How wasted-audio/hvcc works
Main Components Overview
The system should be divided into three main layers:
A. Input Layer
• Pure Data (.pd) files input
• Meta-data JSON files
• Search paths for abstractions
B. Compiler Core
• Heavy Language (hv) representation
• Heavy IR (Intermediate Representation)
• Core processing components
C. Output Layer
• C/C++ generation
• Framework-specific generators
• Build system integration
Detailed Component Breakdown
a) Core Components (hvcc/core):
• hv2ir: Converter from Heavy Language to IR
• JSON definitions (heavy.ir.json, heavy.lang.json)
• Parser components
• Graph processing
b) Interpreters (hvcc/interpreters):
• pd2hv: Pure Data to Heavy Language converter
• Connection handling
• Object management
c) Generators (hvcc/generators):
• ir2c: IR to C/C++ converter
• Framework-specific generators:
• Daisy
• DPF
• JavaScript
• OWL
• PD External
• Unity
• Wwise
Data Flow Representation
Input → Interpreter → Core Processing → Generators → Output
[.pd files] → [pd2hv] → [hv2ir] → [ir2c/framework generators] → [Generated Code]
[Input Sources] → [Compiler Core] → [Output Generators]
• ↓ ↓ ↓
[File System] [Processing Pipeline] [Framework Adapters]
• ↓ ↓ ↓
[Configuration] [Optimizations] [Build Integration]
Implementation Notes
• Use standardized UML notation where appropriate
• Version/configuration management touchpoints
• Test integration points
• External dependency connections
This system design should capture both the high-level architecture and the detailed relationships between components while maintaining clarity and usefulness for development purposes.
Remember to include notes about the various supported platforms and frameworks as auxiliary information, as these affect the overall system design and generation paths.