1// Copyright 2012-2015 The Rust Project Developers. See the COPYRIGHT
2// file at the top-level directory of this distribution and at
3// http://rust-lang.org/COPYRIGHT.
4//
5// Licensed under the Apache License, Version 2.0 <LICENSE-APACHE or
6// http://www.apache.org/licenses/LICENSE-2.0> or the MIT license
7// <LICENSE-MIT or http://opensource.org/licenses/MIT>, at your
8// option. This file may not be copied, modified, or distributed
9// except according to those terms.
10
11//! Determine if a `char` is a valid identifier for a parser and/or lexer according to
12//! [Unicode Standard Annex #31](http://www.unicode.org/reports/tr31/) rules.
13//!
14//! ```rust
15//! extern crate unicode_xid;
16//!
17//! use unicode_xid::UnicodeXID;
18//!
19//! fn main() {
20//! let ch = 'a';
21//! println!("Is {} a valid start of an identifier? {}", ch, UnicodeXID::is_xid_start(ch));
22//! }
23//! ```
24//!
25//! # features
26//!
27//! unicode-xid supports a `no_std` feature. This eliminates dependence
28//! on std, and instead uses equivalent functions from core.
29//!
30
31#![forbid(unsafe_code)]
32#![deny(missing_docs)]
33#![doc(
34 html_logo_url = "https://unicode-rs.github.io/unicode-rs_sm.png",
35 html_favicon_url = "https://unicode-rs.github.io/unicode-rs_sm.png"
36)]
37#![no_std]
38#![cfg_attr(feature = "bench", feature(test, unicode_internals))]
39
40#[cfg(test)]
41#[macro_use]
42extern crate std;
43
44#[cfg(feature = "bench")]
45extern crate test;
46
47use tables::derived_property;
48pub use tables::UNICODE_VERSION;
49
50mod tables;
51
52#[cfg(test)]
53mod tests;
54
55/// Methods for determining if a character is a valid identifier character.
56pub trait UnicodeXID {
57 /// Returns whether the specified character satisfies the 'XID_Start'
58 /// Unicode property.
59 ///
60 /// 'XID_Start' is a Unicode Derived Property specified in
61 /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications),
62 /// mostly similar to ID_Start but modified for closure under NFKx.
63 fn is_xid_start(self) -> bool;
64
65 /// Returns whether the specified `char` satisfies the 'XID_Continue'
66 /// Unicode property.
67 ///
68 /// 'XID_Continue' is a Unicode Derived Property specified in
69 /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications),
70 /// mostly similar to 'ID_Continue' but modified for closure under NFKx.
71 fn is_xid_continue(self) -> bool;
72}
73
74impl UnicodeXID for char {
75 #[inline]
76 fn is_xid_start(self) -> bool {
77 // Fast-path for ascii idents
78 ('a' <= self && self <= 'z')
79 || ('A' <= self && self <= 'Z')
80 || (self > '\x7f' && derived_property::XID_Start(self))
81 }
82
83 #[inline]
84 fn is_xid_continue(self) -> bool {
85 // Fast-path for ascii idents
86 ('a' <= self && self <= 'z')
87 || ('A' <= self && self <= 'Z')
88 || ('0' <= self && self <= '9')
89 || self == '_'
90 || (self > '\x7f' && derived_property::XID_Continue(self))
91 }
92}
93