opencc

Open Chinese Convert (OpenCC) is a comprehensive open-source project designed for high-quality text conversion between different Chinese character sets and regional writing conventions. It supports conversion between Traditional Chinese, Simplified Chinese, Japanese Shinjitai (new character forms), and specific regional wording preferences found in Mainland China, Taiwan, and Hong Kong. Key features include: - Deterministic, dictionary-based conversion that does not rely on Large Language Models (LLMs), ensuring stable, predictable, and fast offline results. - Decoupled dictionaries and libraries, allowing for custom modifications and extensions. - Strict distinction between simplified-traditional mappings and character variant mappings, adhering to the principle of "separate whenever distinguishable." - Support for regional idiomatic expressions and character variants (e.g., converting "鼠标" to "滑鼠"). - Limited support for conversion between Japanese Shinjitai and Kyujitai (old character forms). The project provides multiple interfaces and tools: - **Libraries**: Available for C++, C, Python, and Node.js, enabling integration into various applications. - **Command Line Interface (CLI)**: A native CLI tool for direct text conversion, segmentation inspection, and ambiguity detection via diagnostic modes (`--segmentation`, `--inspect`, `--ambiguities`). - **Prebuilt Binaries**: Available for Windows, Linux, and macOS through package managers like Homebrew, WinGet, and Debian packages. - **Online Converter**: A web-based interface at opencc.js.org. OpenCC includes a wide range of configuration files for specific conversion paths, such as Simplified to Traditional (Standard, Taiwan, Hong Kong), and cross-regional conversions with phrase adjustments. It also supports experimental plugins like `opencc-jieba` for advanced segmentation. The software is licensed under the Apache License 2.0 and is widely used in projects like input methods (ibus-pinyin, fcitx), dictionaries (GoldenDict), and research databases.