Release Date: July 2025

Feature Status: Production Ready

Version: 1.0 Breaking Changes: None

Feature Summary

HyperFlow introduces the Python Transform node, a powerful multi-input/multi-output processing node designed for advanced data routing and transformation. Unlike the template-based Data Transform node with its multiple transform modes, Python Transform is a pure Python execution environment that accepts up to 3 inputs and provides 3 separate outputs with dual-exit flow control, enabling sophisticated data flow patterns, conditional branching, and stateful visual programming workflows.

Core Features

Python Transform Node

• 3 configurable inputs: Mix and match text, JSON, content, segments, and knowledge data types
• 3 independent outputs: Route different results to different downstream nodes for complex workflows
• Dual process flow exits: Conditional branching with exit variable (1 or 2)
• NameSpace data access: Structured inputs provided as objects with dotted attribute syntax as well as regular dict keys
• Smart data mapping: Leverages the enhanced intelligent data type conversion framework
• Substantial Python environment: Access to PIL, matplotlib, numpy, pandas, and more

Multi-Input Processing

• Input flexibility: Each input can be a different data type
• Named access: Reference inputs as variables input1, input2, input3 in code
• Optional inputs: Handle missing inputs gracefully with None checks
• Type intelligence: Automatic conversion based on selected input port and output port types
• NameSpace objects: JSON, Content, Segments, Knowledge inputs have dotted attribute access

Triple Output Routing