uroman-rs is a complete rewrite of the original uroman (Universal Romanizer) in Rust. It provides high-speed, accurate romanization for a vast number of languages and writing systems, faithfully reproducing the behavior of the original implementation.

As a reimplementation, it is designed to be a drop-in replacement that passes the original's comprehensive test suite. This means its romanization logic, including its strengths and limitations, is identical to the original. For effective use, we recommend reviewing the original authors' documentation on Reversibility and Known Limitations.

In the same spirit of fidelity, this project respects the licensing of the original uroman software. uroman-rs is licensed under the Apache License 2.0, and includes the original's license as required. For full details, please refer to the License section.

• Performance: Achieves approximately 27x the speed of the standard Python version, making it ideal for large-scale data processing. (See Benchmark)
• Robustness: Fixes several edge-case bugs present in the original implementation, ensuring safer processing of diverse inputs. (See Bug Fixes)
• Self-Contained: A pure Rust implementation with no dependency on external runtimes. It compiles to a single, portable binary.
• High Fidelity: Faithfully reproduces the behavior of the original uroman and passes its test suite.
• Rich Output Formats: Supports multiple output formats, including simple strings (str) and structured JSON data (edges, alts, lattice).
• Versatile: Can be used as a standalone Command-Line Interface (CLI) tool or as a library in your Rust applications.

The uroman-rs project is available as a crate named uroman. You can use it both as a command-line tool and as a library in your Rust projects.

To install the uroman-rs command-line tool, run the following:

cargo install uroman -F cli

This will install the executable as uroman-rs on your system.

Add uroman-rs to your project's Cargo.toml.

cargo add uroman

uroman-rs can be used directly from your terminal.

Show sample conversions: See examples of how various scripts are romanized.

uroman-rs --sample

View all options:

Display the help message for a full list of commands and flags.

uroman-rs --help

Use in REPL mode:

Run uroman-rs without any arguments to process input line by line. Press Ctrl+D to exit.

$ uroman-rs
>> こんにちは、世界！
konnichiha, shijie!
>> ᚺᚨᛚᛚᛟ ᚹᛟᚱᛚᛞ
hallo world
>> (Ctrl+D)

// Uroman::new() is infallible and does not return a `Result`.
let uroman = Uroman::new();

let romanized_string/*: String*/ = uroman.romanize_string::<rom_format::Str>(
    "✨ユーロマン✨",
    Some("jpn"),
).to_string();

assert_eq!(romanized_string, "✨yuuroman✨");
println!("{romanized_string}");

For more advanced examples, please see the examples/ directory.

Performance was measured against the original Python implementation using hyperfine.

• Test File: multi-script.txt from the original uroman repository.
• Environment: Intel Core i7-14700, WSL2 (Ubuntu 24.04)

uroman-rs aims to be not only a faithful reimplementation but also a more robust and accurate one. It handles several edge cases that can cause the original uroman.py script to crash or produce incorrect output.

For example, the original script panics on inputs with incomplete fractional patterns like "百分之" ("percent of..."). This occurs because the script expects a subsequent number but does not safely handle cases where one is not found, leading to a NoneType attribute error. This issue has been reported to the original author (see isi-nlp/uroman#16).

$ uv run uroman.py "百分之多少"
Traceback (most recent call last):
  ...
AttributeError: 'NoneType' object has no attribute 'value'

In contrast, uroman-rs handles this input safely and provides a reasonable fallback romanization:

$ uroman-rs "百分之多少"
baifenzhiduoshao

In addition to improving stability, uroman-rs also corrects certain romanization errors found in the original implementation. A notable example is the handling of the Tibetan letter འ (U+0F60, TIBETAN LETTER -A).

The original script incorrectly romanizes this character, which represents a consonantal onset with the inherent vowel a, by omitting the vowel sound entirely.

$ uv run uroman.py "འ"
'

uroman-rs outputs the linguistically correct romanization, representing both the 'a-chung consonant (as an apostrophe) and the inherent vowel sound.

$ uroman-rs "འ"
'a

The uroman.py implementation fails to distinguish between the low-tone onset འ ('a-chung) and the high-tone onset ཨ ('a-chen) when followed by the vowel ེ (e).

The original script produces the same output for both འེ and ཨེ.

$ uv run uroman.py "ཨེ"
e
$ uv run uroman.py "འེ"
e

In contrast, uroman-rs correctly preserves the distinct consonant onset of འ, maintaining the distinction between the two characters as intended by the script.

$ uroman-rs "ཨེ"
e
$ uroman-rs "འེ"
'e

This project is licensed under the Apache License, Version 2.0.

uroman-rs is a Rust implementation of the original uroman software by Ulf Hermjakob. As such, it is a derivative work and includes the original license notice in the NOTICE file.

Please be aware that any academic publication of projects using uroman-rs should acknowledge the use of the original uroman software as specified in its license. For details, please see the NOTICE file.