/iOS-app-localizer_GPT

Easily translate iOS apps into multiple languages

Primary LanguagePython

Python Localization Script with GPT Support

This script is designed to help in localizing an iOS application using OpenAI's GPT models. It reads a .xcstrings file containing localization strings and generates translations for specified target languages while preserving the format, placeholders, and comments.

Localization Scripts

Please note: only supports gpt-4-1106-preview, gpt-4o-2024-05-13 and gpt-3.5-turbo-1106 models, which can give back an answer as JSON. I've found that gpt-4-1106-preview gives the best translation quality, so I use it by default everywhere.

Terminal animation

Prerequisites

To run this script, you will need:

  • Python 3.x
  • openai Python package
  • An OpenAI API key

Installation

Before you can run the script, make sure to install the necessary Python packages:

pip3 install openai tiktoken argparse tqdm

Usage

To use the script, run it from the command line with the required arguments.

python3 localize_strings.py --gpt_api_key YOUR_GPT_API_KEY --file PATH_TO_XCSTRINGS_FILE --localize_from SOURCE_LANG_CODES --localize_to TARGET_LANG_CODES

For localize_release_notes you need to install Fastlane. You also need to provide fastlane_api_key_path with path to JSON file:

{
  "key_id": "CQC6F7C12K",
  "issuer_id": "39a1de7c-d01c-41se-e052-532c7c1ja4p1",
  "key": "-----BEGIN PRIVATE KEY-----\nMAGTAgEASBcGRyqG...\n-----END PRIVATE KEY-----",
  "duration": 1200,
  "in_house": false 
}

Available Arguments

  • --gpt_api_key: Your GPT API key (required).
  • --gpt_model: The GPT model you want to use (optional, default is gpt-3.5-turbo-1106).
  • --file: The path to the .xcstrings file you want to localize (required).
  • --out_file: The path to the output .xcstrings file (optional, will overwrite the original file if not provided).
  • --localize_from: A comma-separated list of source language codes (required).
  • --localize_to: A comma-separated list of target language codes (required).
  • --app_description: A short description of your application for better context understanding (optional).
  • --max_input_token_count: Maximum number of tokens for the model (optional).

Example

python3 localize_strings.py --gpt_api_key "your-api-key" --file "./Localizable.xcstrings" --localize_from "en,de" --localize_to "es,fr,it" --app_description "A productivity app for managing tasks."

This command will translate the content of Localizable.xcstrings from English to Spanish, French, and Italian.

Output

The script will generate a new .xcstrings file with the translations added. If the --out_file argument is not provided, the original file will be overwritten.

Notes

  • Ensure that your GPT API key has sufficient quota for processing the translations.
  • The script preserves placeholders such as %@ and %d and attempts to maintain the text length and formatting.
  • Translations that need review are marked as such in the resulting file.

Contributing

Feel free to fork the project and submit pull requests with improvements or report any issues you encounter.

License

This script is open-sourced under the MIT License. See the LICENSE file for more information.

Contact

For any additional questions or comments, please open an issue in the repository.

Happy Localizing! 🌍📲