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};
21 use iter::{range_step, Iterator, RangeStep};
22 use slice::SlicePrelude;
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
68 pub const MAX: char = '\U0010ffff';
70 /// Converts from `u32` to a `char`
72 #[unstable = "pending decisions about costructors for primitives"]
73 pub fn from_u32(i: u32) -> Option<char> {
74 // catch out-of-bounds and surrogates
75 if (i > MAX as u32) || (i >= 0xD800 && i <= 0xDFFF) {
78 Some(unsafe { transmute(i) })
83 /// Checks if a `char` parses as a numeric digit in the given radix
85 /// Compared to `is_numeric()`, this function only recognizes the
86 /// characters `0-9`, `a-z` and `A-Z`.
90 /// Returns `true` if `c` is a valid digit under `radix`, and `false`
95 /// Panics if given a `radix` > 36.
99 /// This just wraps `to_digit()`.
102 #[deprecated = "use the Char::is_digit method"]
103 pub fn is_digit_radix(c: char, radix: uint) -> bool {
108 /// Converts a `char` to the corresponding digit
112 /// If `c` is between '0' and '9', the corresponding value
113 /// between 0 and 9. If `c` is 'a' or 'A', 10. If `c` is
114 /// 'b' or 'B', 11, etc. Returns none if the `char` does not
115 /// refer to a digit in the given radix.
119 /// Panics if given a `radix` outside the range `[0..36]`.
122 #[deprecated = "use the Char::to_digit method"]
123 pub fn to_digit(c: char, radix: uint) -> Option<uint> {
128 /// Converts a number to the character representing it
132 /// Returns `Some(char)` if `num` represents one digit under `radix`,
133 /// using one character of `0-9` or `a-z`, or `None` if it doesn't.
137 /// Panics if given an `radix` > 36.
140 #[unstable = "pending decisions about costructors for primitives"]
141 pub fn from_digit(num: uint, radix: uint) -> Option<char> {
143 panic!("from_digit: radix is to high (maximum 36)");
148 Some(transmute(('0' as uint + num) as u32))
150 Some(transmute(('a' as uint + num - 10u) as u32))
159 /// Returns the hexadecimal Unicode escape of a `char`
161 /// The rules are as follows:
163 /// - chars in [0,0xff] get 2-digit escapes: `\\xNN`
164 /// - chars in [0x100,0xffff] get 4-digit escapes: `\\uNNNN`
165 /// - chars above 0x10000 get 8-digit escapes: `\\UNNNNNNNN`
167 #[deprecated = "use the Char::escape_unicode method"]
168 pub fn escape_unicode(c: char, f: |char|) {
169 for char in c.escape_unicode() {
175 /// Returns a 'default' ASCII and C++11-like literal escape of a `char`
177 /// The default is chosen with a bias toward producing literals that are
178 /// legal in a variety of languages, including C++11 and similar C-family
179 /// languages. The exact rules are:
181 /// - Tab, CR and LF are escaped as '\t', '\r' and '\n' respectively.
182 /// - Single-quote, double-quote and backslash chars are backslash-escaped.
183 /// - Any other chars in the range [0x20,0x7e] are not escaped.
184 /// - Any other chars are given hex Unicode escapes; see `escape_unicode`.
186 #[deprecated = "use the Char::escape_default method"]
187 pub fn escape_default(c: char, f: |char|) {
188 for c in c.escape_default() {
193 /// Returns the amount of bytes this `char` would need if encoded in UTF-8
195 #[deprecated = "use the Char::len_utf8 method"]
196 pub fn len_utf8_bytes(c: char) -> uint {
200 /// Basic `char` manipulations.
201 #[experimental = "trait organization may change"]
203 /// Checks if a `char` parses as a numeric digit in the given radix.
205 /// Compared to `is_numeric()`, this function only recognizes the characters
206 /// `0-9`, `a-z` and `A-Z`.
210 /// Returns `true` if `c` is a valid digit under `radix`, and `false`
215 /// Panics if given a radix > 36.
216 #[deprecated = "use is_digit"]
217 fn is_digit_radix(self, radix: uint) -> bool;
219 /// Checks if a `char` parses as a numeric digit in the given radix.
221 /// Compared to `is_numeric()`, this function only recognizes the characters
222 /// `0-9`, `a-z` and `A-Z`.
226 /// Returns `true` if `c` is a valid digit under `radix`, and `false`
231 /// Panics if given a radix > 36.
232 #[unstable = "pending error conventions"]
233 fn is_digit(self, radix: uint) -> bool;
235 /// Converts a character to the corresponding digit.
239 /// If `c` is between '0' and '9', the corresponding value between 0 and
240 /// 9. If `c` is 'a' or 'A', 10. If `c` is 'b' or 'B', 11, etc. Returns
241 /// none if the character does not refer to a digit in the given radix.
245 /// Panics if given a radix outside the range [0..36].
246 #[unstable = "pending error conventions, trait organization"]
247 fn to_digit(self, radix: uint) -> Option<uint>;
249 /// Converts a number to the character representing it.
253 /// Returns `Some(char)` if `num` represents one digit under `radix`,
254 /// using one character of `0-9` or `a-z`, or `None` if it doesn't.
258 /// Panics if given a radix > 36.
259 #[deprecated = "use the char::from_digit free function"]
260 fn from_digit(num: uint, radix: uint) -> Option<Self>;
262 /// Converts from `u32` to a `char`
263 #[deprecated = "use the char::from_u32 free function"]
264 fn from_u32(i: u32) -> Option<char>;
266 /// Returns an iterator that yields the hexadecimal Unicode escape
267 /// of a character, as `char`s.
269 /// The rules are as follows:
271 /// * Characters in [0,0xff] get 2-digit escapes: `\\xNN`
272 /// * Characters in [0x100,0xffff] get 4-digit escapes: `\\uNNNN`.
273 /// * Characters above 0x10000 get 8-digit escapes: `\\UNNNNNNNN`.
274 #[unstable = "pending error conventions, trait organization"]
275 fn escape_unicode(self) -> UnicodeEscapedChars;
277 /// Returns an iterator that yields the 'default' ASCII and
278 /// C++11-like literal escape of a character, as `char`s.
280 /// The default is chosen with a bias toward producing literals that are
281 /// legal in a variety of languages, including C++11 and similar C-family
282 /// languages. The exact rules are:
284 /// * Tab, CR and LF are escaped as '\t', '\r' and '\n' respectively.
285 /// * Single-quote, double-quote and backslash chars are backslash-
287 /// * Any other chars in the range [0x20,0x7e] are not escaped.
288 /// * Any other chars are given hex Unicode escapes; see `escape_unicode`.
289 #[unstable = "pending error conventions, trait organization"]
290 fn escape_default(self) -> DefaultEscapedChars;
292 /// Returns the amount of bytes this character would need if encoded in
294 #[deprecated = "use len_utf8"]
295 fn len_utf8_bytes(self) -> uint;
297 /// Returns the amount of bytes this character would need if encoded in
299 #[unstable = "pending trait organization"]
300 fn len_utf8(self) -> uint;
302 /// Returns the amount of bytes this character would need if encoded in
304 #[unstable = "pending trait organization"]
305 fn len_utf16(self) -> uint;
307 /// Encodes this character as UTF-8 into the provided byte buffer,
308 /// and then returns the number of bytes written.
310 /// If the buffer is not large enough, nothing will be written into it
311 /// and a `None` will be returned.
312 #[unstable = "pending trait organization"]
313 fn encode_utf8(&self, dst: &mut [u8]) -> Option<uint>;
315 /// Encodes this character as UTF-16 into the provided `u16` buffer,
316 /// and then returns the number of `u16`s written.
318 /// If the buffer is not large enough, nothing will be written into it
319 /// and a `None` will be returned.
320 #[unstable = "pending trait organization"]
321 fn encode_utf16(&self, dst: &mut [u16]) -> Option<uint>;
324 #[experimental = "trait is experimental"]
326 #[deprecated = "use is_digit"]
327 fn is_digit_radix(self, radix: uint) -> bool { self.is_digit(radix) }
329 #[unstable = "pending trait organization"]
330 fn is_digit(self, radix: uint) -> bool {
331 match self.to_digit(radix) {
337 #[unstable = "pending trait organization"]
338 fn to_digit(self, radix: uint) -> Option<uint> {
340 panic!("to_digit: radix is too high (maximum 36)");
342 let val = match self {
343 '0' ... '9' => self as uint - ('0' as uint),
344 'a' ... 'z' => self as uint + 10u - ('a' as uint),
345 'A' ... 'Z' => self as uint + 10u - ('A' as uint),
348 if val < radix { Some(val) }
352 #[deprecated = "use the char::from_digit free function"]
353 fn from_digit(num: uint, radix: uint) -> Option<char> { from_digit(num, radix) }
356 #[deprecated = "use the char::from_u32 free function"]
357 fn from_u32(i: u32) -> Option<char> { from_u32(i) }
359 #[unstable = "pending error conventions, trait organization"]
360 fn escape_unicode(self) -> UnicodeEscapedChars {
361 UnicodeEscapedChars { c: self, state: UnicodeEscapedCharsState::Backslash }
364 #[unstable = "pending error conventions, trait organization"]
365 fn escape_default(self) -> DefaultEscapedChars {
366 let init_state = match self {
367 '\t' => DefaultEscapedCharsState::Backslash('t'),
368 '\r' => DefaultEscapedCharsState::Backslash('r'),
369 '\n' => DefaultEscapedCharsState::Backslash('n'),
370 '\\' => DefaultEscapedCharsState::Backslash('\\'),
371 '\'' => DefaultEscapedCharsState::Backslash('\''),
372 '"' => DefaultEscapedCharsState::Backslash('"'),
373 '\x20' ... '\x7e' => DefaultEscapedCharsState::Char(self),
374 _ => DefaultEscapedCharsState::Unicode(self.escape_unicode())
376 DefaultEscapedChars { state: init_state }
380 #[deprecated = "use len_utf8"]
381 fn len_utf8_bytes(self) -> uint { self.len_utf8() }
384 #[unstable = "pending trait organization"]
385 fn len_utf8(self) -> uint {
386 let code = self as u32;
388 _ if code < MAX_ONE_B => 1u,
389 _ if code < MAX_TWO_B => 2u,
390 _ if code < MAX_THREE_B => 3u,
396 #[unstable = "pending trait organization"]
397 fn len_utf16(self) -> uint {
398 let ch = self as u32;
399 if (ch & 0xFFFF_u32) == ch { 1 } else { 2 }
403 #[unstable = "pending error conventions, trait organization"]
404 fn encode_utf8<'a>(&self, dst: &'a mut [u8]) -> Option<uint> {
405 // Marked #[inline] to allow llvm optimizing it away
406 let code = *self as u32;
407 if code < MAX_ONE_B && dst.len() >= 1 {
410 } else if code < MAX_TWO_B && dst.len() >= 2 {
411 dst[0] = (code >> 6u & 0x1F_u32) as u8 | TAG_TWO_B;
412 dst[1] = (code & 0x3F_u32) as u8 | TAG_CONT;
414 } else if code < MAX_THREE_B && dst.len() >= 3 {
415 dst[0] = (code >> 12u & 0x0F_u32) as u8 | TAG_THREE_B;
416 dst[1] = (code >> 6u & 0x3F_u32) as u8 | TAG_CONT;
417 dst[2] = (code & 0x3F_u32) as u8 | TAG_CONT;
419 } else if dst.len() >= 4 {
420 dst[0] = (code >> 18u & 0x07_u32) as u8 | TAG_FOUR_B;
421 dst[1] = (code >> 12u & 0x3F_u32) as u8 | TAG_CONT;
422 dst[2] = (code >> 6u & 0x3F_u32) as u8 | TAG_CONT;
423 dst[3] = (code & 0x3F_u32) as u8 | TAG_CONT;
431 #[unstable = "pending error conventions, trait organization"]
432 fn encode_utf16(&self, dst: &mut [u16]) -> Option<uint> {
433 // Marked #[inline] to allow llvm optimizing it away
434 let mut ch = *self as u32;
435 if (ch & 0xFFFF_u32) == ch && dst.len() >= 1 {
436 // The BMP falls through (assuming non-surrogate, as it should)
439 } else if dst.len() >= 2 {
440 // Supplementary planes break into surrogates.
442 dst[0] = 0xD800_u16 | ((ch >> 10) as u16);
443 dst[1] = 0xDC00_u16 | ((ch as u16) & 0x3FF_u16);
451 /// An iterator over the characters that represent a `char`, as escaped by
452 /// Rust's unicode escaping rules.
453 pub struct UnicodeEscapedChars {
455 state: UnicodeEscapedCharsState
458 enum UnicodeEscapedCharsState {
461 Value(RangeStep<i32>),
464 impl Iterator<char> for UnicodeEscapedChars {
465 fn next(&mut self) -> Option<char> {
467 UnicodeEscapedCharsState::Backslash => {
468 self.state = UnicodeEscapedCharsState::Type;
471 UnicodeEscapedCharsState::Type => {
472 let (typechar, pad) = if self.c <= '\x7f' { ('x', 2) }
473 else if self.c <= '\uffff' { ('u', 4) }
475 self.state = UnicodeEscapedCharsState::Value(range_step(4 * (pad - 1), -1, -4i32));
478 UnicodeEscapedCharsState::Value(ref mut range_step) => match range_step.next() {
480 let offset = offset as uint;
481 let v = match ((self.c as i32) >> offset) & 0xf {
482 i @ 0 ... 9 => '0' as i32 + i,
483 i => 'a' as i32 + (i - 10)
485 Some(unsafe { transmute(v) })
493 /// An iterator over the characters that represent a `char`, escaped
494 /// for maximum portability.
495 pub struct DefaultEscapedChars {
496 state: DefaultEscapedCharsState
499 enum DefaultEscapedCharsState {
503 Unicode(UnicodeEscapedChars),
506 impl Iterator<char> for DefaultEscapedChars {
507 fn next(&mut self) -> Option<char> {
509 DefaultEscapedCharsState::Backslash(c) => {
510 self.state = DefaultEscapedCharsState::Char(c);
513 DefaultEscapedCharsState::Char(c) => {
514 self.state = DefaultEscapedCharsState::Done;
517 DefaultEscapedCharsState::Done => None,
518 DefaultEscapedCharsState::Unicode(ref mut iter) => iter.next()