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.
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.
11 //! Character manipulation.
13 //! For more details, see ::unicode::char (a.k.a. std::char)
15 #![allow(non_snake_case)]
16 #![doc(primitive = "char")]
20 use option::Option::{None, Some};
24 // UTF-8 ranges and tags for encoding characters
25 static TAG_CONT: u8 = 0b1000_0000u8;
26 static TAG_TWO_B: u8 = 0b1100_0000u8;
27 static TAG_THREE_B: u8 = 0b1110_0000u8;
28 static TAG_FOUR_B: u8 = 0b1111_0000u8;
29 static MAX_ONE_B: u32 = 0x80u32;
30 static MAX_TWO_B: u32 = 0x800u32;
31 static MAX_THREE_B: u32 = 0x10000u32;
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
66 /// The highest valid code point
67 #[stable(feature = "rust1", since = "1.0.0")]
68 pub const MAX: char = '\u{10ffff}';
70 /// Converts a `u32` to an `Option<char>`.
77 /// let c = char::from_u32(10084); // produces `Some(❤)`
78 /// assert_eq!(c, Some('❤'));
81 /// An invalid character:
86 /// let none = char::from_u32(1114112);
87 /// assert_eq!(none, None);
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) {
96 Some(unsafe { transmute(i) })
100 /// Converts a number to the character representing it.
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.
109 /// Panics if given an `radix` > 36.
116 /// let c = char::from_digit(4, 10);
118 /// assert_eq!(c, Some('4'));
121 #[unstable(feature = "core", reason = "pending integer conventions")]
122 pub fn from_digit(num: uint, radix: uint) -> Option<char> {
124 panic!("from_digit: radix is too high (maximum 36)");
129 Some(transmute(('0' as uint + num) as u32))
131 Some(transmute(('a' as uint + num - 10) as u32))
139 /// Basic `char` manipulations.
140 #[stable(feature = "rust1", since = "1.0.0")]
142 /// Checks if a `char` parses as a numeric digit in the given radix.
144 /// Compared to `is_numeric()`, this function only recognizes the characters
145 /// `0-9`, `a-z` and `A-Z`.
149 /// Returns `true` if `c` is a valid digit under `radix`, and `false`
154 /// Panics if given a radix > 36.
161 /// assert!(c.is_digit(10));
163 /// assert!('f'.is_digit(16));
165 #[unstable(feature = "core",
166 reason = "pending integer conventions")]
167 fn is_digit(self, radix: uint) -> bool;
169 /// Converts a character to the corresponding digit.
173 /// If `c` is between '0' and '9', the corresponding value between 0 and
174 /// 9. If `c` is 'a' or 'A', 10. If `c` is 'b' or 'B', 11, etc. Returns
175 /// none if the character does not refer to a digit in the given radix.
179 /// Panics if given a radix outside the range [0..36].
186 /// assert_eq!(c.to_digit(10), Some(1));
188 /// assert_eq!('f'.to_digit(16), Some(15));
190 #[unstable(feature = "core",
191 reason = "pending integer conventions")]
192 fn to_digit(self, radix: uint) -> Option<uint>;
194 /// Returns an iterator that yields the hexadecimal Unicode escape of a character, as `char`s.
196 /// All characters are escaped with Rust syntax of the form `\\u{NNNN}` where `NNNN` is the
197 /// shortest hexadecimal representation of the code point.
202 /// for i in '❤'.escape_unicode() {
203 /// println!("{}", i);
220 /// Collecting into a `String`:
223 /// let heart: String = '❤'.escape_unicode().collect();
225 /// assert_eq!(heart, r"\u{2764}");
227 #[stable(feature = "rust1", since = "1.0.0")]
228 fn escape_unicode(self) -> EscapeUnicode;
230 /// Returns an iterator that yields the 'default' ASCII and
231 /// C++11-like literal escape of a character, as `char`s.
233 /// The default is chosen with a bias toward producing literals that are
234 /// legal in a variety of languages, including C++11 and similar C-family
235 /// languages. The exact rules are:
237 /// * Tab, CR and LF are escaped as '\t', '\r' and '\n' respectively.
238 /// * Single-quote, double-quote and backslash chars are backslash-
240 /// * Any other chars in the range [0x20,0x7e] are not escaped.
241 /// * Any other chars are given hex Unicode escapes; see `escape_unicode`.
246 /// for i in '"'.escape_default() {
247 /// println!("{}", i);
258 /// Collecting into a `String`:
261 /// let quote: String = '"'.escape_default().collect();
263 /// assert_eq!(quote, "\\\"");
265 #[stable(feature = "rust1", since = "1.0.0")]
266 fn escape_default(self) -> EscapeDefault;
268 /// Returns the number of bytes this character would need if encoded in UTF-8.
273 /// let n = 'ß'.len_utf8();
275 /// assert_eq!(n, 2);
277 #[stable(feature = "rust1", since = "1.0.0")]
278 fn len_utf8(self) -> uint;
280 /// Returns the number of bytes this character would need if encoded in UTF-16.
285 /// let n = 'ß'.len_utf16();
287 /// assert_eq!(n, 1);
289 #[stable(feature = "rust1", since = "1.0.0")]
290 fn len_utf16(self) -> uint;
292 /// Encodes this character as UTF-8 into the provided byte buffer, and then returns the number
293 /// of bytes written.
295 /// If the buffer is not large enough, nothing will be written into it and a `None` will be
300 /// In both of these examples, 'ß' takes two bytes to encode.
303 /// let mut b = [0; 2];
305 /// let result = 'ß'.encode_utf8(&mut b);
307 /// assert_eq!(result, Some(2));
310 /// A buffer that's too small:
313 /// let mut b = [0; 1];
315 /// let result = 'ß'.encode_utf8(&mut b);
317 /// assert_eq!(result, None);
319 #[stable(feature = "rust1", since = "1.0.0")]
320 fn encode_utf8(self, dst: &mut [u8]) -> Option<uint>;
322 /// Encodes this character as UTF-16 into the provided `u16` buffer, and then returns the
323 /// number of `u16`s written.
325 /// If the buffer is not large enough, nothing will be written into it and a `None` will be
330 /// In both of these examples, 'ß' takes one byte to encode.
333 /// let mut b = [0; 1];
335 /// let result = 'ß'.encode_utf16(&mut b);
337 /// assert_eq!(result, Some(1));
340 /// A buffer that's too small:
343 /// let mut b = [0; 0];
345 /// let result = 'ß'.encode_utf8(&mut b);
347 /// assert_eq!(result, None);
349 #[stable(feature = "rust1", since = "1.0.0")]
350 fn encode_utf16(self, dst: &mut [u16]) -> Option<uint>;
353 #[stable(feature = "rust1", since = "1.0.0")]
354 impl CharExt for char {
355 #[unstable(feature = "core",
356 reason = "pending integer conventions")]
357 fn is_digit(self, radix: uint) -> bool {
358 self.to_digit(radix).is_some()
361 #[unstable(feature = "core",
362 reason = "pending integer conventions")]
363 fn to_digit(self, radix: uint) -> Option<uint> {
365 panic!("to_digit: radix is too high (maximum 36)");
367 let val = match self {
368 '0' ... '9' => self as uint - ('0' as uint),
369 'a' ... 'z' => self as uint + 10 - ('a' as uint),
370 'A' ... 'Z' => self as uint + 10 - ('A' as uint),
373 if val < radix { Some(val) }
377 #[stable(feature = "rust1", since = "1.0.0")]
378 fn escape_unicode(self) -> EscapeUnicode {
379 EscapeUnicode { c: self, state: EscapeUnicodeState::Backslash }
382 #[stable(feature = "rust1", since = "1.0.0")]
383 fn escape_default(self) -> EscapeDefault {
384 let init_state = match self {
385 '\t' => EscapeDefaultState::Backslash('t'),
386 '\r' => EscapeDefaultState::Backslash('r'),
387 '\n' => EscapeDefaultState::Backslash('n'),
388 '\\' => EscapeDefaultState::Backslash('\\'),
389 '\'' => EscapeDefaultState::Backslash('\''),
390 '"' => EscapeDefaultState::Backslash('"'),
391 '\x20' ... '\x7e' => EscapeDefaultState::Char(self),
392 _ => EscapeDefaultState::Unicode(self.escape_unicode())
394 EscapeDefault { state: init_state }
398 #[stable(feature = "rust1", since = "1.0.0")]
399 fn len_utf8(self) -> uint {
400 let code = self as u32;
402 _ if code < MAX_ONE_B => 1,
403 _ if code < MAX_TWO_B => 2,
404 _ if code < MAX_THREE_B => 3,
410 #[stable(feature = "rust1", since = "1.0.0")]
411 fn len_utf16(self) -> uint {
412 let ch = self as u32;
413 if (ch & 0xFFFF_u32) == ch { 1 } else { 2 }
417 #[unstable(feature = "core",
418 reason = "pending decision about Iterator/Writer/Reader")]
419 fn encode_utf8(self, dst: &mut [u8]) -> Option<uint> {
420 encode_utf8_raw(self as u32, dst)
424 #[unstable(feature = "core",
425 reason = "pending decision about Iterator/Writer/Reader")]
426 fn encode_utf16(self, dst: &mut [u16]) -> Option<uint> {
427 encode_utf16_raw(self as u32, dst)
431 /// Encodes a raw u32 value as UTF-8 into the provided byte buffer,
432 /// and then returns the number of bytes written.
434 /// If the buffer is not large enough, nothing will be written into it
435 /// and a `None` will be returned.
437 #[unstable(feature = "core")]
438 pub fn encode_utf8_raw(code: u32, dst: &mut [u8]) -> Option<uint> {
439 // Marked #[inline] to allow llvm optimizing it away
440 if code < MAX_ONE_B && dst.len() >= 1 {
443 } else if code < MAX_TWO_B && dst.len() >= 2 {
444 dst[0] = (code >> 6u & 0x1F_u32) as u8 | TAG_TWO_B;
445 dst[1] = (code & 0x3F_u32) as u8 | TAG_CONT;
447 } else if code < MAX_THREE_B && dst.len() >= 3 {
448 dst[0] = (code >> 12u & 0x0F_u32) as u8 | TAG_THREE_B;
449 dst[1] = (code >> 6u & 0x3F_u32) as u8 | TAG_CONT;
450 dst[2] = (code & 0x3F_u32) as u8 | TAG_CONT;
452 } else if dst.len() >= 4 {
453 dst[0] = (code >> 18u & 0x07_u32) as u8 | TAG_FOUR_B;
454 dst[1] = (code >> 12u & 0x3F_u32) as u8 | TAG_CONT;
455 dst[2] = (code >> 6u & 0x3F_u32) as u8 | TAG_CONT;
456 dst[3] = (code & 0x3F_u32) as u8 | TAG_CONT;
463 /// Encodes a raw u32 value as UTF-16 into the provided `u16` buffer,
464 /// and then returns the number of `u16`s written.
466 /// If the buffer is not large enough, nothing will be written into it
467 /// and a `None` will be returned.
469 #[unstable(feature = "core")]
470 pub fn encode_utf16_raw(mut ch: u32, dst: &mut [u16]) -> Option<uint> {
471 // Marked #[inline] to allow llvm optimizing it away
472 if (ch & 0xFFFF_u32) == ch && dst.len() >= 1 {
473 // The BMP falls through (assuming non-surrogate, as it should)
476 } else if dst.len() >= 2 {
477 // Supplementary planes break into surrogates.
479 dst[0] = 0xD800_u16 | ((ch >> 10) as u16);
480 dst[1] = 0xDC00_u16 | ((ch as u16) & 0x3FF_u16);
487 /// An iterator over the characters that represent a `char`, as escaped by
488 /// Rust's unicode escaping rules.
490 #[stable(feature = "rust1", since = "1.0.0")]
491 pub struct EscapeUnicode {
493 state: EscapeUnicodeState
497 #[unstable(feature = "core")]
498 enum EscapeUnicodeState {
507 #[stable(feature = "rust1", since = "1.0.0")]
508 impl Iterator for EscapeUnicode {
511 fn next(&mut self) -> Option<char> {
513 EscapeUnicodeState::Backslash => {
514 self.state = EscapeUnicodeState::Type;
517 EscapeUnicodeState::Type => {
518 self.state = EscapeUnicodeState::LeftBrace;
521 EscapeUnicodeState::LeftBrace => {
523 while (self.c as u32) >> (4 * (n + 1)) != 0 {
526 self.state = EscapeUnicodeState::Value(n);
529 EscapeUnicodeState::Value(offset) => {
530 let v = match ((self.c as i32) >> (offset * 4)) & 0xf {
531 i @ 0 ... 9 => '0' as i32 + i,
532 i => 'a' as i32 + (i - 10)
535 self.state = EscapeUnicodeState::RightBrace;
537 self.state = EscapeUnicodeState::Value(offset - 1);
539 Some(unsafe { transmute(v) })
541 EscapeUnicodeState::RightBrace => {
542 self.state = EscapeUnicodeState::Done;
545 EscapeUnicodeState::Done => None,
550 /// An iterator over the characters that represent a `char`, escaped
551 /// for maximum portability.
553 #[stable(feature = "rust1", since = "1.0.0")]
554 pub struct EscapeDefault {
555 state: EscapeDefaultState
559 #[unstable(feature = "core")]
560 enum EscapeDefaultState {
564 Unicode(EscapeUnicode),
567 #[stable(feature = "rust1", since = "1.0.0")]
568 impl Iterator for EscapeDefault {
571 fn next(&mut self) -> Option<char> {
573 EscapeDefaultState::Backslash(c) => {
574 self.state = EscapeDefaultState::Char(c);
577 EscapeDefaultState::Char(c) => {
578 self.state = EscapeDefaultState::Done;
581 EscapeDefaultState::Done => None,
582 EscapeDefaultState::Unicode(ref mut iter) => iter.next()