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 // ANDROID: Unconditionally use std to allow building as a dylib. 42 // #[macro_use] 43 #[cfg(android_dylib)] 44 extern crate std; 45 46 #[cfg(feature = "bench")] 47 extern crate test; 48 49 use tables::derived_property; 50 pub use tables::UNICODE_VERSION; 51 52 mod tables; 53 54 #[cfg(test)] 55 mod tests; 56 57 /// Methods for determining if a character is a valid identifier character. 58 pub trait UnicodeXID { 59 /// Returns whether the specified character satisfies the 'XID_Start' 60 /// Unicode property. 61 /// 62 /// 'XID_Start' is a Unicode Derived Property specified in 63 /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications), 64 /// mostly similar to ID_Start but modified for closure under NFKx. is_xid_start(self) -> bool65 fn is_xid_start(self) -> bool; 66 67 /// Returns whether the specified `char` satisfies the 'XID_Continue' 68 /// Unicode property. 69 /// 70 /// 'XID_Continue' is a Unicode Derived Property specified in 71 /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications), 72 /// mostly similar to 'ID_Continue' but modified for closure under NFKx. is_xid_continue(self) -> bool73 fn is_xid_continue(self) -> bool; 74 } 75 76 impl UnicodeXID for char { 77 #[inline] is_xid_start(self) -> bool78 fn is_xid_start(self) -> bool { 79 // Fast-path for ascii idents 80 ('a' <= self && self <= 'z') 81 || ('A' <= self && self <= 'Z') 82 || (self > '\x7f' && derived_property::XID_Start(self)) 83 } 84 85 #[inline] is_xid_continue(self) -> bool86 fn is_xid_continue(self) -> bool { 87 // Fast-path for ascii idents 88 ('a' <= self && self <= 'z') 89 || ('A' <= self && self <= 'Z') 90 || ('0' <= self && self <= '9') 91 || self == '_' 92 || (self > '\x7f' && derived_property::XID_Continue(self)) 93 } 94 } 95