DataLayerGenerator

A compile-time C++20 data layer framework with Python/Jinja2 code generation for embedded and host applications.

The library is header-only. You describe your data model (datapoints, groups, types, enums, structs) in JSON or YAML files, run the generator, and get a fully type-safe datalayer.h header with constinit objects, range-checked access, optional persistence, optional fmt formatters, and optional Python bindings.

Getting Started

• The Basics
  • Overview
  • Architecture
  • C++ Framework API

Model File Reference

• Model File Reference
  • groups.json
  • datapoints.json
  • structs.json
  • enums.json
  • types.json
  • YAML Support

Code Generator

• Code Generator
  • Requirements
  • Invocation
  • Output Files
  • CMake Integration
  • Validation
  • Templates

C++ API Reference

• Helper Types
  • Version
  • FixedString
  • PersistenceType
  • Access Tag Types
  • BaseType
  • CheckResult
  • Helper Utilities
• C++20 Concepts
  • Access Concepts
  • Type Concepts
  • Range Concept
• Enumerations
  • Framework Enumerations
  • Model-Defined Enumerations
• File Persistence (Serialization)
  • Enabling Persistence
  • Version-Aware Upgrade
  • API Reference

Packaging & Integration

• Packaging & Integration
  • Installing the Package from Source
  • Conan Options
  • CMake Integration for Consumers
  • Building the Documentation
  • Cross-Compilation (ARM Cortex-M4)

Examples

• Examples
  • Reading and Writing a Datapoint
  • Range-Checked Alias Types
  • Struct Datapoints
  • Array Datapoints
  • Dispatcher — Runtime-ID Access
  • Printing the Data Layer Structure
  • File Persistence
  • Python Bindings