Loop Node

Loop nodes allow you to iterate through each element in an array and apply transformations or operations to them. They work similarly to JavaScript’s map() function, taking an array as input and outputting a processed array.

Overview

Loop nodes create a sub-workflow that executes once for each item in the input array. Each iteration processes one array element and produces a corresponding output element, maintaining the same array structure.

How Loops Work

You can think of loop nodes as JavaScript’s ["x","y"].map() function:

• Input: An array of values
• Processing: Each array element is processed through a sub-workflow
• Output: A new array where each element corresponds to the processed result

Loop nodes contain an embedded sub-workflow consisting of two required nodes:

1. Start Loop Once - The entry point for each loop cycle
2. End Loop Once - The exit point that defines the output of each iteration

Sub-Workflow Structure

Each loop node contains:

Start Loop Once Node

This node provides two variables for each iteration:

• $loopStart.item - The value of the current array element
• $loopStart.index - The index of the current array element (starting from 0)

End Loop Once Node

This node determines what is returned from each iteration. The value exported by this node will become the corresponding element in the output array.

Configure the data to be exported from each iteration. The selected output variable determines the content in the final result array.

Configuration Instructions

Node Name

Set a descriptive name for your loop operation.

Loop Variable

Select the variable containing the array to iterate over. The variable must be:

• Array type (can be parsed as a JSON array)
• For example: ["a", "b", "c"] will iterate 3 times
• Non-array type values will be treated as single-element arrays

Note

The loop variable must be a data type that can be parsed as a JSON array. If it’s not an array type, such as content “a”, it will be treated as [“a”], and the loop will execute only once. Loops execute a maximum of 500 times. If the loop count exceeds 500, the loop will stop.

Loop Interval

Set the delay time between iterations in seconds. Defaults to 0 (no delay).

Tip

Setting an interval greater than 0 adds a delay between each loop iteration, which is useful for:

• Adhering to rate limits when making API calls
• Avoiding system resource overload when processing large datasets
• Debugging workflows by slowing down execution

Output

The loop node outputs $loop1.result - an array containing the results of each iteration.

Usage Examples

1. Text Processing

Process a string array, applying transformations to each string:

• Input: ["hello", "world", "flowai"]
• Sub-workflow: Convert to uppercase
• Output: ["HELLO", "WORLD", "FLOWAI"]

2. Data Enrichment

Enhance array data with additional information:

• Input: [{"id": 1}, {"id": 2}, {"id": 3}]
• Sub-workflow: Get user details for each ID
• Output: [{"id": 1, "name": "John"}, {"id": 2, "name": "Jane"}, ...]

3. API Batch Processing

Make multiple API calls with different parameters:

• Input: ["product1", "product2", "product3"]
• Sub-workflow: Get price for each product
• Output: [99.99, 149.99, 79.99]

Best Practices

1. Performance: Be aware of the loop count limit (maximum 500 iterations)
2. Rate Limiting: Use loop intervals when making external API calls
3. Error Handling: Consider how to handle failures in individual iterations
4. Data Types: Ensure loop variables are properly formatted as arrays
5. Output Structure: Plan the data structure needed by the End Loop Once node

Common Patterns

Simple Value Transformation

Use when you need to transform each array element:

• Configure End Loop Once to output the transformed value
• Maintain the same array structure

Data Aggregation

Use when you need to collect information from each element:

• Configure End Loop Once to output enriched objects
• Combine original data with newly computed data

Filtering with Processing

Combine with conditional nodes to selectively process array elements:

• Use conditionals in the sub-workflow
• Output different values based on conditions