Data Processing Nodes

Data Processing nodes transform upstream raw content into the format required by downstream nodes. This guide covers all 5 Data Processing nodes.

These nodes do not automatically change the source variable. Except for Variable Manager, save the processed result to a new output variable.

Data Processing nodes

Variable Manager

Creates or updates one or more variables in the middle of a workflow, using the same variable editor as Start Flow.

Variable Manager configuration

No.	Setting	Description
1	Variable name	Enter a valid, unique variable name
2	Type	Select a data type such as string, number, boolean, array, or object
3	Default value	Enter a value that matches the selected type
4	Note	Optionally describe the variable purpose
5	Add variable	Add another variable to the current node

A later definition overrides an earlier variable with the same name. Values created here are available only on downstream reachable paths. See Variables and Data for names, types, and JSON rules.

Extract Text

Uses a regular expression to extract content from a selected variable and saves the result.

Extract Text configuration

No.	Setting	Description
1	Data	Select the upstream variable containing the source text
2	Extraction rule	Enter a valid regular expression or use a common-rule shortcut
3	Ignore case	Enable case-insensitive matching
4	First match only	Enable to return the first match; disable to return all matches
5	Save to	Enter a valid output variable name

For this source text:

Order: ZY-20260724, Status: Complete

Use this regular expression to extract the order ID:

ZY-\d+

Regular Expressions

• Regular Expressions Tutorial
• Common Regular Expressions

Parse JSON

Parses JSON text from a selected variable into an object and saves the result.

Parse JSON configuration

No.	Setting	Description
1	Source variable	Select the variable containing valid JSON text
2	Save to	Enter the output variable for the parsed object

The input must be valid JSON. For example:

{"name":"Sample product","price":99.9,"in_stock":true}

If the input contains a prefix, extra log text, or incomplete JSON, clean it with Extract Text before parsing.

Extract Field

Reads a field from an array, object, or unknown-type variable and saves it.

Extract Field configuration

No.	Setting	Description
1	Data	Select an array, object, or unknown-type input variable
2	Key	Enter an object key or array index; variables are supported
3	Save to	Enter a valid output variable name

Object example:

{"username":"demo@example.com","roles":["admin","editor"]}

Enter username to get demo@example.com. Array indexes are not the same as the user-facing natural order used by element selectors, so follow the node hint and verify with test data.

Extract Random Field

Selects a random field from an array, object, or unknown-type variable and saves it.

Extract Random Field configuration

No.	Setting	Description
1	Data	Select an array, object, or unknown-type input variable
2	Save to	Enter a valid output variable name

An empty input cannot produce a useful result, so test for existence with IF or guarantee data upstream. For repeatable or ordered behavior, use Extract Field or For Data.

Processing Examples

Parse an API Response

Listen for Response → Parse JSON → Extract Field → Save to Excel

1. Save the endpoint response to response_text.
2. Parse it as JSON and save it to response_data.
3. Extract the required field from response_data.
4. Write the final variable to Excel.

Write the transformed result to Excel

Read an ID from Text

Get Clipboard Content → Extract Text → Enter Text

1. Read the clipboard text.
2. Extract the business ID with a regular expression.
3. Enter the output variable in the page form.

Select a Random Material

1. Create an array variable named messages in Start Flow.
2. Add Extract Random Field and select messages as the input.
3. Save the result to current_message.
4. Reference ${current_message} in Enter Text.

Common Issues

JSON Parsing Fails

Check for straight double quotes, valid braces, and commas, with no prefix or suffix outside the JSON. Empty or truncated responses also fail parsing.

Extract Field Returns Empty

Inspect the actual upstream structure and check object key capitalization or the array index. Object keys are case-sensitive.

A Variable Still Has Its Old Value

Confirm that the consumer is downstream from Variable Manager and no separate branch bypasses the update. A later definition applies only after that node executes.

Random Extraction Is Unstable

This is expected. Use Extract Field for deterministic debugging or For Data to process every item.

Related Guides

• Variables and Data: Review variable types, naming, and overrides.
• Data Collection and Storage Nodes: Collect page, request, and file data.
• Flow Control Nodes: Process data with conditions and loops.