Data Source Nodes

Data source nodes are the starting points of your workflows. They pull data from your connections, existing views, or other flows into your pipeline for transformation and processing.

Entity Node

The Entity node pulls data from a specific table, view, or endpoint in your connection.

Adding an Entity

1. Click Entity in the toolbar, or drag from the Element Panel
2. The Entity Selector opens showing your connections
3. Browse or search for the entity you want
4. Select it and click Add (or double-click to add immediately)

Tip: Hold Cmd/Ctrl while clicking to select multiple entities at once, then add them all together.

Entity Configuration

When you select an Entity node, the configuration panel shows:

Connection Info

• The connection name and type
• Entity name (table/view/endpoint)

Schema

• List of available columns with data types
• Column descriptions if available

Query Options (for database entities)

• Row Limit - Maximum rows to fetch (useful for previewing large tables)
• Custom WHERE - Add conditions to filter at the database level

Working with Entity Schemas

When you add an entity, FactoryThread loads its schema (columns and types). This schema:

• Propagates downstream - Connected nodes know what fields are available
• Updates automatically - If you refresh the connection, schema changes are detected
• Shows in preview - Column names and types appear in data previews

Database vs File Entities

Database Entities (SQL Server, PostgreSQL, etc.):

• Pull data with SQL queries
• Support query optimization (row limits, WHERE clauses)
• Can handle very large tables efficiently

File Entities (Excel, CSV):

• Pull entire sheet/file contents
• Schema inferred from headers and data
• Best for smaller datasets

API Entities (REST, OData):

• Call the API endpoint
• Handle pagination automatically
• Pass parameters defined in the connection

View Node

The View node includes an existing view as a data source, allowing you to reuse transformation logic.

Adding a View

1. Click View in the toolbar, or drag from the Element Panel
2. The View Selector opens showing your views
3. Select the view you want to include
4. Click Add

Why Use View Nodes?

Reusability Define a transformation once and use it in multiple workflows:

• Create a view that normalizes customer data
• Include that view in multiple flows
• When the logic changes, update it once

Modularity Break complex logic into manageable pieces:

• Each view handles one transformation concern
• Chain views together for complex pipelines
• Easier to test and debug

Organization Keep workflows clean and understandable:

• Abstract complex transformations behind a named view
• The flow shows "what" happens, views show "how"

View Parameters

Some views accept parameters for dynamic behavior:

1. Select the View node
2. In the configuration panel, you'll see any defined parameters
3. Enter values or map them from upstream data

Call Flow Node

The Call Flow node executes another flow and uses its output as data in your current workflow.

Adding a Call Flow

1. Click Action in the toolbar
2. Select Call Flow
3. Choose the flow to call
4. Configure any input parameters

When to Use Call Flow

Sequential Processing When one flow's output becomes another's input:

• Flow A extracts and cleans data
• Flow B calls Flow A and adds business logic
• Flow B's output is the final result

Shared Processing When multiple flows need the same preprocessing:

• Create a "common processing" flow
• Each consumer flow calls it
• Changes to common logic apply everywhere

Error Isolation To contain failures:

• If the called flow fails, you can handle the error
• The calling flow can decide how to proceed

Call Flow Configuration

Flow Selection Choose which flow to call from your deployed flows.

Input Parameters Map values to the called flow's input parameters:

• Static values
• Fields from upstream nodes
• Expressions

Timeout Set how long to wait for the called flow to complete.

Dynamic Call Flow

For advanced use cases, the Dynamic Call Flow node lets you determine which flow to call at runtime:

1. Add a Dynamic Call Flow node
2. Map a field that contains the flow ID or name
3. The specific flow is determined when the workflow runs

This is useful for:

• Configuration-driven processing
• A/B testing different flows
• Multi-tenant scenarios

Structured Response: Dynamic Call Flow returns { status, response } where:

• status (boolean): true on success, false on failure
• response (array|string): Result data on success, error message on failure

Use a Condition node after Dynamic Call Flow to check ${status} before processing results.

Constant Node

The Constant node provides static data that you define directly in the workflow.

Adding a Constant

1. Drag Constant from the Element Panel
2. Define the schema (columns) you want
3. Add rows of data

Use Cases

Lookup Tables Small reference tables that don't need a database:

• Status mappings
• Code translations
• Configuration values

Test Data When building and testing workflows:

• Create sample data to test transformations
• No connection required

Default Values Provide fallback data when other sources are empty.

Configuration

Schema Definition Define columns with:

• Column name
• Data type (text, number, date, boolean)

Row Data Add rows manually in a spreadsheet-like interface.

Best Practices

Choose the Right Data Source

Need	Use
Data from a database/file	Entity node
Reusable transformation	View node
Output from another flow	Call Flow node
Static reference data	Constant node

Optimize Data Loading

For large tables:

• Use row limits during development
• Add WHERE conditions to filter at the source
• Consider creating database views for complex queries

For APIs:

• Understand pagination behavior
• Set appropriate timeouts
• Handle rate limits gracefully

Name Data Sources Clearly

Good names describe what data is being pulled:

Bad	Good
Entity1	Customer Orders
View1	Normalized Addresses
CallFlow1	Load Product Catalog

Handle Schema Changes

When underlying schemas change:

1. Connections can be refreshed to pick up changes
2. Downstream nodes may need reconfiguration
3. Preview your workflow to verify data still flows correctly

Troubleshooting

"Connection failed"

If an entity node shows a connection error:

• Check the connection is healthy in Connections
• Verify credentials haven't expired
• Test network connectivity

"Schema not found"

If the schema can't be loaded:

• Refresh the connection to rediscover entities
• Verify the table/endpoint still exists
• Check user permissions on the data source

"Circular dependency"

If using Call Flow or View nodes:

• You can't create cycles (A calls B, B calls A)
• Verify your flow/view references don't loop

Next Steps

• Transform Nodes - Process your data
• Action Nodes - Output results
• Connections Overview - Manage data sources