src/libunicode/char.rs

   1 // Copyright 2012-2014 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 //! Character manipulation (`char` type, Unicode Scalar Value)
  12 //!
  13 //! This module provides the `CharExt` trait, as well as its
  14 //! implementation for the primitive `char` type, in order to allow
  15 //! basic character manipulation.
  16 //!
  17 //! A `char` actually represents a
  18 //! *[Unicode Scalar
  19 //! Value](http://www.unicode.org/glossary/#unicode_scalar_value)*, as it can
  20 //! contain any Unicode code point except high-surrogate and low-surrogate code
  21 //! points.
  22 //!
  23 //! As such, only values in the ranges \[0x0,0xD7FF\] and \[0xE000,0x10FFFF\]
  24 //! (inclusive) are allowed. A `char` can always be safely cast to a `u32`;
  25 //! however the converse is not always true due to the above range limits
  26 //! and, as such, should be performed via the `from_u32` function.
  27
  28 #![stable(feature = "rust1", since = "1.0.0")]
  29 #![doc(primitive = "char")]
  30
  31 use core::char::CharExt as C;
  32 use core::option::Option::{self, Some};
  33 use core::iter::Iterator;
  34 use tables::{derived_property, property, general_category, conversions, charwidth};
  35
  36 // stable reexports
  37 pub use core::char::{MAX, from_u32, from_digit, EscapeUnicode, EscapeDefault};
  38
  39 // unstable reexports
  40 pub use normalize::{decompose_canonical, decompose_compatible, compose};
  41 pub use tables::normalization::canonical_combining_class;
  42 pub use tables::UNICODE_VERSION;
  43
  44 /// An iterator over the lowercase mapping of a given character, returned from
  45 /// the [`to_lowercase` method](../primitive.char.html#method.to_lowercase) on
  46 /// characters.
  47 #[stable(feature = "rust1", since = "1.0.0")]
  48 pub struct ToLowercase(Option<char>);
  49
  50 #[stable(feature = "rust1", since = "1.0.0")]
  51 impl Iterator for ToLowercase {
  52     type Item = char;
  53     fn next(&mut self) -> Option<char> { self.0.take() }
  54 }
  55
  56 /// An iterator over the uppercase mapping of a given character, returned from
  57 /// the [`to_uppercase` method](../primitive.char.html#method.to_uppercase) on
  58 /// characters.
  59 #[stable(feature = "rust1", since = "1.0.0")]
  60 pub struct ToUppercase(Option<char>);
  61
  62 #[stable(feature = "rust1", since = "1.0.0")]
  63 impl Iterator for ToUppercase {
  64     type Item = char;
  65     fn next(&mut self) -> Option<char> { self.0.take() }
  66 }
  67
  68 #[stable(feature = "rust1", since = "1.0.0")]
  69 #[lang = "char"]
  70 impl char {
  71     /// Checks if a `char` parses as a numeric digit in the given radix.
  72     ///
  73     /// Compared to `is_numeric()`, this function only recognizes the characters
  74     /// `0-9`, `a-z` and `A-Z`.
  75     ///
  76     /// # Return value
  77     ///
  78     /// Returns `true` if `c` is a valid digit under `radix`, and `false`
  79     /// otherwise.
  80     ///
  81     /// # Panics
  82     ///
  83     /// Panics if given a radix > 36.
  84     ///
  85     /// # Examples
  86     ///
  87     /// ```
  88     /// let c = '1';
  89     ///
  90     /// assert!(c.is_digit(10));
  91     ///
  92     /// assert!('f'.is_digit(16));
  93     /// ```
  94     #[stable(feature = "rust1", since = "1.0.0")]
  95     #[inline]
  96     pub fn is_digit(self, radix: u32) -> bool { C::is_digit(self, radix) }
  97
  98     /// Converts a character to the corresponding digit.
  99     ///
 100     /// # Return value
 101     ///
 102     /// If `c` is between '0' and '9', the corresponding value between 0 and
 103     /// 9. If `c` is 'a' or 'A', 10. If `c` is 'b' or 'B', 11, etc. Returns
 104     /// none if the character does not refer to a digit in the given radix.
 105     ///
 106     /// # Panics
 107     ///
 108     /// Panics if given a radix outside the range [0..36].
 109     ///
 110     /// # Examples
 111     ///
 112     /// ```
 113     /// let c = '1';
 114     ///
 115     /// assert_eq!(c.to_digit(10), Some(1));
 116     ///
 117     /// assert_eq!('f'.to_digit(16), Some(15));
 118     /// ```
 119     #[stable(feature = "rust1", since = "1.0.0")]
 120     pub fn to_digit(self, radix: u32) -> Option<u32> { C::to_digit(self, radix) }
 121
 122     /// Returns an iterator that yields the hexadecimal Unicode escape of a
 123     /// character, as `char`s.
 124     ///
 125     /// All characters are escaped with Rust syntax of the form `\\u{NNNN}`
 126     /// where `NNNN` is the shortest hexadecimal representation of the code
 127     /// point.
 128     ///
 129     /// # Examples
 130     ///
 131     /// ```
 132     /// for i in '❤'.escape_unicode() {
 133     ///     println!("{}", i);
 134     /// }
 135     /// ```
 136     ///
 137     /// This prints:
 138     ///
 139     /// ```text
 140     /// \
 141     /// u
 142     /// {
 143     /// 2
 144     /// 7
 145     /// 6
 146     /// 4
 147     /// }
 148     /// ```
 149     ///
 150     /// Collecting into a `String`:
 151     ///
 152     /// ```
 153     /// let heart: String = '❤'.escape_unicode().collect();
 154     ///
 155     /// assert_eq!(heart, r"\u{2764}");
 156     /// ```
 157     #[stable(feature = "rust1", since = "1.0.0")]
 158     pub fn escape_unicode(self) -> EscapeUnicode { C::escape_unicode(self) }
 159
 160     /// Returns an iterator that yields the 'default' ASCII and
 161     /// C++11-like literal escape of a character, as `char`s.
 162     ///
 163     /// The default is chosen with a bias toward producing literals that are
 164     /// legal in a variety of languages, including C++11 and similar C-family
 165     /// languages. The exact rules are:
 166     ///
 167     /// * Tab, CR and LF are escaped as '\t', '\r' and '\n' respectively.
 168     /// * Single-quote, double-quote and backslash chars are backslash-
 169     ///   escaped.
 170     /// * Any other chars in the range [0x20,0x7e] are not escaped.
 171     /// * Any other chars are given hex Unicode escapes; see `escape_unicode`.
 172     ///
 173     /// # Examples
 174     ///
 175     /// ```
 176     /// for i in '"'.escape_default() {
 177     ///     println!("{}", i);
 178     /// }
 179     /// ```
 180     ///
 181     /// This prints:
 182     ///
 183     /// ```text
 184     /// \
 185     /// "
 186     /// ```
 187     ///
 188     /// Collecting into a `String`:
 189     ///
 190     /// ```
 191     /// let quote: String = '"'.escape_default().collect();
 192     ///
 193     /// assert_eq!(quote, "\\\"");
 194     /// ```
 195     #[stable(feature = "rust1", since = "1.0.0")]
 196     pub fn escape_default(self) -> EscapeDefault { C::escape_default(self) }
 197
 198     /// Returns the number of bytes this character would need if encoded in
 199     /// UTF-8.
 200     ///
 201     /// # Examples
 202     ///
 203     /// ```
 204     /// let n = 'ß'.len_utf8();
 205     ///
 206     /// assert_eq!(n, 2);
 207     /// ```
 208     #[stable(feature = "rust1", since = "1.0.0")]
 209     pub fn len_utf8(self) -> usize { C::len_utf8(self) }
 210
 211     /// Returns the number of 16-bit code units this character would need if
 212     /// encoded in UTF-16.
 213     ///
 214     /// # Examples
 215     ///
 216     /// ```
 217     /// let n = 'ß'.len_utf16();
 218     ///
 219     /// assert_eq!(n, 1);
 220     /// ```
 221     #[stable(feature = "rust1", since = "1.0.0")]
 222     pub fn len_utf16(self) -> usize { C::len_utf16(self) }
 223
 224     /// Encodes this character as UTF-8 into the provided byte buffer, and then
 225     /// returns the number of bytes written.
 226     ///
 227     /// If the buffer is not large enough, nothing will be written into it and a
 228     /// `None` will be returned. A buffer of length four is large enough to
 229     /// encode any `char`.
 230     ///
 231     /// # Examples
 232     ///
 233     /// In both of these examples, 'ß' takes two bytes to encode.
 234     ///
 235     /// ```
 236     /// # #![feature(unicode)]
 237     /// let mut b = [0; 2];
 238     ///
 239     /// let result = 'ß'.encode_utf8(&mut b);
 240     ///
 241     /// assert_eq!(result, Some(2));
 242     /// ```
 243     ///
 244     /// A buffer that's too small:
 245     ///
 246     /// ```
 247     /// # #![feature(unicode)]
 248     /// let mut b = [0; 1];
 249     ///
 250     /// let result = 'ß'.encode_utf8(&mut b);
 251     ///
 252     /// assert_eq!(result, None);
 253     /// ```
 254     #[unstable(feature = "unicode",
 255                reason = "pending decision about Iterator/Writer/Reader")]
 256     pub fn encode_utf8(self, dst: &mut [u8]) -> Option<usize> { C::encode_utf8(self, dst) }
 257
 258     /// Encodes this character as UTF-16 into the provided `u16` buffer, and
 259     /// then returns the number of `u16`s written.
 260     ///
 261     /// If the buffer is not large enough, nothing will be written into it and a
 262     /// `None` will be returned. A buffer of length 2 is large enough to encode
 263     /// any `char`.
 264     ///
 265     /// # Examples
 266     ///
 267     /// In both of these examples, 'ß' takes one `u16` to encode.
 268     ///
 269     /// ```
 270     /// # #![feature(unicode)]
 271     /// let mut b = [0; 1];
 272     ///
 273     /// let result = 'ß'.encode_utf16(&mut b);
 274     ///
 275     /// assert_eq!(result, Some(1));
 276     /// ```
 277     ///
 278     /// A buffer that's too small:
 279     ///
 280     /// ```
 281     /// # #![feature(unicode)]
 282     /// let mut b = [0; 0];
 283     ///
 284     /// let result = 'ß'.encode_utf8(&mut b);
 285     ///
 286     /// assert_eq!(result, None);
 287     /// ```
 288     #[unstable(feature = "unicode",
 289                reason = "pending decision about Iterator/Writer/Reader")]
 290     pub fn encode_utf16(self, dst: &mut [u16]) -> Option<usize> { C::encode_utf16(self, dst) }
 291
 292     /// Returns whether the specified character is considered a Unicode
 293     /// alphabetic code point.
 294     #[stable(feature = "rust1", since = "1.0.0")]
 295     #[inline]
 296     pub fn is_alphabetic(self) -> bool {
 297         match self {
 298             'a' ... 'z' | 'A' ... 'Z' => true,
 299             c if c > '\x7f' => derived_property::Alphabetic(c),
 300             _ => false
 301         }
 302     }
 303
 304     /// Returns whether the specified character satisfies the 'XID_Start'
 305     /// Unicode property.
 306     ///
 307     /// 'XID_Start' is a Unicode Derived Property specified in
 308     /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications),
 309     /// mostly similar to ID_Start but modified for closure under NFKx.
 310     #[unstable(feature = "unicode",
 311                reason = "mainly needed for compiler internals")]
 312     #[inline]
 313     pub fn is_xid_start(self) -> bool { derived_property::XID_Start(self) }
 314
 315     /// Returns whether the specified `char` satisfies the 'XID_Continue'
 316     /// Unicode property.
 317     ///
 318     /// 'XID_Continue' is a Unicode Derived Property specified in
 319     /// [UAX #31](http://unicode.org/reports/tr31/#NFKC_Modifications),
 320     /// mostly similar to 'ID_Continue' but modified for closure under NFKx.
 321     #[unstable(feature = "unicode",
 322                reason = "mainly needed for compiler internals")]
 323     #[inline]
 324     pub fn is_xid_continue(self) -> bool { derived_property::XID_Continue(self) }
 325
 326     /// Indicates whether a character is in lowercase.
 327     ///
 328     /// This is defined according to the terms of the Unicode Derived Core
 329     /// Property `Lowercase`.
 330     #[stable(feature = "rust1", since = "1.0.0")]
 331     #[inline]
 332     pub fn is_lowercase(self) -> bool {
 333         match self {
 334             'a' ... 'z' => true,
 335             c if c > '\x7f' => derived_property::Lowercase(c),
 336             _ => false
 337         }
 338     }
 339
 340     /// Indicates whether a character is in uppercase.
 341     ///
 342     /// This is defined according to the terms of the Unicode Derived Core
 343     /// Property `Uppercase`.
 344     #[stable(feature = "rust1", since = "1.0.0")]
 345     #[inline]
 346     pub fn is_uppercase(self) -> bool {
 347         match self {
 348             'A' ... 'Z' => true,
 349             c if c > '\x7f' => derived_property::Uppercase(c),
 350             _ => false
 351         }
 352     }
 353
 354     /// Indicates whether a character is whitespace.
 355     ///
 356     /// Whitespace is defined in terms of the Unicode Property `White_Space`.
 357     #[stable(feature = "rust1", since = "1.0.0")]
 358     #[inline]
 359     pub fn is_whitespace(self) -> bool {
 360         match self {
 361             ' ' | '\x09' ... '\x0d' => true,
 362             c if c > '\x7f' => property::White_Space(c),
 363             _ => false
 364         }
 365     }
 366
 367     /// Indicates whether a character is alphanumeric.
 368     ///
 369     /// Alphanumericness is defined in terms of the Unicode General Categories
 370     /// 'Nd', 'Nl', 'No' and the Derived Core Property 'Alphabetic'.
 371     #[stable(feature = "rust1", since = "1.0.0")]
 372     #[inline]
 373     pub fn is_alphanumeric(self) -> bool {
 374         self.is_alphabetic() || self.is_numeric()
 375     }
 376
 377     /// Indicates whether a character is a control code point.
 378     ///
 379     /// Control code points are defined in terms of the Unicode General
 380     /// Category `Cc`.
 381     #[stable(feature = "rust1", since = "1.0.0")]
 382     #[inline]
 383     pub fn is_control(self) -> bool { general_category::Cc(self) }
 384
 385     /// Indicates whether the character is numeric (Nd, Nl, or No).
 386     #[stable(feature = "rust1", since = "1.0.0")]
 387     #[inline]
 388     pub fn is_numeric(self) -> bool {
 389         match self {
 390             '0' ... '9' => true,
 391             c if c > '\x7f' => general_category::N(c),
 392             _ => false
 393         }
 394     }
 395
 396     /// Converts a character to its lowercase equivalent.
 397     ///
 398     /// The case-folding performed is the common or simple mapping. See
 399     /// `to_uppercase()` for references and more information.
 400     ///
 401     /// # Return value
 402     ///
 403     /// Returns an iterator which yields the characters corresponding to the
 404     /// lowercase equivalent of the character. If no conversion is possible then
 405     /// the input character is returned.
 406     #[stable(feature = "rust1", since = "1.0.0")]
 407     #[inline]
 408     pub fn to_lowercase(self) -> ToLowercase {
 409         ToLowercase(Some(conversions::to_lower(self)))
 410     }
 411
 412     /// Converts a character to its uppercase equivalent.
 413     ///
 414     /// The case-folding performed is the common or simple mapping: it maps
 415     /// one Unicode codepoint to its uppercase equivalent according to the
 416     /// Unicode database [1]. The additional [`SpecialCasing.txt`] is not yet
 417     /// considered here, but the iterator returned will soon support this form
 418     /// of case folding.
 419     ///
 420     /// A full reference can be found here [2].
 421     ///
 422     /// # Return value
 423     ///
 424     /// Returns an iterator which yields the characters corresponding to the
 425     /// uppercase equivalent of the character. If no conversion is possible then
 426     /// the input character is returned.
 427     ///
 428     /// [1]: ftp://ftp.unicode.org/Public/UNIDATA/UnicodeData.txt
 429     ///
 430     /// [`SpecialCasing`.txt`]: ftp://ftp.unicode.org/Public/UNIDATA/SpecialCasing.txt
 431     ///
 432     /// [2]: http://www.unicode.org/versions/Unicode4.0.0/ch03.pdf#G33992
 433     #[stable(feature = "rust1", since = "1.0.0")]
 434     #[inline]
 435     pub fn to_uppercase(self) -> ToUppercase {
 436         ToUppercase(Some(conversions::to_upper(self)))
 437     }
 438
 439     /// Returns this character's displayed width in columns, or `None` if it is a
 440     /// control character other than `'\x00'`.
 441     ///
 442     /// `is_cjk` determines behavior for characters in the Ambiguous category:
 443     /// if `is_cjk` is `true`, these are 2 columns wide; otherwise, they are 1.
 444     /// In CJK contexts, `is_cjk` should be `true`, else it should be `false`.
 445     /// [Unicode Standard Annex #11](http://www.unicode.org/reports/tr11/)
 446     /// recommends that these characters be treated as 1 column (i.e.,
 447     /// `is_cjk` = `false`) if the context cannot be reliably determined.
 448     #[unstable(feature = "unicode",
 449                reason = "needs expert opinion. is_cjk flag stands out as ugly")]
 450     pub fn width(self, is_cjk: bool) -> Option<usize> { charwidth::width(self, is_cjk) }
 451 }