Drow Translator Azure Function

An Azure Function that translates text between English (Common) and the Drow language. This project is inspired by a 2003 Perl script by Brian Sidharta, focusing on translating English to Drow and vice versa using an SQLite database for the word list.

Features

• Translate text from English to Drow and from Drow to English.
• Handles plurals, possessives, and contractions.
• Utilizes an SQLite database for efficient word lookup.
• Web UI served directly from the function.
• REST API with plain-text and JSON endpoints.
• Returns plain text responses for easy integration with applications like Second Life scripts.
• Configurable for any constructed language — all language-specific settings live in a single Data/language.json file.

Prerequisites

• .NET SDK (version 10 or later)
• Azure Functions Core Tools (for local development and testing)
• Python 3 (for dictionary maintenance scripts only)
• NuGet Packages (restored automatically via dotnet restore):
  • Microsoft.Azure.Functions.Worker and related extensions
  • Microsoft.Data.Sqlite (SQLite database access)
  • Humanizer.Core (pluralization and singularization)
  • Microsoft.Azure.Functions.Worker.Extensions.OpenApi (Swagger / OpenAPI docs for isolated worker)

Installation

1. Clone the Repository

git clone https://github.com/auroris/DrowTranslatascan.git
cd DrowTranslatascan

2. Install Dependencies

dotnet restore

3. Configure Local Settings

Create a local.settings.json in the project root (this file is not committed to source control):

{
  "IsEncrypted": false,
  "Values": {
    "AzureWebJobsStorage": "UseDevelopmentStorage=true",
    "FUNCTIONS_WORKER_RUNTIME": "dotnet-isolated"
  },
  "ConnectionStrings": {}
}

4. Build the Project

dotnet build

5. Run the Function Locally

func start

The function will be available at http://localhost:7071.

Usage

Web UI

Open http://localhost:7071/api/Home (or just http://localhost:7071/ — it redirects) in a browser for the dark-fantasy themed translator UI.

API Endpoints

Plain-text translation

GET|POST /api/Translate?text={text}&lang={Drow|Common}

Returns translated text as text/plain.

Example:

curl "http://localhost:7071/api/Translate?text=Hello,+how+are+you?&lang=Drow"

Response:

Vendui, lu'oh ph' dos?

JSON translation

POST /api/TranslateJson
Content-Type: application/json

{"text": "Hello, how are you?", "lang": "Drow"}

Response:

{
  "translation": "Vendui, lu'oh ph' dos?",
  "originalText": "Hello, how are you?",
  "targetLanguage": "Drow"
}

Parameters

• text: The text to translate.
• lang: Target language — Drow (English → Drow) or Common (Drow → English).

API Documentation

With the function running locally, visit:

• Swagger UI: http://localhost:7071/swagger/index.html
• OpenAPI spec: http://localhost:7071/swagger/v1/swagger.json

Running Tests

cd drowtest
dotnet run

Build Your Own Conlang Translator

Want to adapt this project for your own constructed language? Everything language-specific is driven by Data/language.json — no C# changes required for most conlangs.

• TUTORIAL.md — Step-by-step guide: fork the project, edit the config, provide a dictionary, and run. Includes a worked example building a "Sylvan" translator from scratch.
• LLM_CONLANG_GUIDE.md — Instruction file for LLM assistants (Claude, ChatGPT, etc.) helping you design your conlang's rules, build a dictionary, and configure the translator. Feed this file to your AI assistant to get structured help with phonology, pluralization, cipher tables, and vocabulary generation.

Dictionary Maintenance

The word list lives in Data/drow_dictionary.csv and is compiled into Data/drow_dictionary.db (SQLite). See scripts/README.md for full instructions on finding missing words and adding new entries.

Quick reference:

# Find common English words not yet in the dictionary
python3 scripts/missing_words.py

# Add new entries and rebuild the database
python3 scripts/patch_missing_words.py

Deployment

Deploy the Azure Function to your Azure account using your preferred deployment method (e.g., Visual Studio publish, Azure CLI, or GitHub Actions).

Ensure that the Data folder and the drow_dictionary.db file are included in your deployment package.

Disabling the Azure Functions Default Homepage

By default, Azure Functions shows its own homepage at /. To disable it so that the / → /api/Home redirect works correctly, set the following application setting in the Azure Portal:

Setting	Value
AzureWebJobsDisableHomepage	true

This can be configured under Configuration → Application settings in your Function App's Azure Portal page.

Repository Structure

DrowTranslatascan/
├── Data/                        # Dictionary (CSV + SQLite DB) and language.json config
├── analysis/                    # Corpus analysis scripts and output (phonotactics)
├── drowtest/                    # Test runner project
├── lsl/                         # Second Life LSL script example
├── scripts/                     # Dictionary maintenance scripts (see scripts/README.md)
├── tel'mithrim/                 # Original unmodified Tel'Mithrim word list and license
├── AlgorithmicConverter.cs      # Fallback algorithmic conversion (config-driven)
├── HomeFunction.cs              # Web UI (themed HTML, config-driven)
├── LanguageConfig.cs            # Language configuration model (deserialized from language.json)
├── Program.cs                   # Host configuration and startup
├── ProgramState.cs              # Shared program state (DB path, language config)
├── Translate.cs                 # Translation logic and HTTP trigger functions
├── TranslateModels.cs           # Request/response model types
├── TUTORIAL.md                  # Guide: adapt this project for your own conlang
└── LLM_CONLANG_GUIDE.md         # Instructions for LLM assistants helping build a conlang

License

Code

This project is licensed under the MIT License.

Word List

The original Drow word list is by Tel'Mithrim/Brian Sidharta and is licensed under Tel'Mithrim/Brian Sidharta's Software License Version 1.0. The unmodified original is preserved in tel'mithrim/. The version in Data/ is a derivative work distributed under the same license terms.

Acknowledgments

• Tel'Mithrim/The Grey Company and Brian Sidharta: For developing the original Drow dictionary and Perl script. See tel'mithrim/WORDLIST_LICENSE.md.
• Humanizer Library: For providing pluralization and singularization utilities.

Contributing

Contributions are welcome! Please open an issue or submit a pull request for any improvements or additions.

Contact Information

For questions or support, please open an issue.