Revision control
Copy as Markdown
Other Tools
// Copyright 2012-2015 The Rust Project Developers. See the COPYRIGHT
// file at the top-level directory of this distribution and at
//
// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
// option. This file may not be copied, modified, or distributed
// except according to those terms.
//! Determine displayed width of `char` and `str` types according to
//! rules.
//!
//! ```rust
//! extern crate unicode_width;
//!
//! use unicode_width::UnicodeWidthStr;
//!
//! fn main() {
//! let teststr = "Ｈｅｌｌｏ, ｗｏｒｌｄ!";
//! let width = UnicodeWidthStr::width(teststr);
//! println!("{}", teststr);
//! println!("The above string is {} columns wide.", width);
//! let width = teststr.width_cjk();
//! println!("The above string is {} columns wide (CJK).", width);
//! }
//! ```
//!
//! # features
//!
//! unicode-width supports a `no_std` feature. This eliminates dependence
//! on std, and instead uses equivalent functions from core.
//!
//! # crates.io
//!
//! You can use this package in your project by adding the following
//! to your `Cargo.toml`:
//!
//! ```toml
//! [dependencies]
//! unicode-width = "0.1.5"
//! ```
#![deny(missing_docs, unsafe_code)]
#![cfg_attr(feature = "bench", feature(test))]
#![no_std]
#[cfg(test)]
#[macro_use]
extern crate std;
#[cfg(feature = "bench")]
extern crate test;
use tables::charwidth as cw;
pub use tables::UNICODE_VERSION;
use core::ops::Add;
mod tables;
#[cfg(test)]
mod tests;
/// Methods for determining displayed width of Unicode characters.
pub trait UnicodeWidthChar {
/// Returns the character's displayed width in columns, or `None` if the
/// character is a control character other than `'\x00'`.
///
/// This function treats characters in the Ambiguous category according
/// as 1 column wide. This is consistent with the recommendations for non-CJK
/// contexts, or when the context cannot be reliably determined.
fn width(self) -> Option<usize>;
/// Returns the character's displayed width in columns, or `None` if the
/// character is a control character other than `'\x00'`.
///
/// This function treats characters in the Ambiguous category according
/// as 2 columns wide. This is consistent with the recommendations for
/// CJK contexts.
fn width_cjk(self) -> Option<usize>;
}
impl UnicodeWidthChar for char {
#[inline]
fn width(self) -> Option<usize> { cw::width(self, false) }
#[inline]
fn width_cjk(self) -> Option<usize> { cw::width(self, true) }
}
/// Methods for determining displayed width of Unicode strings.
pub trait UnicodeWidthStr {
/// Returns the string's displayed width in columns.
///
/// Control characters are treated as having zero width.
///
/// This function treats characters in the Ambiguous category according
/// as 1 column wide. This is consistent with the recommendations for
/// non-CJK contexts, or when the context cannot be reliably determined.
fn width<'a>(&'a self) -> usize;
/// Returns the string's displayed width in columns.
///
/// Control characters are treated as having zero width.
///
/// This function treats characters in the Ambiguous category according
/// as 2 column wide. This is consistent with the recommendations for
/// CJK contexts.
fn width_cjk<'a>(&'a self) -> usize;
}
impl UnicodeWidthStr for str {
#[inline]
fn width(&self) -> usize {
self.chars().map(|c| cw::width(c, false).unwrap_or(0)).fold(0, Add::add)
}
#[inline]
fn width_cjk(&self) -> usize {
self.chars().map(|c| cw::width(c, true).unwrap_or(0)).fold(0, Add::add)
}
}