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
68 pub const MAX: char = '\u{10ffff}';
70 /// Converts from `u32` to a `char`
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 /// Converts a number to the character representing it
87 /// Returns `Some(char)` if `num` represents one digit under `radix`,
88 /// using one character of `0-9` or `a-z`, or `None` if it doesn't.
92 /// Panics if given an `radix` > 36.
95 #[unstable = "pending integer conventions"]
96 pub fn from_digit(num: uint, radix: uint) -> Option<char> {
98 panic!("from_digit: radix is too high (maximum 36)");
103 Some(transmute(('0' as uint + num) as u32))
105 Some(transmute(('a' as uint + num - 10u) as u32))
113 /// Basic `char` manipulations.
116 /// Checks if a `char` parses as a numeric digit in the given radix.
118 /// Compared to `is_numeric()`, this function only recognizes the characters
119 /// `0-9`, `a-z` and `A-Z`.
123 /// Returns `true` if `c` is a valid digit under `radix`, and `false`
128 /// Panics if given a radix > 36.
129 #[unstable = "pending integer conventions"]
130 fn is_digit(self, radix: uint) -> bool;
132 /// Converts a character to the corresponding digit.
136 /// If `c` is between '0' and '9', the corresponding value between 0 and
137 /// 9. If `c` is 'a' or 'A', 10. If `c` is 'b' or 'B', 11, etc. Returns
138 /// none if the character does not refer to a digit in the given radix.
142 /// Panics if given a radix outside the range [0..36].
143 #[unstable = "pending integer conventions"]
144 fn to_digit(self, radix: uint) -> Option<uint>;
146 /// Returns an iterator that yields the hexadecimal Unicode escape
147 /// of a character, as `char`s.
149 /// All characters are escaped with Rust syntax of the form `\\u{NNNN}`
150 /// where `NNNN` is the shortest hexadecimal representation of the code
153 fn escape_unicode(self) -> EscapeUnicode;
155 /// Returns an iterator that yields the 'default' ASCII and
156 /// C++11-like literal escape of a character, as `char`s.
158 /// The default is chosen with a bias toward producing literals that are
159 /// legal in a variety of languages, including C++11 and similar C-family
160 /// languages. The exact rules are:
162 /// * Tab, CR and LF are escaped as '\t', '\r' and '\n' respectively.
163 /// * Single-quote, double-quote and backslash chars are backslash-
165 /// * Any other chars in the range [0x20,0x7e] are not escaped.
166 /// * Any other chars are given hex Unicode escapes; see `escape_unicode`.
168 fn escape_default(self) -> EscapeDefault;
170 /// Returns the amount of bytes this character would need if encoded in
173 fn len_utf8(self) -> uint;
175 /// Returns the amount of bytes this character would need if encoded in
178 fn len_utf16(self) -> uint;
180 /// Encodes this character as UTF-8 into the provided byte buffer,
181 /// and then returns the number of bytes written.
183 /// If the buffer is not large enough, nothing will be written into it
184 /// and a `None` will be returned.
186 fn encode_utf8(self, dst: &mut [u8]) -> Option<uint>;
188 /// Encodes this character as UTF-16 into the provided `u16` buffer,
189 /// and then returns the number of `u16`s written.
191 /// If the buffer is not large enough, nothing will be written into it
192 /// and a `None` will be returned.
194 fn encode_utf16(self, dst: &mut [u16]) -> Option<uint>;
198 impl CharExt for char {
199 #[unstable = "pending integer conventions"]
200 fn is_digit(self, radix: uint) -> bool {
201 self.to_digit(radix).is_some()
204 #[unstable = "pending integer conventions"]
205 fn to_digit(self, radix: uint) -> Option<uint> {
207 panic!("to_digit: radix is too high (maximum 36)");
209 let val = match self {
210 '0' ... '9' => self as uint - ('0' as uint),
211 'a' ... 'z' => self as uint + 10u - ('a' as uint),
212 'A' ... 'Z' => self as uint + 10u - ('A' as uint),
215 if val < radix { Some(val) }
220 fn escape_unicode(self) -> EscapeUnicode {
221 EscapeUnicode { c: self, state: EscapeUnicodeState::Backslash }
225 fn escape_default(self) -> EscapeDefault {
226 let init_state = match self {
227 '\t' => EscapeDefaultState::Backslash('t'),
228 '\r' => EscapeDefaultState::Backslash('r'),
229 '\n' => EscapeDefaultState::Backslash('n'),
230 '\\' => EscapeDefaultState::Backslash('\\'),
231 '\'' => EscapeDefaultState::Backslash('\''),
232 '"' => EscapeDefaultState::Backslash('"'),
233 '\x20' ... '\x7e' => EscapeDefaultState::Char(self),
234 _ => EscapeDefaultState::Unicode(self.escape_unicode())
236 EscapeDefault { state: init_state }
241 fn len_utf8(self) -> uint {
242 let code = self as u32;
244 _ if code < MAX_ONE_B => 1u,
245 _ if code < MAX_TWO_B => 2u,
246 _ if code < MAX_THREE_B => 3u,
253 fn len_utf16(self) -> uint {
254 let ch = self as u32;
255 if (ch & 0xFFFF_u32) == ch { 1 } else { 2 }
259 #[unstable = "pending decision about Iterator/Writer/Reader"]
260 fn encode_utf8(self, dst: &mut [u8]) -> Option<uint> {
261 // Marked #[inline] to allow llvm optimizing it away
262 let code = self as u32;
263 if code < MAX_ONE_B && dst.len() >= 1 {
266 } else if code < MAX_TWO_B && dst.len() >= 2 {
267 dst[0] = (code >> 6u & 0x1F_u32) as u8 | TAG_TWO_B;
268 dst[1] = (code & 0x3F_u32) as u8 | TAG_CONT;
270 } else if code < MAX_THREE_B && dst.len() >= 3 {
271 dst[0] = (code >> 12u & 0x0F_u32) as u8 | TAG_THREE_B;
272 dst[1] = (code >> 6u & 0x3F_u32) as u8 | TAG_CONT;
273 dst[2] = (code & 0x3F_u32) as u8 | TAG_CONT;
275 } else if dst.len() >= 4 {
276 dst[0] = (code >> 18u & 0x07_u32) as u8 | TAG_FOUR_B;
277 dst[1] = (code >> 12u & 0x3F_u32) as u8 | TAG_CONT;
278 dst[2] = (code >> 6u & 0x3F_u32) as u8 | TAG_CONT;
279 dst[3] = (code & 0x3F_u32) as u8 | TAG_CONT;
287 #[unstable = "pending decision about Iterator/Writer/Reader"]
288 fn encode_utf16(self, dst: &mut [u16]) -> Option<uint> {
289 // Marked #[inline] to allow llvm optimizing it away
290 let mut ch = self as u32;
291 if (ch & 0xFFFF_u32) == ch && dst.len() >= 1 {
292 // The BMP falls through (assuming non-surrogate, as it should)
295 } else if dst.len() >= 2 {
296 // Supplementary planes break into surrogates.
298 dst[0] = 0xD800_u16 | ((ch >> 10) as u16);
299 dst[1] = 0xDC00_u16 | ((ch as u16) & 0x3FF_u16);
307 /// An iterator over the characters that represent a `char`, as escaped by
308 /// Rust's unicode escaping rules.
311 pub struct EscapeUnicode {
313 state: EscapeUnicodeState
317 enum EscapeUnicodeState {
327 impl Iterator for EscapeUnicode {
330 fn next(&mut self) -> Option<char> {
332 EscapeUnicodeState::Backslash => {
333 self.state = EscapeUnicodeState::Type;
336 EscapeUnicodeState::Type => {
337 self.state = EscapeUnicodeState::LeftBrace;
340 EscapeUnicodeState::LeftBrace => {
342 while (self.c as u32) >> (4 * (n + 1)) != 0 {
345 self.state = EscapeUnicodeState::Value(n);
348 EscapeUnicodeState::Value(offset) => {
349 let v = match ((self.c as i32) >> (offset * 4)) & 0xf {
350 i @ 0 ... 9 => '0' as i32 + i,
351 i => 'a' as i32 + (i - 10)
354 self.state = EscapeUnicodeState::RightBrace;
356 self.state = EscapeUnicodeState::Value(offset - 1);
358 Some(unsafe { transmute(v) })
360 EscapeUnicodeState::RightBrace => {
361 self.state = EscapeUnicodeState::Done;
364 EscapeUnicodeState::Done => None,
369 /// An iterator over the characters that represent a `char`, escaped
370 /// for maximum portability.
373 pub struct EscapeDefault {
374 state: EscapeDefaultState
378 enum EscapeDefaultState {
382 Unicode(EscapeUnicode),
386 impl Iterator for EscapeDefault {
389 fn next(&mut self) -> Option<char> {
391 EscapeDefaultState::Backslash(c) => {
392 self.state = EscapeDefaultState::Char(c);
395 EscapeDefaultState::Char(c) => {
396 self.state = EscapeDefaultState::Done;
399 EscapeDefaultState::Done => None,
400 EscapeDefaultState::Unicode(ref mut iter) => iter.next()