4 use crate::str::from_utf8_unchecked_mut;
5 use crate::unicode::printable::is_printable;
6 use crate::unicode::tables::{conversions, derived_property, general_category, property};
12 /// Checks if a `char` is a digit in the given radix.
14 /// A 'radix' here is sometimes also called a 'base'. A radix of two
15 /// indicates a binary number, a radix of ten, decimal, and a radix of
16 /// sixteen, hexadecimal, to give some common values. Arbitrary
17 /// radices are supported.
19 /// Compared to `is_numeric()`, this function only recognizes the characters
20 /// `0-9`, `a-z` and `A-Z`.
22 /// 'Digit' is defined to be only the following characters:
28 /// For a more comprehensive understanding of 'digit', see [`is_numeric`][is_numeric].
30 /// [is_numeric]: #method.is_numeric
34 /// Panics if given a radix larger than 36.
41 /// assert!('1'.is_digit(10));
42 /// assert!('f'.is_digit(16));
43 /// assert!(!'f'.is_digit(10));
46 /// Passing a large radix, causing a panic:
51 /// let result = thread::spawn(|| {
56 /// assert!(result.is_err());
58 #[stable(feature = "rust1", since = "1.0.0")]
60 pub fn is_digit(self, radix: u32) -> bool {
61 self.to_digit(radix).is_some()
64 /// Converts a `char` to a digit in the given radix.
66 /// A 'radix' here is sometimes also called a 'base'. A radix of two
67 /// indicates a binary number, a radix of ten, decimal, and a radix of
68 /// sixteen, hexadecimal, to give some common values. Arbitrary
69 /// radices are supported.
71 /// 'Digit' is defined to be only the following characters:
79 /// Returns `None` if the `char` does not refer to a digit in the given radix.
83 /// Panics if given a radix larger than 36.
90 /// assert_eq!('1'.to_digit(10), Some(1));
91 /// assert_eq!('f'.to_digit(16), Some(15));
94 /// Passing a non-digit results in failure:
97 /// assert_eq!('f'.to_digit(10), None);
98 /// assert_eq!('z'.to_digit(16), None);
101 /// Passing a large radix, causing a panic:
106 /// let result = thread::spawn(|| {
107 /// '1'.to_digit(37);
110 /// assert!(result.is_err());
112 #[stable(feature = "rust1", since = "1.0.0")]
114 pub fn to_digit(self, radix: u32) -> Option<u32> {
115 assert!(radix <= 36, "to_digit: radix is too high (maximum 36)");
117 // the code is split up here to improve execution speed for cases where
118 // the `radix` is constant and 10 or smaller
119 let val = if radix <= 10 {
121 '0'..='9' => self as u32 - '0' as u32,
126 '0'..='9' => self as u32 - '0' as u32,
127 'a'..='z' => self as u32 - 'a' as u32 + 10,
128 'A'..='Z' => self as u32 - 'A' as u32 + 10,
133 if val < radix { Some(val) } else { None }
136 /// Returns an iterator that yields the hexadecimal Unicode escape of a
137 /// character as `char`s.
139 /// This will escape characters with the Rust syntax of the form
140 /// `\u{NNNNNN}` where `NNNNNN` is a hexadecimal representation.
147 /// for c in '❤'.escape_unicode() {
153 /// Using `println!` directly:
156 /// println!("{}", '❤'.escape_unicode());
159 /// Both are equivalent to:
162 /// println!("\\u{{2764}}");
165 /// Using `to_string`:
168 /// assert_eq!('❤'.escape_unicode().to_string(), "\\u{2764}");
170 #[stable(feature = "rust1", since = "1.0.0")]
172 pub fn escape_unicode(self) -> EscapeUnicode {
175 // or-ing 1 ensures that for c==0 the code computes that one
176 // digit should be printed and (which is the same) avoids the
177 // (31 - 32) underflow
178 let msb = 31 - (c | 1).leading_zeros();
180 // the index of the most significant hex digit
181 let ms_hex_digit = msb / 4;
184 state: EscapeUnicodeState::Backslash,
185 hex_digit_idx: ms_hex_digit as usize,
189 /// An extended version of `escape_debug` that optionally permits escaping
190 /// Extended Grapheme codepoints. This allows us to format characters like
191 /// nonspacing marks better when they're at the start of a string.
193 pub(crate) fn escape_debug_ext(self, escape_grapheme_extended: bool) -> EscapeDebug {
194 let init_state = match self {
195 '\t' => EscapeDefaultState::Backslash('t'),
196 '\r' => EscapeDefaultState::Backslash('r'),
197 '\n' => EscapeDefaultState::Backslash('n'),
198 '\\' | '\'' | '"' => EscapeDefaultState::Backslash(self),
199 _ if escape_grapheme_extended && self.is_grapheme_extended() => {
200 EscapeDefaultState::Unicode(self.escape_unicode())
202 _ if is_printable(self) => EscapeDefaultState::Char(self),
203 _ => EscapeDefaultState::Unicode(self.escape_unicode()),
205 EscapeDebug(EscapeDefault { state: init_state })
208 /// Returns an iterator that yields the literal escape code of a character
211 /// This will escape the characters similar to the `Debug` implementations
212 /// of `str` or `char`.
219 /// for c in '\n'.escape_debug() {
225 /// Using `println!` directly:
228 /// println!("{}", '\n'.escape_debug());
231 /// Both are equivalent to:
237 /// Using `to_string`:
240 /// assert_eq!('\n'.escape_debug().to_string(), "\\n");
242 #[stable(feature = "char_escape_debug", since = "1.20.0")]
244 pub fn escape_debug(self) -> EscapeDebug {
245 self.escape_debug_ext(true)
248 /// Returns an iterator that yields the literal escape code of a character
251 /// The default is chosen with a bias toward producing literals that are
252 /// legal in a variety of languages, including C++11 and similar C-family
253 /// languages. The exact rules are:
255 /// * Tab is escaped as `\t`.
256 /// * Carriage return is escaped as `\r`.
257 /// * Line feed is escaped as `\n`.
258 /// * Single quote is escaped as `\'`.
259 /// * Double quote is escaped as `\"`.
260 /// * Backslash is escaped as `\\`.
261 /// * Any character in the 'printable ASCII' range `0x20` .. `0x7e`
262 /// inclusive is not escaped.
263 /// * All other characters are given hexadecimal Unicode escapes; see
264 /// [`escape_unicode`][escape_unicode].
266 /// [escape_unicode]: #method.escape_unicode
273 /// for c in '"'.escape_default() {
279 /// Using `println!` directly:
282 /// println!("{}", '"'.escape_default());
286 /// Both are equivalent to:
289 /// println!("\\\"");
292 /// Using `to_string`:
295 /// assert_eq!('"'.escape_default().to_string(), "\\\"");
297 #[stable(feature = "rust1", since = "1.0.0")]
299 pub fn escape_default(self) -> EscapeDefault {
300 let init_state = match self {
301 '\t' => EscapeDefaultState::Backslash('t'),
302 '\r' => EscapeDefaultState::Backslash('r'),
303 '\n' => EscapeDefaultState::Backslash('n'),
304 '\\' | '\'' | '"' => EscapeDefaultState::Backslash(self),
305 '\x20'..='\x7e' => EscapeDefaultState::Char(self),
306 _ => EscapeDefaultState::Unicode(self.escape_unicode()),
308 EscapeDefault { state: init_state }
311 /// Returns the number of bytes this `char` would need if encoded in UTF-8.
313 /// That number of bytes is always between 1 and 4, inclusive.
320 /// let len = 'A'.len_utf8();
321 /// assert_eq!(len, 1);
323 /// let len = 'ß'.len_utf8();
324 /// assert_eq!(len, 2);
326 /// let len = 'ℝ'.len_utf8();
327 /// assert_eq!(len, 3);
329 /// let len = '💣'.len_utf8();
330 /// assert_eq!(len, 4);
333 /// The `&str` type guarantees that its contents are UTF-8, and so we can compare the length it
334 /// would take if each code point was represented as a `char` vs in the `&str` itself:
338 /// let eastern = '東';
339 /// let capital = '京';
341 /// // both can be represented as three bytes
342 /// assert_eq!(3, eastern.len_utf8());
343 /// assert_eq!(3, capital.len_utf8());
345 /// // as a &str, these two are encoded in UTF-8
346 /// let tokyo = "東京";
348 /// let len = eastern.len_utf8() + capital.len_utf8();
350 /// // we can see that they take six bytes total...
351 /// assert_eq!(6, tokyo.len());
353 /// // ... just like the &str
354 /// assert_eq!(len, tokyo.len());
356 #[stable(feature = "rust1", since = "1.0.0")]
358 pub fn len_utf8(self) -> usize {
359 let code = self as u32;
360 if code < MAX_ONE_B {
362 } else if code < MAX_TWO_B {
364 } else if code < MAX_THREE_B {
371 /// Returns the number of 16-bit code units this `char` would need if
372 /// encoded in UTF-16.
374 /// See the documentation for [`len_utf8`] for more explanation of this
375 /// concept. This function is a mirror, but for UTF-16 instead of UTF-8.
377 /// [`len_utf8`]: #method.len_utf8
384 /// let n = 'ß'.len_utf16();
385 /// assert_eq!(n, 1);
387 /// let len = '💣'.len_utf16();
388 /// assert_eq!(len, 2);
390 #[stable(feature = "rust1", since = "1.0.0")]
392 pub fn len_utf16(self) -> usize {
393 let ch = self as u32;
394 if (ch & 0xFFFF) == ch { 1 } else { 2 }
397 /// Encodes this character as UTF-8 into the provided byte buffer,
398 /// and then returns the subslice of the buffer that contains the encoded character.
402 /// Panics if the buffer is not large enough.
403 /// A buffer of length four is large enough to encode any `char`.
407 /// In both of these examples, 'ß' takes two bytes to encode.
410 /// let mut b = [0; 2];
412 /// let result = 'ß'.encode_utf8(&mut b);
414 /// assert_eq!(result, "ß");
416 /// assert_eq!(result.len(), 2);
419 /// A buffer that's too small:
424 /// let result = thread::spawn(|| {
425 /// let mut b = [0; 1];
428 /// 'ß'.encode_utf8(&mut b);
431 /// assert!(result.is_err());
433 #[stable(feature = "unicode_encode_char", since = "1.15.0")]
435 pub fn encode_utf8(self, dst: &mut [u8]) -> &mut str {
436 let code = self as u32;
437 // SAFETY: each arm checks the size of the slice and only uses `get_unchecked` unsafe ops
439 let len = if code < MAX_ONE_B && !dst.is_empty() {
440 *dst.get_unchecked_mut(0) = code as u8;
442 } else if code < MAX_TWO_B && dst.len() >= 2 {
443 *dst.get_unchecked_mut(0) = (code >> 6 & 0x1F) as u8 | TAG_TWO_B;
444 *dst.get_unchecked_mut(1) = (code & 0x3F) as u8 | TAG_CONT;
446 } else if code < MAX_THREE_B && dst.len() >= 3 {
447 *dst.get_unchecked_mut(0) = (code >> 12 & 0x0F) as u8 | TAG_THREE_B;
448 *dst.get_unchecked_mut(1) = (code >> 6 & 0x3F) as u8 | TAG_CONT;
449 *dst.get_unchecked_mut(2) = (code & 0x3F) as u8 | TAG_CONT;
451 } else if dst.len() >= 4 {
452 *dst.get_unchecked_mut(0) = (code >> 18 & 0x07) as u8 | TAG_FOUR_B;
453 *dst.get_unchecked_mut(1) = (code >> 12 & 0x3F) as u8 | TAG_CONT;
454 *dst.get_unchecked_mut(2) = (code >> 6 & 0x3F) as u8 | TAG_CONT;
455 *dst.get_unchecked_mut(3) = (code & 0x3F) as u8 | TAG_CONT;
459 "encode_utf8: need {} bytes to encode U+{:X}, but the buffer has {}",
460 from_u32_unchecked(code).len_utf8(),
465 from_utf8_unchecked_mut(dst.get_unchecked_mut(..len))
469 /// Encodes this character as UTF-16 into the provided `u16` buffer,
470 /// and then returns the subslice of the buffer that contains the encoded character.
474 /// Panics if the buffer is not large enough.
475 /// A buffer of length 2 is large enough to encode any `char`.
479 /// In both of these examples, '𝕊' takes two `u16`s to encode.
482 /// let mut b = [0; 2];
484 /// let result = '𝕊'.encode_utf16(&mut b);
486 /// assert_eq!(result.len(), 2);
489 /// A buffer that's too small:
494 /// let result = thread::spawn(|| {
495 /// let mut b = [0; 1];
498 /// '𝕊'.encode_utf16(&mut b);
501 /// assert!(result.is_err());
503 #[stable(feature = "unicode_encode_char", since = "1.15.0")]
505 pub fn encode_utf16(self, dst: &mut [u16]) -> &mut [u16] {
506 let mut code = self as u32;
507 // SAFETY: each arm checks whether there are enough bits to write into
509 if (code & 0xFFFF) == code && !dst.is_empty() {
510 // The BMP falls through (assuming non-surrogate, as it should)
511 *dst.get_unchecked_mut(0) = code as u16;
512 slice::from_raw_parts_mut(dst.as_mut_ptr(), 1)
513 } else if dst.len() >= 2 {
514 // Supplementary planes break into surrogates.
516 *dst.get_unchecked_mut(0) = 0xD800 | ((code >> 10) as u16);
517 *dst.get_unchecked_mut(1) = 0xDC00 | ((code as u16) & 0x3FF);
518 slice::from_raw_parts_mut(dst.as_mut_ptr(), 2)
521 "encode_utf16: need {} units to encode U+{:X}, but the buffer has {}",
522 from_u32_unchecked(code).len_utf16(),
530 /// Returns `true` if this `char` has the `Alphabetic` property.
532 /// `Alphabetic` is described in Chapter 4 (Character Properties) of the [Unicode Standard] and
533 /// specified in the [Unicode Character Database][ucd] [`DerivedCoreProperties.txt`].
535 /// [Unicode Standard]: https://www.unicode.org/versions/latest/
536 /// [ucd]: https://www.unicode.org/reports/tr44/
537 /// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
544 /// assert!('a'.is_alphabetic());
545 /// assert!('京'.is_alphabetic());
548 /// // love is many things, but it is not alphabetic
549 /// assert!(!c.is_alphabetic());
551 #[stable(feature = "rust1", since = "1.0.0")]
553 pub fn is_alphabetic(self) -> bool {
555 'a'..='z' | 'A'..='Z' => true,
556 c => c > '\x7f' && derived_property::Alphabetic(c),
560 /// Returns `true` if this `char` has the `Lowercase` property.
562 /// `Lowercase` is described in Chapter 4 (Character Properties) of the [Unicode Standard] and
563 /// specified in the [Unicode Character Database][ucd] [`DerivedCoreProperties.txt`].
565 /// [Unicode Standard]: https://www.unicode.org/versions/latest/
566 /// [ucd]: https://www.unicode.org/reports/tr44/
567 /// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
574 /// assert!('a'.is_lowercase());
575 /// assert!('δ'.is_lowercase());
576 /// assert!(!'A'.is_lowercase());
577 /// assert!(!'Δ'.is_lowercase());
579 /// // The various Chinese scripts do not have case, and so:
580 /// assert!(!'中'.is_lowercase());
582 #[stable(feature = "rust1", since = "1.0.0")]
584 pub fn is_lowercase(self) -> bool {
587 c => c > '\x7f' && derived_property::Lowercase(c),
591 /// Returns `true` if this `char` has the `Uppercase` property.
593 /// `Uppercase` is described in Chapter 4 (Character Properties) of the [Unicode Standard] and
594 /// specified in the [Unicode Character Database][ucd] [`DerivedCoreProperties.txt`].
596 /// [Unicode Standard]: https://www.unicode.org/versions/latest/
597 /// [ucd]: https://www.unicode.org/reports/tr44/
598 /// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
605 /// assert!(!'a'.is_uppercase());
606 /// assert!(!'δ'.is_uppercase());
607 /// assert!('A'.is_uppercase());
608 /// assert!('Δ'.is_uppercase());
610 /// // The various Chinese scripts do not have case, and so:
611 /// assert!(!'中'.is_uppercase());
613 #[stable(feature = "rust1", since = "1.0.0")]
615 pub fn is_uppercase(self) -> bool {
618 c => c > '\x7f' && derived_property::Uppercase(c),
622 /// Returns `true` if this `char` has the `White_Space` property.
624 /// `White_Space` is specified in the [Unicode Character Database][ucd] [`PropList.txt`].
626 /// [ucd]: https://www.unicode.org/reports/tr44/
627 /// [`PropList.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/PropList.txt
634 /// assert!(' '.is_whitespace());
636 /// // a non-breaking space
637 /// assert!('\u{A0}'.is_whitespace());
639 /// assert!(!'越'.is_whitespace());
641 #[stable(feature = "rust1", since = "1.0.0")]
643 pub fn is_whitespace(self) -> bool {
645 ' ' | '\x09'..='\x0d' => true,
646 c => c > '\x7f' && property::White_Space(c),
650 /// Returns `true` if this `char` satisfies either [`is_alphabetic()`] or [`is_numeric()`].
652 /// [`is_alphabetic()`]: #method.is_alphabetic
653 /// [`is_numeric()`]: #method.is_numeric
660 /// assert!('٣'.is_alphanumeric());
661 /// assert!('7'.is_alphanumeric());
662 /// assert!('৬'.is_alphanumeric());
663 /// assert!('¾'.is_alphanumeric());
664 /// assert!('①'.is_alphanumeric());
665 /// assert!('K'.is_alphanumeric());
666 /// assert!('و'.is_alphanumeric());
667 /// assert!('藏'.is_alphanumeric());
669 #[stable(feature = "rust1", since = "1.0.0")]
671 pub fn is_alphanumeric(self) -> bool {
672 self.is_alphabetic() || self.is_numeric()
675 /// Returns `true` if this `char` has the general category for control codes.
677 /// Control codes (code points with the general category of `Cc`) are described in Chapter 4
678 /// (Character Properties) of the [Unicode Standard] and specified in the [Unicode Character
679 /// Database][ucd] [`UnicodeData.txt`].
681 /// [Unicode Standard]: https://www.unicode.org/versions/latest/
682 /// [ucd]: https://www.unicode.org/reports/tr44/
683 /// [`UnicodeData.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
690 /// // U+009C, STRING TERMINATOR
691 /// assert!('
\9c'.is_control());
692 /// assert!(!'q'.is_control());
694 #[stable(feature = "rust1", since = "1.0.0")]
696 pub fn is_control(self) -> bool {
697 general_category::Cc(self)
700 /// Returns `true` if this `char` has the `Grapheme_Extend` property.
702 /// `Grapheme_Extend` is described in [Unicode Standard Annex #29 (Unicode Text
703 /// Segmentation)][uax29] and specified in the [Unicode Character Database][ucd]
704 /// [`DerivedCoreProperties.txt`].
706 /// [uax29]: https://www.unicode.org/reports/tr29/
707 /// [ucd]: https://www.unicode.org/reports/tr44/
708 /// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
710 pub(crate) fn is_grapheme_extended(self) -> bool {
711 derived_property::Grapheme_Extend(self)
714 /// Returns `true` if this `char` has one of the general categories for numbers.
716 /// The general categories for numbers (`Nd` for decimal digits, `Nl` for letter-like numeric
717 /// characters, and `No` for other numeric characters) are specified in the [Unicode Character
718 /// Database][ucd] [`UnicodeData.txt`].
720 /// [Unicode Standard]: https://www.unicode.org/versions/latest/
721 /// [ucd]: https://www.unicode.org/reports/tr44/
722 /// [`UnicodeData.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
729 /// assert!('٣'.is_numeric());
730 /// assert!('7'.is_numeric());
731 /// assert!('৬'.is_numeric());
732 /// assert!('¾'.is_numeric());
733 /// assert!('①'.is_numeric());
734 /// assert!(!'K'.is_numeric());
735 /// assert!(!'و'.is_numeric());
736 /// assert!(!'藏'.is_numeric());
738 #[stable(feature = "rust1", since = "1.0.0")]
740 pub fn is_numeric(self) -> bool {
743 c => c > '\x7f' && general_category::N(c),
747 /// Returns an iterator that yields the lowercase mapping of this `char` as one or more
750 /// If this `char` does not have a lowercase mapping, the iterator yields the same `char`.
752 /// If this `char` has a one-to-one lowercase mapping given by the [Unicode Character
753 /// Database][ucd] [`UnicodeData.txt`], the iterator yields that `char`.
755 /// [ucd]: https://www.unicode.org/reports/tr44/
756 /// [`UnicodeData.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
758 /// If this `char` requires special considerations (e.g. multiple `char`s) the iterator yields
759 /// the `char`(s) given by [`SpecialCasing.txt`].
761 /// [`SpecialCasing.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/SpecialCasing.txt
763 /// This operation performs an unconditional mapping without tailoring. That is, the conversion
764 /// is independent of context and language.
766 /// In the [Unicode Standard], Chapter 4 (Character Properties) discusses case mapping in
767 /// general and Chapter 3 (Conformance) discusses the default algorithm for case conversion.
769 /// [Unicode Standard]: https://www.unicode.org/versions/latest/
776 /// for c in 'İ'.to_lowercase() {
782 /// Using `println!` directly:
785 /// println!("{}", 'İ'.to_lowercase());
788 /// Both are equivalent to:
791 /// println!("i\u{307}");
794 /// Using `to_string`:
797 /// assert_eq!('C'.to_lowercase().to_string(), "c");
799 /// // Sometimes the result is more than one character:
800 /// assert_eq!('İ'.to_lowercase().to_string(), "i\u{307}");
802 /// // Characters that do not have both uppercase and lowercase
803 /// // convert into themselves.
804 /// assert_eq!('山'.to_lowercase().to_string(), "山");
806 #[stable(feature = "rust1", since = "1.0.0")]
808 pub fn to_lowercase(self) -> ToLowercase {
809 ToLowercase(CaseMappingIter::new(conversions::to_lower(self)))
812 /// Returns an iterator that yields the uppercase mapping of this `char` as one or more
815 /// If this `char` does not have a uppercase mapping, the iterator yields the same `char`.
817 /// If this `char` has a one-to-one uppercase mapping given by the [Unicode Character
818 /// Database][ucd] [`UnicodeData.txt`], the iterator yields that `char`.
820 /// [ucd]: https://www.unicode.org/reports/tr44/
821 /// [`UnicodeData.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
823 /// If this `char` requires special considerations (e.g. multiple `char`s) the iterator yields
824 /// the `char`(s) given by [`SpecialCasing.txt`].
826 /// [`SpecialCasing.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/SpecialCasing.txt
828 /// This operation performs an unconditional mapping without tailoring. That is, the conversion
829 /// is independent of context and language.
831 /// In the [Unicode Standard], Chapter 4 (Character Properties) discusses case mapping in
832 /// general and Chapter 3 (Conformance) discusses the default algorithm for case conversion.
834 /// [Unicode Standard]: https://www.unicode.org/versions/latest/
841 /// for c in 'ß'.to_uppercase() {
847 /// Using `println!` directly:
850 /// println!("{}", 'ß'.to_uppercase());
853 /// Both are equivalent to:
859 /// Using `to_string`:
862 /// assert_eq!('c'.to_uppercase().to_string(), "C");
864 /// // Sometimes the result is more than one character:
865 /// assert_eq!('ß'.to_uppercase().to_string(), "SS");
867 /// // Characters that do not have both uppercase and lowercase
868 /// // convert into themselves.
869 /// assert_eq!('山'.to_uppercase().to_string(), "山");
874 /// In Turkish, the equivalent of 'i' in Latin has five forms instead of two:
876 /// * 'Dotless': I / ı, sometimes written ï
877 /// * 'Dotted': İ / i
879 /// Note that the lowercase dotted 'i' is the same as the Latin. Therefore:
882 /// let upper_i = 'i'.to_uppercase().to_string();
885 /// The value of `upper_i` here relies on the language of the text: if we're
886 /// in `en-US`, it should be `"I"`, but if we're in `tr_TR`, it should
887 /// be `"İ"`. `to_uppercase()` does not take this into account, and so:
890 /// let upper_i = 'i'.to_uppercase().to_string();
892 /// assert_eq!(upper_i, "I");
895 /// holds across languages.
896 #[stable(feature = "rust1", since = "1.0.0")]
898 pub fn to_uppercase(self) -> ToUppercase {
899 ToUppercase(CaseMappingIter::new(conversions::to_upper(self)))
902 /// Checks if the value is within the ASCII range.
908 /// let non_ascii = '❤';
910 /// assert!(ascii.is_ascii());
911 /// assert!(!non_ascii.is_ascii());
913 #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
914 #[rustc_const_stable(feature = "const_ascii_methods_on_intrinsics", since = "1.32.0")]
916 pub const fn is_ascii(&self) -> bool {
920 /// Makes a copy of the value in its ASCII upper case equivalent.
922 /// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z',
923 /// but non-ASCII letters are unchanged.
925 /// To uppercase the value in-place, use [`make_ascii_uppercase`].
927 /// To uppercase ASCII characters in addition to non-ASCII characters, use
928 /// [`to_uppercase`].
934 /// let non_ascii = '❤';
936 /// assert_eq!('A', ascii.to_ascii_uppercase());
937 /// assert_eq!('❤', non_ascii.to_ascii_uppercase());
940 /// [`make_ascii_uppercase`]: #method.make_ascii_uppercase
941 /// [`to_uppercase`]: #method.to_uppercase
942 #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
944 pub fn to_ascii_uppercase(&self) -> char {
945 if self.is_ascii() { (*self as u8).to_ascii_uppercase() as char } else { *self }
948 /// Makes a copy of the value in its ASCII lower case equivalent.
950 /// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z',
951 /// but non-ASCII letters are unchanged.
953 /// To lowercase the value in-place, use [`make_ascii_lowercase`].
955 /// To lowercase ASCII characters in addition to non-ASCII characters, use
956 /// [`to_lowercase`].
962 /// let non_ascii = '❤';
964 /// assert_eq!('a', ascii.to_ascii_lowercase());
965 /// assert_eq!('❤', non_ascii.to_ascii_lowercase());
968 /// [`make_ascii_lowercase`]: #method.make_ascii_lowercase
969 /// [`to_lowercase`]: #method.to_lowercase
970 #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
972 pub fn to_ascii_lowercase(&self) -> char {
973 if self.is_ascii() { (*self as u8).to_ascii_lowercase() as char } else { *self }
976 /// Checks that two values are an ASCII case-insensitive match.
978 /// Equivalent to `to_ascii_lowercase(a) == to_ascii_lowercase(b)`.
983 /// let upper_a = 'A';
984 /// let lower_a = 'a';
985 /// let lower_z = 'z';
987 /// assert!(upper_a.eq_ignore_ascii_case(&lower_a));
988 /// assert!(upper_a.eq_ignore_ascii_case(&upper_a));
989 /// assert!(!upper_a.eq_ignore_ascii_case(&lower_z));
991 #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
993 pub fn eq_ignore_ascii_case(&self, other: &char) -> bool {
994 self.to_ascii_lowercase() == other.to_ascii_lowercase()
997 /// Converts this type to its ASCII upper case equivalent in-place.
999 /// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z',
1000 /// but non-ASCII letters are unchanged.
1002 /// To return a new uppercased value without modifying the existing one, use
1003 /// [`to_ascii_uppercase`].
1008 /// let mut ascii = 'a';
1010 /// ascii.make_ascii_uppercase();
1012 /// assert_eq!('A', ascii);
1015 /// [`to_ascii_uppercase`]: #method.to_ascii_uppercase
1016 #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
1018 pub fn make_ascii_uppercase(&mut self) {
1019 *self = self.to_ascii_uppercase();
1022 /// Converts this type to its ASCII lower case equivalent in-place.
1024 /// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z',
1025 /// but non-ASCII letters are unchanged.
1027 /// To return a new lowercased value without modifying the existing one, use
1028 /// [`to_ascii_lowercase`].
1033 /// let mut ascii = 'A';
1035 /// ascii.make_ascii_lowercase();
1037 /// assert_eq!('a', ascii);
1040 /// [`to_ascii_lowercase`]: #method.to_ascii_lowercase
1041 #[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
1043 pub fn make_ascii_lowercase(&mut self) {
1044 *self = self.to_ascii_lowercase();
1047 /// Checks if the value is an ASCII alphabetic character:
1049 /// - U+0041 'A' ..= U+005A 'Z', or
1050 /// - U+0061 'a' ..= U+007A 'z'.
1055 /// let uppercase_a = 'A';
1056 /// let uppercase_g = 'G';
1060 /// let percent = '%';
1061 /// let space = ' ';
1063 /// let esc: char = 0x1b_u8.into();
1065 /// assert!(uppercase_a.is_ascii_alphabetic());
1066 /// assert!(uppercase_g.is_ascii_alphabetic());
1067 /// assert!(a.is_ascii_alphabetic());
1068 /// assert!(g.is_ascii_alphabetic());
1069 /// assert!(!zero.is_ascii_alphabetic());
1070 /// assert!(!percent.is_ascii_alphabetic());
1071 /// assert!(!space.is_ascii_alphabetic());
1072 /// assert!(!lf.is_ascii_alphabetic());
1073 /// assert!(!esc.is_ascii_alphabetic());
1075 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1077 pub fn is_ascii_alphabetic(&self) -> bool {
1078 self.is_ascii() && (*self as u8).is_ascii_alphabetic()
1081 /// Checks if the value is an ASCII uppercase character:
1082 /// U+0041 'A' ..= U+005A 'Z'.
1087 /// let uppercase_a = 'A';
1088 /// let uppercase_g = 'G';
1092 /// let percent = '%';
1093 /// let space = ' ';
1095 /// let esc: char = 0x1b_u8.into();
1097 /// assert!(uppercase_a.is_ascii_uppercase());
1098 /// assert!(uppercase_g.is_ascii_uppercase());
1099 /// assert!(!a.is_ascii_uppercase());
1100 /// assert!(!g.is_ascii_uppercase());
1101 /// assert!(!zero.is_ascii_uppercase());
1102 /// assert!(!percent.is_ascii_uppercase());
1103 /// assert!(!space.is_ascii_uppercase());
1104 /// assert!(!lf.is_ascii_uppercase());
1105 /// assert!(!esc.is_ascii_uppercase());
1107 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1109 pub fn is_ascii_uppercase(&self) -> bool {
1110 self.is_ascii() && (*self as u8).is_ascii_uppercase()
1113 /// Checks if the value is an ASCII lowercase character:
1114 /// U+0061 'a' ..= U+007A 'z'.
1119 /// let uppercase_a = 'A';
1120 /// let uppercase_g = 'G';
1124 /// let percent = '%';
1125 /// let space = ' ';
1127 /// let esc: char = 0x1b_u8.into();
1129 /// assert!(!uppercase_a.is_ascii_lowercase());
1130 /// assert!(!uppercase_g.is_ascii_lowercase());
1131 /// assert!(a.is_ascii_lowercase());
1132 /// assert!(g.is_ascii_lowercase());
1133 /// assert!(!zero.is_ascii_lowercase());
1134 /// assert!(!percent.is_ascii_lowercase());
1135 /// assert!(!space.is_ascii_lowercase());
1136 /// assert!(!lf.is_ascii_lowercase());
1137 /// assert!(!esc.is_ascii_lowercase());
1139 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1141 pub fn is_ascii_lowercase(&self) -> bool {
1142 self.is_ascii() && (*self as u8).is_ascii_lowercase()
1145 /// Checks if the value is an ASCII alphanumeric character:
1147 /// - U+0041 'A' ..= U+005A 'Z', or
1148 /// - U+0061 'a' ..= U+007A 'z', or
1149 /// - U+0030 '0' ..= U+0039 '9'.
1154 /// let uppercase_a = 'A';
1155 /// let uppercase_g = 'G';
1159 /// let percent = '%';
1160 /// let space = ' ';
1162 /// let esc: char = 0x1b_u8.into();
1164 /// assert!(uppercase_a.is_ascii_alphanumeric());
1165 /// assert!(uppercase_g.is_ascii_alphanumeric());
1166 /// assert!(a.is_ascii_alphanumeric());
1167 /// assert!(g.is_ascii_alphanumeric());
1168 /// assert!(zero.is_ascii_alphanumeric());
1169 /// assert!(!percent.is_ascii_alphanumeric());
1170 /// assert!(!space.is_ascii_alphanumeric());
1171 /// assert!(!lf.is_ascii_alphanumeric());
1172 /// assert!(!esc.is_ascii_alphanumeric());
1174 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1176 pub fn is_ascii_alphanumeric(&self) -> bool {
1177 self.is_ascii() && (*self as u8).is_ascii_alphanumeric()
1180 /// Checks if the value is an ASCII decimal digit:
1181 /// U+0030 '0' ..= U+0039 '9'.
1186 /// let uppercase_a = 'A';
1187 /// let uppercase_g = 'G';
1191 /// let percent = '%';
1192 /// let space = ' ';
1194 /// let esc: char = 0x1b_u8.into();
1196 /// assert!(!uppercase_a.is_ascii_digit());
1197 /// assert!(!uppercase_g.is_ascii_digit());
1198 /// assert!(!a.is_ascii_digit());
1199 /// assert!(!g.is_ascii_digit());
1200 /// assert!(zero.is_ascii_digit());
1201 /// assert!(!percent.is_ascii_digit());
1202 /// assert!(!space.is_ascii_digit());
1203 /// assert!(!lf.is_ascii_digit());
1204 /// assert!(!esc.is_ascii_digit());
1206 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1208 pub fn is_ascii_digit(&self) -> bool {
1209 self.is_ascii() && (*self as u8).is_ascii_digit()
1212 /// Checks if the value is an ASCII hexadecimal digit:
1214 /// - U+0030 '0' ..= U+0039 '9', or
1215 /// - U+0041 'A' ..= U+0046 'F', or
1216 /// - U+0061 'a' ..= U+0066 'f'.
1221 /// let uppercase_a = 'A';
1222 /// let uppercase_g = 'G';
1226 /// let percent = '%';
1227 /// let space = ' ';
1229 /// let esc: char = 0x1b_u8.into();
1231 /// assert!(uppercase_a.is_ascii_hexdigit());
1232 /// assert!(!uppercase_g.is_ascii_hexdigit());
1233 /// assert!(a.is_ascii_hexdigit());
1234 /// assert!(!g.is_ascii_hexdigit());
1235 /// assert!(zero.is_ascii_hexdigit());
1236 /// assert!(!percent.is_ascii_hexdigit());
1237 /// assert!(!space.is_ascii_hexdigit());
1238 /// assert!(!lf.is_ascii_hexdigit());
1239 /// assert!(!esc.is_ascii_hexdigit());
1241 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1243 pub fn is_ascii_hexdigit(&self) -> bool {
1244 self.is_ascii() && (*self as u8).is_ascii_hexdigit()
1247 /// Checks if the value is an ASCII punctuation character:
1249 /// - U+0021 ..= U+002F `! " # $ % & ' ( ) * + , - . /`, or
1250 /// - U+003A ..= U+0040 `: ; < = > ? @`, or
1251 /// - U+005B ..= U+0060 ``[ \ ] ^ _ ` ``, or
1252 /// - U+007B ..= U+007E `{ | } ~`
1257 /// let uppercase_a = 'A';
1258 /// let uppercase_g = 'G';
1262 /// let percent = '%';
1263 /// let space = ' ';
1265 /// let esc: char = 0x1b_u8.into();
1267 /// assert!(!uppercase_a.is_ascii_punctuation());
1268 /// assert!(!uppercase_g.is_ascii_punctuation());
1269 /// assert!(!a.is_ascii_punctuation());
1270 /// assert!(!g.is_ascii_punctuation());
1271 /// assert!(!zero.is_ascii_punctuation());
1272 /// assert!(percent.is_ascii_punctuation());
1273 /// assert!(!space.is_ascii_punctuation());
1274 /// assert!(!lf.is_ascii_punctuation());
1275 /// assert!(!esc.is_ascii_punctuation());
1277 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1279 pub fn is_ascii_punctuation(&self) -> bool {
1280 self.is_ascii() && (*self as u8).is_ascii_punctuation()
1283 /// Checks if the value is an ASCII graphic character:
1284 /// U+0021 '!' ..= U+007E '~'.
1289 /// let uppercase_a = 'A';
1290 /// let uppercase_g = 'G';
1294 /// let percent = '%';
1295 /// let space = ' ';
1297 /// let esc: char = 0x1b_u8.into();
1299 /// assert!(uppercase_a.is_ascii_graphic());
1300 /// assert!(uppercase_g.is_ascii_graphic());
1301 /// assert!(a.is_ascii_graphic());
1302 /// assert!(g.is_ascii_graphic());
1303 /// assert!(zero.is_ascii_graphic());
1304 /// assert!(percent.is_ascii_graphic());
1305 /// assert!(!space.is_ascii_graphic());
1306 /// assert!(!lf.is_ascii_graphic());
1307 /// assert!(!esc.is_ascii_graphic());
1309 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1311 pub fn is_ascii_graphic(&self) -> bool {
1312 self.is_ascii() && (*self as u8).is_ascii_graphic()
1315 /// Checks if the value is an ASCII whitespace character:
1316 /// U+0020 SPACE, U+0009 HORIZONTAL TAB, U+000A LINE FEED,
1317 /// U+000C FORM FEED, or U+000D CARRIAGE RETURN.
1319 /// Rust uses the WhatWG Infra Standard's [definition of ASCII
1320 /// whitespace][infra-aw]. There are several other definitions in
1321 /// wide use. For instance, [the POSIX locale][pct] includes
1322 /// U+000B VERTICAL TAB as well as all the above characters,
1323 /// but—from the very same specification—[the default rule for
1324 /// "field splitting" in the Bourne shell][bfs] considers *only*
1325 /// SPACE, HORIZONTAL TAB, and LINE FEED as whitespace.
1327 /// If you are writing a program that will process an existing
1328 /// file format, check what that format's definition of whitespace is
1329 /// before using this function.
1331 /// [infra-aw]: https://infra.spec.whatwg.org/#ascii-whitespace
1332 /// [pct]: http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap07.html#tag_07_03_01
1333 /// [bfs]: http://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_06_05
1338 /// let uppercase_a = 'A';
1339 /// let uppercase_g = 'G';
1343 /// let percent = '%';
1344 /// let space = ' ';
1346 /// let esc: char = 0x1b_u8.into();
1348 /// assert!(!uppercase_a.is_ascii_whitespace());
1349 /// assert!(!uppercase_g.is_ascii_whitespace());
1350 /// assert!(!a.is_ascii_whitespace());
1351 /// assert!(!g.is_ascii_whitespace());
1352 /// assert!(!zero.is_ascii_whitespace());
1353 /// assert!(!percent.is_ascii_whitespace());
1354 /// assert!(space.is_ascii_whitespace());
1355 /// assert!(lf.is_ascii_whitespace());
1356 /// assert!(!esc.is_ascii_whitespace());
1358 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1360 pub fn is_ascii_whitespace(&self) -> bool {
1361 self.is_ascii() && (*self as u8).is_ascii_whitespace()
1364 /// Checks if the value is an ASCII control character:
1365 /// U+0000 NUL ..= U+001F UNIT SEPARATOR, or U+007F DELETE.
1366 /// Note that most ASCII whitespace characters are control
1367 /// characters, but SPACE is not.
1372 /// let uppercase_a = 'A';
1373 /// let uppercase_g = 'G';
1377 /// let percent = '%';
1378 /// let space = ' ';
1380 /// let esc: char = 0x1b_u8.into();
1382 /// assert!(!uppercase_a.is_ascii_control());
1383 /// assert!(!uppercase_g.is_ascii_control());
1384 /// assert!(!a.is_ascii_control());
1385 /// assert!(!g.is_ascii_control());
1386 /// assert!(!zero.is_ascii_control());
1387 /// assert!(!percent.is_ascii_control());
1388 /// assert!(!space.is_ascii_control());
1389 /// assert!(lf.is_ascii_control());
1390 /// assert!(esc.is_ascii_control());
1392 #[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1394 pub fn is_ascii_control(&self) -> bool {
1395 self.is_ascii() && (*self as u8).is_ascii_control()