IBus

IBus
Name	IBus
Developer	Hayasi Masahiro; GNOME Foundation; various contributors
Initial release	2003
Programming language	C, C++
Operating system	Linux, BSD, Unix-like
License	GNU Lesser General Public License

IBus is an input method framework for Unix-like operating systems that provides a modular infrastructure for inputting complex scripts, non-Latin alphabets, and language-specific text composition. It mediates between applications and input method engines, handling keystroke processing, candidate selection, and context-aware composition for writing systems such as Chinese language, Japanese language, Korean language, Hindi language, Arabic language, and many Cyrillic script-based languages. The project interacts with desktop environments, display servers, and accessibility stacks to deliver consistent input behavior across GNOME, KDE Plasma, X.Org Server, Wayland, and terminal applications.

Overview

IBus is a daemon-oriented architecture that separates input method logic from client applications, enabling runtime loading of input method engines (IMEs) and flexible configuration. It uses inter-process communication to present a unified input method API to toolkits like GTK+, Qt (software), FLTK, and Electron (software framework), and integrates with display servers such as X.Org Server and Wayland (display server protocol). The framework supports features like on-the-fly language switching, candidate window rendering, and composition buffering, facilitating input for scripts including Han characters, Hiragana, Katakana, Devanagari, Arabic script, Hebrew language, and Thai language.

History

Work on the software began in the early 2000s in response to limitations in existing input frameworks and the need for a cross-desktop solution compatible with X.Org Server and emerging desktop environments like GNOME and KDE. Early adopters included distributions such as Debian, Ubuntu, Fedora, and openSUSE, which packaged the framework alongside internationalization stacks. Over time, contributions from individual maintainers, corporate contributors, and foundations like the GNOME Foundation and projects such as Red Hat and Intel Corporation expanded support, added engines for Pinyin and Mozc, and addressed integration with upcoming systems like Wayland (display server protocol) and accessibility tools including Orca (screen reader). The project evolved through revisions to DBus-based APIs, transitions in toolkit support, and collaborations with input engine projects such as fcitx, Mozc (project), and SCIM developers.

Architecture and Components

The system is composed of a central daemon, client libraries, input method engines, and frontends. The daemon exposes services over D-Bus to manage IME lifecycles and process key events, while client-side libraries for GTK+ and Qt (software) provide integration points for applications. Engines implement language-specific algorithms—examples include statistical models like n-gram language models, rule-based converters, and dictionary lookup systems derived from resources such as CC-CEDICT and Unihan. Frontends and UI modules render candidate lists and composition windows and may use libraries like Pango, Cairo, and Fontconfig for text shaping and font fallback. Session management ties into desktop components such as GDM (GNOME Display Manager) and systemd user services.

Input Method Engines and Supported Languages

A wide ecosystem of engines provides support for scripts and languages. Prominent engines include Pinyin-based engines for Mandarin Chinese, Wubi layout engines, Mozc (project) for Japanese language derived from a major search company’s open-source variant, Hangul engines for Korean language, and engines for South Asian scripts like Devanagari supporting Hindi language and Nepali language. Other engines implement input for Arabic language, Hebrew language, Thai language, Vietnamese language, and various Cyrillic script languages such as Russian language, Ukrainian language, and Bulgarian language. Community contributions have enabled specialized input methods for minority languages and scholarly transcription systems including IPA-based tools and input for Sanskrit and historic scripts registered in Unicode.

Configuration and Usage

Configuration can be performed through graphical utilities bundled with desktop environments—examples include control panels in GNOME and KDE Plasma—or via command-line tools and configuration files stored in user XDG directories. Users select engines, switch keyboard layouts, and define triggers and hotkeys for composition. Integration points include per-application configuration for terminal emulators like GNOME Terminal, Konsole, and text editors such as Vim and Emacs. Accessibility settings interact with screen readers like Orca (screen reader) to announce candidate lists and composition state. Advanced users script behavior using D-Bus interfaces and interact with engine internals for custom dictionary loading and personalization.

Integration and Platform Support

The framework supports multiple display server protocols and desktop environments, working on Linux kernel-based systems, BSD variants such as FreeBSD, and other Unix-like environments. Client-side integration is available for GTK+ and Qt (software) toolkits, enabling support in applications such as Firefox, Chromium (web browser), LibreOffice, Evolution (software), and Thunderbird. Terminal support relies on compatibility layers and toolkit-specific input handling. The project aligns with accessibility and internationalization standards like ICU and Unicode to ensure proper text handling and script rendering across platforms.

Development and Community

Development occurs in public repositories with contributors from distributions, companies, and independent developers. The community interacts through mailing lists, issue trackers, and code hosting platforms used by projects such as GNOME, Debian, Ubuntu, Fedora, and openSUSE. Contribution areas include engine development, API stabilization, documentation, and testing against internationalization suites from organizations like Unicode Consortium and localization projects coordinated with initiatives such as Transifex and Launchpad. The project collaborates with related input frameworks and engine projects including fcitx, Mozc (project), and SCIM to share best practices and interoperability work.

Category:Input methods