DSM

The Digital Substrate Model (DSM) is a small declarative language for describing data models — namespaces, concepts, structures, attachments, function pools. It is the modeling layer of the dsviper ecosystem and the entry point of the code-generation pipeline.

DSM is standalone: authoring a .dsm file does not require Kibo, the Viper C++ runtime, or dsviper. You can use it as a typed contract between teams, or feed it into code generation when typed code is needed. See Value Chains for where DSM sits in the broader picture.

Place in the ecosystem

• Depends on — DSM is the root of the toolchain value chain.

• Consumed by — Kibo and dsviper

• Source repository — digital-substrate/dsm. Carries the ANTLR4 grammar (DSM.g4) and the canonical JSON wire format — producer-neutral, implemented by Viper C++, Kibo, and the IDE plugins.

• Related repositories —

  • digital-substrate/dsm-samples — reference DSM models (Tuto, Ge, Re).

  • digital-substrate/dsm-vscode — VS Code plugin.

  • digital-substrate/dsm-jetbrains — JetBrains plugin.

  Install instructions: see IDE Integration.

Quickstart

A minimal DSM file, exercising the five fundamental notions (namespace, concept, key, document, attachment):

namespace Tuto {f529bc42-0618-4f54-a3fb-d55f95c5ad03} {

    concept User;

    struct Login {
        string nickname;
        string password;
    };

    attachment<User, Login> login;

};

Validate the syntax:

python3 tools/dsm_util.py check model.dsm

Generate a typed package with Kibo: a kibo.toml beside the model states the generation once, and kibo-project renders it with the kibo-template-viper pack:

[project]
definitions = "model.dsm"
infrastructure = "model"

[generator]
templates = "2"

[target.python]
features = ["Base"]
output = "."

python3 kibo_project.py generate

The same file declares C++ or TypeScript targets for the Viper C++ runtime and the Node binding. See Templated Features for the features of each target, and Using your generated SDK for the generated surface.

To use a model without code generation — load it through dsviper’s dynamic API at runtime — see DSM Processing.

Topics

• Introduction to DSM
  • The Five Fundamental Notions
  • Data Definitions Extracted from Code
  • The Data Model is a Set of Sealed Definitions
  • A Complete Data Model
• Types and Structures
  • Primitive Types
  • Enumerations
  • Structures
  • Mathematical Types
  • Generic Types
  • Special Types
  • Default Values
• Concepts and Hierarchies
  • Basic Concepts
  • Concept Inheritance
  • Key Polymorphism
  • Clubs: Grouping Unrelated Concepts
  • Concept vs Club
  • RuntimeId Generation
• Attachments
  • Basic Attachments
  • Recommended Pattern: Multiple Attachments
  • Single-Attachment Pattern
  • Schema Evolution: Adding Features
  • Attachment Types
  • Optional vs Required
  • Design Guidelines
• Function Pools
  • Two Types of Function Pools
  • Function Pool (Pure Functions)
  • Attachment Function Pool
  • Graph Editor Pool Examples
  • Raptor Editor Pool Example
  • Function Documentation
  • Parameter Types
  • XArray Operations
  • Pool Namespacing
  • Design Guidelines
• IDE Integration
  • VS Code Extension
  • Build Task Configuration
  • Optional: Theme Customization
  • JetBrains Plugin
  • Development Workflow
  • What’s Next
• DSM Samples
  • Pattern Comparison

Status

Part of DevKit 1.2.x (LTS, feature-locked).