Generate L10n - VSCode Extension

Automate localization of Dart/Flutter projects using AI/LLMs directly from VSCode.

Download video for better quality

Table of Contents

• Overview
• Features
• Installation
• Usage
• Configuration
• Commands
• Tree View
• Technical Details
• Troubleshooting
• Contributing
• License

Overview

generate-l10n is a VSCode extension designed to simplify the localization process for Dart/Flutter projects. By leveraging LLM models (Mistral, OpenAI, Google), it automates the tedious task of extracting strings to .arb files and updating .dart code with the correct localization keys.

Whether you want to process entire files or just a specific snippet of text, the extension handles the context, key generation, and translation for you.

Features

• Smart Selection (New!): Highlight text in a Dart file and use "Quick Fix" (lightbulb) to localize it instantly.
• Contextual Intelligence: The AI detects the source language and checks existing .arb files to reuse existing keys or create new ones consistently.
• Batch Processing: Interactive tree view to select and process multiple files at once.
• Multi-Provider: Support for Mistral, OpenAI, and Google Gemini models.
• Automated Workflow: Optionally executes flutter gen-l10n immediately after modification.
• Native Integration: Seamlessly integrated into the VSCode Activity Bar and Command Palette.

Installation

1. Open VSCode.
2. Navigate to the Extensions sidebar.
3. Search for generate-l10n.
4. Click Install.
5. Once installed, the L10n Generator icon appears in the activity bar.

No additional setup is required beyond the API key.

Usage

1. Localize Selected Text (Quick Action)

• Highlight a string in your .dart file.
• Click the Lightbulb 💡 icon, right-click or press Cmd+. / Ctrl+..
• Select "auto-l10n" (replaces text with key) or "auto-l10n & execute" (replaces text + runs flutter gen-l10n).
• The AI will automatically update all your .arb files with the translated values.

2. Localize Entire Files (Batch)

• Click the L10n Generator icon in the activity bar.
• Expand the /lib folder tree and check the files you want to process.
• Press the Play ▶ button.
• The extension will refactor the files to use AppLocalizations.

Configuration

Note: The API key is required for the extension to function.

The model choice is crucial. Most of the tests were carried out using mistral-large-2411 and gemini-2.5-flash. Smaller or older models might not get the work done perfectly.

Auto-detection

The extension automatically detects your Flutter project name from pubspec.yaml when activated. If the packageName setting is empty, it will be populated automatically and stored in your workspace settings.

You can manually re-detect the project name emptying it in the config tab.

Commands & UI

Command Palette (Ctrl+Shift+P)

• Process Selected Files: Starts batch localization.
• Configure Extension: Quick access to settings.
• Refresh View: Updates the file tree structure.

Tree View

• Displays your /lib hierarchy (excluding l10n folder and non-Dart files).
• Provides a visual way to manage bulk localization tasks.

Technical Details

The extension follows a modular Object-Oriented Programming (OOP) architecture for high maintainability:

• Core Logic: Separated into L10nProcessor for file handling and LLMService for AI logic.
• REST Integration: Migrated to a clean Fetch API implementation for LLM requests with strict JSON output parsing.
• Providers: Uses CodeActionProvider to inject localization commands directly into the VSCode editor UI.
• Robust Parsing: Handles LLM responses with built-in sanitization (markdown block removal, JSON validation).
• Automation: Built-in utility to trigger flutter gen-l10n via the VSCode Terminal API.

Troubleshooting

• No workspace open: Open a Flutter project folder in VSCode.
• Missing API key: Set it via Settings (generateL10n.apiKey) or Configure Extension command.
• No files checked: Ensure you select at least one Dart file.
• Flutter gen-l10n fails: Verify Flutter SDK is installed and added to PATH.
• Language Detection: If the AI misidentifies the source language, ensure your existing .arb files follow the standard naming convention (e.g., app_en.arb, app_fr.arb).
• Terminal Errors: If flutter gen-l10n fails, ensure Flutter is in your system PATH and your l10n.yaml is correctly configured.
• API Limits: Check your LLM provider dashboard if you receive empty responses or "429 Too Many Requests" errors.

Contributing

Contributions are welcome!

1. Fork the repository.
2. Install dependencies: npm install.
3. Run tests: npm run test:unit.
4. Submit a PR describing your changes (logic refactoring, new providers, etc.).

License

MIT License © 2026