src/libcore/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.
  12 //!
  13 //! For more details, see ::unicode::char (a.k.a. std::char)
  14
  15 #![allow(non_snake_case)]
  16 #![doc(primitive = "char")]
  17
  18 use iter::Iterator;
  19 use mem::transmute;
  20 use option::Option::{None, Some};
  21 use option::Option;
  22 use slice::SliceExt;
  23
  24 // UTF-8 ranges and tags for encoding characters
  25 const TAG_CONT: u8    = 0b1000_0000;
  26 const TAG_TWO_B: u8   = 0b1100_0000;
  27 const TAG_THREE_B: u8 = 0b1110_0000;
  28 const TAG_FOUR_B: u8  = 0b1111_0000;
  29 const MAX_ONE_B: u32   =     0x80;
  30 const MAX_TWO_B: u32   =    0x800;
  31 const MAX_THREE_B: u32 =  0x10000;
  32
  33 /*
  34     Lu  Uppercase_Letter        an uppercase letter
  35     Ll  Lowercase_Letter        a lowercase letter
  36     Lt  Titlecase_Letter        a digraphic character, with first part uppercase
  37     Lm  Modifier_Letter         a modifier letter
  38     Lo  Other_Letter            other letters, including syllables and ideographs
  39     Mn  Nonspacing_Mark         a nonspacing combining mark (zero advance width)
  40     Mc  Spacing_Mark            a spacing combining mark (positive advance width)
  41     Me  Enclosing_Mark          an enclosing combining mark
  42     Nd  Decimal_Number          a decimal digit
  43     Nl  Letter_Number           a letterlike numeric character
  44     No  Other_Number            a numeric character of other type
  45     Pc  Connector_Punctuation   a connecting punctuation mark, like a tie
  46     Pd  Dash_Punctuation        a dash or hyphen punctuation mark
  47     Ps  Open_Punctuation        an opening punctuation mark (of a pair)
  48     Pe  Close_Punctuation       a closing punctuation mark (of a pair)
  49     Pi  Initial_Punctuation     an initial quotation mark
  50     Pf  Final_Punctuation       a final quotation mark
  51     Po  Other_Punctuation       a punctuation mark of other type
  52     Sm  Math_Symbol             a symbol of primarily mathematical use
  53     Sc  Currency_Symbol         a currency sign
  54     Sk  Modifier_Symbol         a non-letterlike modifier symbol
  55     So  Other_Symbol            a symbol of other type
  56     Zs  Space_Separator         a space character (of various non-zero widths)
  57     Zl  Line_Separator          U+2028 LINE SEPARATOR only
  58     Zp  Paragraph_Separator     U+2029 PARAGRAPH SEPARATOR only
  59     Cc  Control                 a C0 or C1 control code
  60     Cf  Format                  a format control character
  61     Cs  Surrogate               a surrogate code point
  62     Co  Private_Use             a private-use character
  63     Cn  Unassigned              a reserved unassigned code point or a noncharacter
  64 */
  65
  66 /// The highest valid code point
  67 #[stable(feature = "rust1", since = "1.0.0")]
  68 pub const MAX: char = '\u{10ffff}';
  69
  70 /// Converts a `u32` to an `Option<char>`.
  71 ///
  72 /// # Examples
  73 ///
  74 /// ```
  75 /// use std::char;
  76 ///
  77 /// let c = char::from_u32(10084); // produces `Some(❤)`
  78 /// assert_eq!(c, Some('❤'));
  79 /// ```
  80 ///
  81 /// An invalid character:
  82 ///
  83 /// ```
  84 /// use std::char;
  85 ///
  86 /// let none = char::from_u32(1114112);
  87 /// assert_eq!(none, None);
  88 /// ```
  89 #[inline]
  90 #[stable(feature = "rust1", since = "1.0.0")]
  91 pub fn from_u32(i: u32) -> Option<char> {
  92     // catch out-of-bounds and surrogates
  93     if (i > MAX as u32) || (i >= 0xD800 && i <= 0xDFFF) {
  94         None
  95     } else {
  96         Some(unsafe { transmute(i) })
  97     }
  98 }
  99
 100 /// Converts a number to the character representing it.
 101 ///
 102 /// # Return value
 103 ///
 104 /// Returns `Some(char)` if `num` represents one digit under `radix`,
 105 /// using one character of `0-9` or `a-z`, or `None` if it doesn't.
 106 ///
 107 /// # Panics
 108 ///
 109 /// Panics if given an `radix` > 36.
 110 ///
 111 /// # Examples
 112 ///
 113 /// ```
 114 /// use std::char;
 115 ///
 116 /// let c = char::from_digit(4, 10);
 117 ///
 118 /// assert_eq!(c, Some('4'));
 119 /// ```
 120 #[inline]
 121 #[stable(feature = "rust1", since = "1.0.0")]
 122 pub fn from_digit(num: u32, radix: u32) -> Option<char> {
 123     if radix > 36 {
 124         panic!("from_digit: radix is too high (maximum 36)");
 125     }
 126     if num < radix {
 127         unsafe {
 128             if num < 10 {
 129                 Some(transmute('0' as u32 + num))
 130             } else {
 131                 Some(transmute('a' as u32 + num - 10))
 132             }
 133         }
 134     } else {
 135         None
 136     }
 137 }
 138
 139 // NB: the stabilization and documentation for this trait is in
 140 // unicode/char.rs, not here
 141 #[allow(missing_docs)] // docs in libunicode/u_char.rs
 142 #[doc(hidden)]
 143 pub trait CharExt {
 144     fn is_digit(self, radix: u32) -> bool;
 145     fn to_digit(self, radix: u32) -> Option<u32>;
 146     fn escape_unicode(self) -> EscapeUnicode;
 147     fn escape_default(self) -> EscapeDefault;
 148     fn len_utf8(self) -> usize;
 149     fn len_utf16(self) -> usize;
 150     fn encode_utf8(self, dst: &mut [u8]) -> Option<usize>;
 151     fn encode_utf16(self, dst: &mut [u16]) -> Option<usize>;
 152 }
 153
 154 impl CharExt for char {
 155     fn is_digit(self, radix: u32) -> bool {
 156         self.to_digit(radix).is_some()
 157     }
 158
 159     fn to_digit(self, radix: u32) -> Option<u32> {
 160         if radix > 36 {
 161             panic!("to_digit: radix is too high (maximum 36)");
 162         }
 163         let val = match self {
 164           '0' ... '9' => self as u32 - '0' as u32,
 165           'a' ... 'z' => self as u32 - 'a' as u32 + 10,
 166           'A' ... 'Z' => self as u32 - 'A' as u32 + 10,
 167           _ => return None,
 168         };
 169         if val < radix { Some(val) }
 170         else { None }
 171     }
 172
 173     fn escape_unicode(self) -> EscapeUnicode {
 174         EscapeUnicode { c: self, state: EscapeUnicodeState::Backslash }
 175     }
 176
 177     fn escape_default(self) -> EscapeDefault {
 178         let init_state = match self {
 179             '\t' => EscapeDefaultState::Backslash('t'),
 180             '\r' => EscapeDefaultState::Backslash('r'),
 181             '\n' => EscapeDefaultState::Backslash('n'),
 182             '\\' => EscapeDefaultState::Backslash('\\'),
 183             '\'' => EscapeDefaultState::Backslash('\''),
 184             '"'  => EscapeDefaultState::Backslash('"'),
 185             '\x20' ... '\x7e' => EscapeDefaultState::Char(self),
 186             _ => EscapeDefaultState::Unicode(self.escape_unicode())
 187         };
 188         EscapeDefault { state: init_state }
 189     }
 190
 191     #[inline]
 192     fn len_utf8(self) -> usize {
 193         let code = self as u32;
 194         if code < MAX_ONE_B {
 195             1
 196         } else if code < MAX_TWO_B {
 197             2
 198         } else if code < MAX_THREE_B {
 199             3
 200         } else {
 201             4
 202         }
 203     }
 204
 205     #[inline]
 206     fn len_utf16(self) -> usize {
 207         let ch = self as u32;
 208         if (ch & 0xFFFF) == ch { 1 } else { 2 }
 209     }
 210
 211     #[inline]
 212     fn encode_utf8(self, dst: &mut [u8]) -> Option<usize> {
 213         encode_utf8_raw(self as u32, dst)
 214     }
 215
 216     #[inline]
 217     fn encode_utf16(self, dst: &mut [u16]) -> Option<usize> {
 218         encode_utf16_raw(self as u32, dst)
 219     }
 220 }
 221
 222 /// Encodes a raw u32 value as UTF-8 into the provided byte buffer,
 223 /// and then returns the number of bytes written.
 224 ///
 225 /// If the buffer is not large enough, nothing will be written into it
 226 /// and a `None` will be returned.
 227 #[inline]
 228 pub fn encode_utf8_raw(code: u32, dst: &mut [u8]) -> Option<usize> {
 229     // Marked #[inline] to allow llvm optimizing it away
 230     if code < MAX_ONE_B && !dst.is_empty() {
 231         dst[0] = code as u8;
 232         Some(1)
 233     } else if code < MAX_TWO_B && dst.len() >= 2 {
 234         dst[0] = (code >> 6 & 0x1F) as u8 | TAG_TWO_B;
 235         dst[1] = (code & 0x3F) as u8 | TAG_CONT;
 236         Some(2)
 237     } else if code < MAX_THREE_B && dst.len() >= 3  {
 238         dst[0] = (code >> 12 & 0x0F) as u8 | TAG_THREE_B;
 239         dst[1] = (code >>  6 & 0x3F) as u8 | TAG_CONT;
 240         dst[2] = (code & 0x3F) as u8 | TAG_CONT;
 241         Some(3)
 242     } else if dst.len() >= 4 {
 243         dst[0] = (code >> 18 & 0x07) as u8 | TAG_FOUR_B;
 244         dst[1] = (code >> 12 & 0x3F) as u8 | TAG_CONT;
 245         dst[2] = (code >>  6 & 0x3F) as u8 | TAG_CONT;
 246         dst[3] = (code & 0x3F) as u8 | TAG_CONT;
 247         Some(4)
 248     } else {
 249         None
 250     }
 251 }
 252
 253 /// Encodes a raw u32 value as UTF-16 into the provided `u16` buffer,
 254 /// and then returns the number of `u16`s written.
 255 ///
 256 /// If the buffer is not large enough, nothing will be written into it
 257 /// and a `None` will be returned.
 258 #[inline]
 259 pub fn encode_utf16_raw(mut ch: u32, dst: &mut [u16]) -> Option<usize> {
 260     // Marked #[inline] to allow llvm optimizing it away
 261     if (ch & 0xFFFF) == ch && !dst.is_empty() {
 262         // The BMP falls through (assuming non-surrogate, as it should)
 263         dst[0] = ch as u16;
 264         Some(1)
 265     } else if dst.len() >= 2 {
 266         // Supplementary planes break into surrogates.
 267         ch -= 0x1_0000;
 268         dst[0] = 0xD800 | ((ch >> 10) as u16);
 269         dst[1] = 0xDC00 | ((ch as u16) & 0x3FF);
 270         Some(2)
 271     } else {
 272         None
 273     }
 274 }
 275
 276 /// An iterator over the characters that represent a `char`, as escaped by
 277 /// Rust's unicode escaping rules.
 278 #[derive(Clone)]
 279 #[stable(feature = "rust1", since = "1.0.0")]
 280 pub struct EscapeUnicode {
 281     c: char,
 282     state: EscapeUnicodeState
 283 }
 284
 285 #[derive(Clone)]
 286 enum EscapeUnicodeState {
 287     Backslash,
 288     Type,
 289     LeftBrace,
 290     Value(usize),
 291     RightBrace,
 292     Done,
 293 }
 294
 295 #[stable(feature = "rust1", since = "1.0.0")]
 296 impl Iterator for EscapeUnicode {
 297     type Item = char;
 298
 299     fn next(&mut self) -> Option<char> {
 300         match self.state {
 301             EscapeUnicodeState::Backslash => {
 302                 self.state = EscapeUnicodeState::Type;
 303                 Some('\\')
 304             }
 305             EscapeUnicodeState::Type => {
 306                 self.state = EscapeUnicodeState::LeftBrace;
 307                 Some('u')
 308             }
 309             EscapeUnicodeState::LeftBrace => {
 310                 let mut n = 0;
 311                 while (self.c as u32) >> (4 * (n + 1)) != 0 {
 312                     n += 1;
 313                 }
 314                 self.state = EscapeUnicodeState::Value(n);
 315                 Some('{')
 316             }
 317             EscapeUnicodeState::Value(offset) => {
 318                 let v = match ((self.c as i32) >> (offset * 4)) & 0xf {
 319                     i @ 0 ... 9 => '0' as i32 + i,
 320                     i => 'a' as i32 + (i - 10)
 321                 };
 322                 if offset == 0 {
 323                     self.state = EscapeUnicodeState::RightBrace;
 324                 } else {
 325                     self.state = EscapeUnicodeState::Value(offset - 1);
 326                 }
 327                 Some(unsafe { transmute(v) })
 328             }
 329             EscapeUnicodeState::RightBrace => {
 330                 self.state = EscapeUnicodeState::Done;
 331                 Some('}')
 332             }
 333             EscapeUnicodeState::Done => None,
 334         }
 335     }
 336 }
 337
 338 /// An iterator over the characters that represent a `char`, escaped
 339 /// for maximum portability.
 340 #[derive(Clone)]
 341 #[stable(feature = "rust1", since = "1.0.0")]
 342 pub struct EscapeDefault {
 343     state: EscapeDefaultState
 344 }
 345
 346 #[derive(Clone)]
 347 enum EscapeDefaultState {
 348     Backslash(char),
 349     Char(char),
 350     Done,
 351     Unicode(EscapeUnicode),
 352 }
 353
 354 #[stable(feature = "rust1", since = "1.0.0")]
 355 impl Iterator for EscapeDefault {
 356     type Item = char;
 357
 358     fn next(&mut self) -> Option<char> {
 359         match self.state {
 360             EscapeDefaultState::Backslash(c) => {
 361                 self.state = EscapeDefaultState::Char(c);
 362                 Some('\\')
 363             }
 364             EscapeDefaultState::Char(c) => {
 365                 self.state = EscapeDefaultState::Done;
 366                 Some(c)
 367             }
 368             EscapeDefaultState::Done => None,
 369             EscapeDefaultState::Unicode(ref mut iter) => iter.next()
 370         }
 371     }
 372 }