Regex separators use PHP's preg_split() - test with sample data.

Handle Edge Cases: Consider empty values, missing pieces, or inconsistent formatting.

Notes

  • Split value requires a string; split into keys requires an array
  • Empty values result in an empty array
  • Special separators ({*nl*}, {*tab*}) are converted to actual characters
  • Regex patterns must be valid for preg_split()
  • Index in custom names is 0-based (first item is 0)
  • Indexed names create sequential fields from the starting number
  • Column type transformations apply to each individual split value
  • Wildcards in indexed key template are case-sensitive