Weighted-Draw-System

A Unity 6 project with two reusable, domain-agnostic modules:

• ProbabilityEngine: condition-aware weighted selection
• SchemaImporter: schema-driven CSV/JSON data import

This repository is structured as a Unity project.

What This Repo Contains

• Assets/Scripts/ProbabilityEngine: generic runtime selection engine
• Assets/Scripts/SchemaImporter: schema-based parsing/import pipeline
• Assets/Tests/EditMode/ProbabilityEngine: edit mode tests for probability logic
• Assets/Tests/EditMode/SchemaImporter: edit mode tests for importer/parser behavior

Core Features

ProbabilityEngine

Use ProbabilityEngine<TState, TValue> when you need to:

• filter options by conditions against runtime state
• select one valid option by weighted randomness
• keep logic generic across gameplay contexts (loot, AI, events, etc.)

Detailed documentation:

• Assets/Scripts/ProbabilityEngine/Docs/ProbabilityEngine.md

SchemaImporter

Use SchemaImporter when you need to:

• import CSV/JSON into typed DataRecord rows
• enforce required fields and type conversion via schema
• parse condition expressions from data files into structured condition objects

Detailed documentation:

• Assets/Scripts/SchemaImporter/Docs/SchemaImporter.md

Start Here

1. Define your data table/file (CSV or JSON).
2. Create and configure a DataSchemaSO asset.
3. Validate imported records and fix warnings.
4. Tune weights/conditions for balancing.

Read first:

• Assets/Scripts/SchemaImporter/Docs/SchemaImporter.md
• Assets/Scripts/ProbabilityEngine/Docs/ProbabilityEngine.md

Then:

1. Integrate SchemaImporter to load and validate data.
2. Map imported records into ProbabilityItem<TState, TValue>.
3. Implement IGameState and custom ICondition<TState> where needed.
4. Evaluate with GetValidChoices(...) and EvaluateRandom(...).

Read first:

• Assets/Scripts/SchemaImporter/Docs/SchemaImporter.md
• Assets/Scripts/ProbabilityEngine/Docs/ProbabilityEngine.md

Minimal Workflow Example

// 1) Import data from schema
List<DataRecord> records = DynamicDataImporter.ImportFromSchema(schema);

// 2) Build randomiser facade (schema-driven condition + weight columns)
var randomiser = new RandomiserSystem(records, schema);

// 3) Evaluate using runtime context
var context = new Dictionary<string, object>
{
    { "playerLevel", 10 },
    { "hasWeapon", 1 }
};

DataRecord selected = randomiser.EvaluateRandom(context);

Project Requirements

• Unity 6 (project currently targets 6000.4 in existing docs)

Testing

Use Unity Test Runner (Edit Mode) for module verification:

• Assets/Tests/EditMode/ProbabilityEngine
• Assets/Tests/EditMode/SchemaImporter

The feature docs also reference the test files that define current expected behavior.

Contributing

When changing behavior:

• update or add tests in Assets/Tests/EditMode/...
• update feature docs in:
  • Assets/Scripts/ProbabilityEngine/Docs/ProbabilityEngine.md
  • Assets/Scripts/SchemaImporter/Docs/SchemaImporter.md
• keep top-level README.md as a navigation and onboarding guide

License

See LICENSE.