methods.rs source code [crates/core/src/char/methods.rs]

1	//! impl char {}
2
3	use super::*;
4	use crate::panic::const_panic;
5	use crate::slice;
6	use crate::str::from_utf8_unchecked_mut;
7	use crate::unicode::printable::is_printable;
8	use crate::unicode::{self, conversions};
9
10	impl char {
11	/// The lowest valid code point a `char` can have, `'\0'`.
12	///
13	/// Unlike integer types, `char` actually has a gap in the middle,
14	/// meaning that the range of possible `char`s is smaller than you
15	/// might expect. Ranges of `char` will automatically hop this gap
16	/// for you:
17	///
18	/// ```
19	/// let dist = u32::from(char::MAX) - u32::from(char::MIN);
20	/// let size = (char::MIN..=char::MAX).count() as u32;
21	/// assert!(size < dist);
22	/// ```
23	///
24	/// Despite this gap, the `MIN` and [`MAX`] values can be used as bounds for
25	/// all `char` values.
26	///
27	/// [`MAX`]: char::MAX
28	///
29	/// # Examples
30	///
31	/// ```
32	/// # fn something_which_returns_char() -> char { 'a' }
33	/// let c: char = something_which_returns_char();
34	/// assert!(char::MIN <= c);
35	///
36	/// let value_at_min = u32::from(char::MIN);
37	/// assert_eq!(char::from_u32(value_at_min), Some('`\0`'));
38	/// ```
39	#[stable(feature = "char_min", since = "1.83.0")]
40	pub const MIN: char = '`\0`';
41
42	/// The highest valid code point a `char` can have, `'\u{10FFFF}'`.
43	///
44	/// Unlike integer types, `char` actually has a gap in the middle,
45	/// meaning that the range of possible `char`s is smaller than you
46	/// might expect. Ranges of `char` will automatically hop this gap
47	/// for you:
48	///
49	/// ```
50	/// let dist = u32::from(char::MAX) - u32::from(char::MIN);
51	/// let size = (char::MIN..=char::MAX).count() as u32;
52	/// assert!(size < dist);
53	/// ```
54	///
55	/// Despite this gap, the [`MIN`] and `MAX` values can be used as bounds for
56	/// all `char` values.
57	///
58	/// [`MIN`]: char::MIN
59	///
60	/// # Examples
61	///
62	/// ```
63	/// # fn something_which_returns_char() -> char { 'a' }
64	/// let c: char = something_which_returns_char();
65	/// assert!(c <= char::MAX);
66	///
67	/// let value_at_max = u32::from(char::MAX);
68	/// assert_eq!(char::from_u32(value_at_max), Some('`\u{10FFFF}`'));
69	/// assert_eq!(char::from_u32(value_at_max + `1`), None);
70	/// ```
71	#[stable(feature = "assoc_char_consts", since = "1.52.0")]
72	pub const MAX: char = '`\u{10FFFF}`';
73
74	/// The maximum number of bytes required to [encode](char::encode_utf8) a `char` to
75	/// UTF-8 encoding.
76	#[unstable(feature = "char_max_len", issue = "121714")]
77	pub const MAX_LEN_UTF8: usize = `4`;
78
79	/// The maximum number of two-byte units required to [encode](char::encode_utf16) a `char`
80	/// to UTF-16 encoding.
81	#[unstable(feature = "char_max_len", issue = "121714")]
82	pub const MAX_LEN_UTF16: usize = `2`;
83
84	/// `U+FFFD REPLACEMENT CHARACTER` (�) is used in Unicode to represent a
85	/// decoding error.
86	///
87	/// It can occur, for example, when giving ill-formed UTF-8 bytes to
88	/// [`String::from_utf8_lossy`](../std/string/struct.String.html#method.from_utf8_lossy).
89	#[stable(feature = "assoc_char_consts", since = "1.52.0")]
90	pub const REPLACEMENT_CHARACTER: char = '`\u{FFFD}`';
91
92	/// The version of [Unicode](https://www.unicode.org/) that the Unicode parts of
93	/// `char` and `str` methods are based on.
94	///
95	/// New versions of Unicode are released regularly and subsequently all methods
96	/// in the standard library depending on Unicode are updated. Therefore the
97	/// behavior of some `char` and `str` methods and the value of this constant
98	/// changes over time. This is not* considered to be a breaking change.*
99	///
100	/// The version numbering scheme is explained in
101	/// [Unicode 11.0 or later, Section 3.1 Versions of the Unicode Standard](https://www.unicode.org/versions/Unicode11.0.0/ch03.pdf#page=4).
102	#[stable(feature = "assoc_char_consts", since = "1.52.0")]
103	pub const UNICODE_VERSION: (u8, u8, u8) = crate::unicode::UNICODE_VERSION;
104
105	/// Creates an iterator over the native endian UTF-16 encoded code points in `iter`,
106	/// returning unpaired surrogates as `Err`s.
107	///
108	/// # Examples
109	///
110	/// Basic usage:
111	///
112	/// ```
113	/// // 𝄞mus<invalid>ic<invalid>
114	/// let v = [
115	/// `0xD834`, `0xDD1E`, `0x006d`, `0x0075`, `0x0073`, `0xDD1E`, `0x0069`, `0x0063`, `0xD834`,
116	/// ];
117	///
118	/// assert_eq!(
119	/// char::decode_utf16(v)
120	/// .map(\|r\| r.map_err(\|e\| e.unpaired_surrogate()))
121	/// .collect::<Vec<_>>(),
122	/// vec![
123	/// Ok('𝄞'),
124	/// Ok('m'), Ok('u'), Ok('s'),
125	/// Err(`0xDD1E`),
126	/// Ok('i'), Ok('c'),
127	/// Err(`0xD834`)
128	/// ]
129	/// );
130	/// ```
131	///
132	/// A lossy decoder can be obtained by replacing `Err` results with the replacement character:
133	///
134	/// ```
135	/// // 𝄞mus<invalid>ic<invalid>
136	/// let v = [
137	/// `0xD834`, `0xDD1E`, `0x006d`, `0x0075`, `0x0073`, `0xDD1E`, `0x0069`, `0x0063`, `0xD834`,
138	/// ];
139	///
140	/// assert_eq!(
141	/// char::decode_utf16(v)
142	/// .map(\|r\| r.unwrap_or(char::REPLACEMENT_CHARACTER))
143	/// .collect::<String>(),
144	/// "𝄞mus�ic�"
145	/// );
146	/// ```
147	#[stable(feature = "assoc_char_funcs", since = "1.52.0")]
148	#[inline]
149	pub fn decode_utf16<I: IntoIterator<Item = u16>>(iter: I) -> DecodeUtf16<I::IntoIter> {
150	super::decode::decode_utf16(iter)
151	}
152
153	/// Converts a `u32` to a `char`.
154	///
155	/// Note that all `char`s are valid [`u32`]s, and can be cast to one with
156	/// [`as`](../std/keyword.as.html):
157	///
158	/// ```
159	/// let c = '💯';
160	/// let i = c as u32;
161	///
162	/// assert_eq!(`128175`, i);
163	/// ```
164	///
165	/// However, the reverse is not true: not all valid [`u32`]s are valid
166	/// `char`s. `from_u32()` will return `None` if the input is not a valid value
167	/// for a `char`.
168	///
169	/// For an unsafe version of this function which ignores these checks, see
170	/// [`from_u32_unchecked`].
171	///
172	/// [`from_u32_unchecked`]: #method.from_u32_unchecked
173	///
174	/// # Examples
175	///
176	/// Basic usage:
177	///
178	/// ```
179	/// let c = char::from_u32(`0x2764`);
180	///
181	/// assert_eq!(Some('❤'), c);
182	/// ```
183	///
184	/// Returning `None` when the input is not a valid `char`:
185	///
186	/// ```
187	/// let c = char::from_u32(`0x110000`);
188	///
189	/// assert_eq!(None, c);
190	/// ```
191	#[stable(feature = "assoc_char_funcs", since = "1.52.0")]
192	#[rustc_const_stable(feature = "const_char_convert", since = "1.67.0")]
193	#[must_use]
194	#[inline]
195	pub const fn from_u32(i: u32) -> Option<char> {
196	super::convert::from_u32(i)
197	}
198
199	/// Converts a `u32` to a `char`, ignoring validity.
200	///
201	/// Note that all `char`s are valid [`u32`]s, and can be cast to one with
202	/// `as`:
203	///
204	/// ```
205	/// let c = '💯';
206	/// let i = c as u32;
207	///
208	/// assert_eq!(`128175`, i);
209	/// ```
210	///
211	/// However, the reverse is not true: not all valid [`u32`]s are valid
212	/// `char`s. `from_u32_unchecked()` will ignore this, and blindly cast to
213	/// `char`, possibly creating an invalid one.
214	///
215	/// # Safety
216	///
217	/// This function is unsafe, as it may construct invalid `char` values.
218	///
219	/// For a safe version of this function, see the [`from_u32`] function.
220	///
221	/// [`from_u32`]: #method.from_u32
222	///
223	/// # Examples
224	///
225	/// Basic usage:
226	///
227	/// ```
228	/// let c = unsafe { char::from_u32_unchecked(`0x2764`) };
229	///
230	/// assert_eq!('❤', c);
231	/// ```
232	#[stable(feature = "assoc_char_funcs", since = "1.52.0")]
233	#[rustc_const_stable(feature = "const_char_from_u32_unchecked", since = "1.81.0")]
234	#[must_use]
235	#[inline]
236	pub const unsafe fn from_u32_unchecked(i: u32) -> char {
237	// SAFETY: the safety contract must be upheld by the caller.
238	unsafe { super::convert::from_u32_unchecked(i) }
239	}
240
241	/// Converts a digit in the given radix to a `char`.
242	///
243	/// A 'radix' here is sometimes also called a 'base'. A radix of two
244	/// indicates a binary number, a radix of ten, decimal, and a radix of
245	/// sixteen, hexadecimal, to give some common values. Arbitrary
246	/// radices are supported.
247	///
248	/// `from_digit()` will return `None` if the input is not a digit in
249	/// the given radix.
250	///
251	/// # Panics
252	///
253	/// Panics if given a radix larger than 36.
254	///
255	/// # Examples
256	///
257	/// Basic usage:
258	///
259	/// ```
260	/// let c = char::from_digit(`4`, `10`);
261	///
262	/// assert_eq!(Some('4'), c);
263	///
264	/// // Decimal 11 is a single digit in base 16
265	/// let c = char::from_digit(`11`, `16`);
266	///
267	/// assert_eq!(Some('b'), c);
268	/// ```
269	///
270	/// Returning `None` when the input is not a digit:
271	///
272	/// ```
273	/// let c = char::from_digit(`20`, `10`);
274	///
275	/// assert_eq!(None, c);
276	/// ```
277	///
278	/// Passing a large radix, causing a panic:
279	///
280	/// ```should_panic
281	/// // this panics
282	/// let _c = char::from_digit(`1`, `37`);
283	/// ```
284	#[stable(feature = "assoc_char_funcs", since = "1.52.0")]
285	#[rustc_const_stable(feature = "const_char_convert", since = "1.67.0")]
286	#[must_use]
287	#[inline]
288	pub const fn from_digit(num: u32, radix: u32) -> Option<char> {
289	super::convert::from_digit(num, radix)
290	}
291
292	/// Checks if a `char` is a digit in the given radix.
293	///
294	/// A 'radix' here is sometimes also called a 'base'. A radix of two
295	/// indicates a binary number, a radix of ten, decimal, and a radix of
296	/// sixteen, hexadecimal, to give some common values. Arbitrary
297	/// radices are supported.
298	///
299	/// Compared to [`is_numeric()`], this function only recognizes the characters
300	/// `0-9`, `a-z` and `A-Z`.
301	///
302	/// 'Digit' is defined to be only the following characters:
303	///
304	/// `0-9`*
305	/// `a-z`*
306	/// `A-Z`*
307	///
308	/// For a more comprehensive understanding of 'digit', see [`is_numeric()`].
309	///
310	/// [`is_numeric()`]: #method.is_numeric
311	///
312	/// # Panics
313	///
314	/// Panics if given a radix smaller than 2 or larger than 36.
315	///
316	/// # Examples
317	///
318	/// Basic usage:
319	///
320	/// ```
321	/// assert!('1'.is_digit(`10`));
322	/// assert!('f'.is_digit(`16`));
323	/// assert!(!'f'.is_digit(`10`));
324	/// ```
325	///
326	/// Passing a large radix, causing a panic:
327	///
328	/// ```should_panic
329	/// // this panics
330	/// '1'.is_digit(`37`);
331	/// ```
332	///
333	/// Passing a small radix, causing a panic:
334	///
335	/// ```should_panic
336	/// // this panics
337	/// '1'.is_digit(`1`);
338	/// ```
339	#[stable(feature = "rust1", since = "1.0.0")]
340	#[rustc_const_stable(feature = "const_char_classify", since = "1.87.0")]
341	#[inline]
342	pub const fn is_digit(self, radix: u32) -> bool {
343	self.to_digit(radix).is_some()
344	}
345
346	/// Converts a `char` to a digit in the given radix.
347	///
348	/// A 'radix' here is sometimes also called a 'base'. A radix of two
349	/// indicates a binary number, a radix of ten, decimal, and a radix of
350	/// sixteen, hexadecimal, to give some common values. Arbitrary
351	/// radices are supported.
352	///
353	/// 'Digit' is defined to be only the following characters:
354	///
355	/// `0-9`*
356	/// `a-z`*
357	/// `A-Z`*
358	///
359	/// # Errors
360	///
361	/// Returns `None` if the `char` does not refer to a digit in the given radix.
362	///
363	/// # Panics
364	///
365	/// Panics if given a radix smaller than 2 or larger than 36.
366	///
367	/// # Examples
368	///
369	/// Basic usage:
370	///
371	/// ```
372	/// assert_eq!('1'.to_digit(`10`), Some(`1`));
373	/// assert_eq!('f'.to_digit(`16`), Some(`15`));
374	/// ```
375	///
376	/// Passing a non-digit results in failure:
377	///
378	/// ```
379	/// assert_eq!('f'.to_digit(`10`), None);
380	/// assert_eq!('z'.to_digit(`16`), None);
381	/// ```
382	///
383	/// Passing a large radix, causing a panic:
384	///
385	/// ```should_panic
386	/// // this panics
387	/// let _ = '1'.to_digit(`37`);
388	/// ```
389	/// Passing a small radix, causing a panic:
390	///
391	/// ```should_panic
392	/// // this panics
393	/// let _ = '1'.to_digit(`1`);
394	/// ```
395	#[stable(feature = "rust1", since = "1.0.0")]
396	#[rustc_const_stable(feature = "const_char_convert", since = "1.67.0")]
397	#[must_use = "this returns the result of the operation, \
398	without modifying the original"]
399	#[inline]
400	pub const fn to_digit(self, radix: u32) -> Option<u32> {
401	assert!(
402	radix >= `2` && radix <= `36`,
403	"to_digit: invalid radix -- radix must be in the range 2 to 36 inclusive"
404	);
405	// check radix to remove letter handling code when radix is a known constant
406	let value = if self > '9' && radix > `10` {
407	// mask to convert ASCII letters to uppercase
408	const TO_UPPERCASE_MASK: u32 = !`0b0010_0000`;
409	// Converts an ASCII letter to its corresponding integer value:
410	// A-Z => 10-35, a-z => 10-35. Other characters produce values >= 36.
411	//
412	// Add Overflow Safety:
413	// By applying the mask after the subtraction, the first addendum is
414	// constrained such that it never exceeds u32::MAX - 0x20.
415	((self as u32).wrapping_sub('A' as u32) & TO_UPPERCASE_MASK) + `10`
416	} else {
417	// convert digit to value, non-digits wrap to values > 36
418	(self as u32).wrapping_sub('0' as u32)
419	};
420	// FIXME(const-hack): once then_some is const fn, use it here
421	if value < radix { Some(value) } else { None }
422	}
423
424	/// Returns an iterator that yields the hexadecimal Unicode escape of a
425	/// character as `char`s.
426	///
427	/// This will escape characters with the Rust syntax of the form
428	/// `\u{NNNNNN}` where `NNNNNN` is a hexadecimal representation.
429	///
430	/// # Examples
431	///
432	/// As an iterator:
433	///
434	/// ```
435	/// for c in '❤'.escape_unicode() {
436	/// print!("{c}");
437	/// }
438	/// println!();
439	/// ```
440	///
441	/// Using `println!` directly:
442	///
443	/// ```
444	/// println!("{}", '❤'.escape_unicode());
445	/// ```
446	///
447	/// Both are equivalent to:
448	///
449	/// ```
450	/// println!("`\\`u{{2764}}");
451	/// ```
452	///
453	/// Using [`to_string`](../std/string/trait.ToString.html#tymethod.to_string):
454	///
455	/// ```
456	/// assert_eq!('❤'.escape_unicode().to_string(), "`\\`u{2764}");
457	/// ```
458	#[must_use = "this returns the escaped char as an iterator, \
459	without modifying the original"]
460	#[stable(feature = "rust1", since = "1.0.0")]
461	#[inline]
462	pub fn escape_unicode(self) -> EscapeUnicode {
463	EscapeUnicode::new(self)
464	}
465
466	/// An extended version of `escape_debug` that optionally permits escaping
467	/// Extended Grapheme codepoints, single quotes, and double quotes. This
468	/// allows us to format characters like nonspacing marks better when they're
469	/// at the start of a string, and allows escaping single quotes in
470	/// characters, and double quotes in strings.
471	#[inline]
472	pub(crate) fn escape_debug_ext(self, args: EscapeDebugExtArgs) -> EscapeDebug {
473	match self {
474	'`\0`' => EscapeDebug::backslash(ascii::Char::Digit0),
475	'`\t`' => EscapeDebug::backslash(ascii::Char::SmallT),
476	'`\r`' => EscapeDebug::backslash(ascii::Char::SmallR),
477	'`\n`' => EscapeDebug::backslash(ascii::Char::SmallN),
478	'`\\`' => EscapeDebug::backslash(ascii::Char::ReverseSolidus),
479	'`\"`' if args.escape_double_quote => EscapeDebug::backslash(ascii::Char::QuotationMark),
480	'`\'`' if args.escape_single_quote => EscapeDebug::backslash(ascii::Char::Apostrophe),
481	_ if args.escape_grapheme_extended && self.is_grapheme_extended() => {
482	EscapeDebug::unicode(self)
483	}
484	_ if is_printable(self) => EscapeDebug::printable(self),
485	_ => EscapeDebug::unicode(self),
486	}
487	}
488
489	/// Returns an iterator that yields the literal escape code of a character
490	/// as `char`s.
491	///
492	/// This will escape the characters similar to the [`Debug`](core::fmt::Debug) implementations
493	/// of `str` or `char`.
494	///
495	/// # Examples
496	///
497	/// As an iterator:
498	///
499	/// ```
500	/// for c in '`\n`'.escape_debug() {
501	/// print!("{c}");
502	/// }
503	/// println!();
504	/// ```
505	///
506	/// Using `println!` directly:
507	///
508	/// ```
509	/// println!("{}", '`\n`'.escape_debug());
510	/// ```
511	///
512	/// Both are equivalent to:
513	///
514	/// ```
515	/// println!("`\\`n");
516	/// ```
517	///
518	/// Using [`to_string`](../std/string/trait.ToString.html#tymethod.to_string):
519	///
520	/// ```
521	/// assert_eq!('`\n`'.escape_debug().to_string(), "`\\`n");
522	/// ```
523	#[must_use = "this returns the escaped char as an iterator, \
524	without modifying the original"]
525	#[stable(feature = "char_escape_debug", since = "1.20.0")]
526	#[inline]
527	pub fn escape_debug(self) -> EscapeDebug {
528	self.escape_debug_ext(EscapeDebugExtArgs::ESCAPE_ALL)
529	}
530
531	/// Returns an iterator that yields the literal escape code of a character
532	/// as `char`s.
533	///
534	/// The default is chosen with a bias toward producing literals that are
535	/// legal in a variety of languages, including C++11 and similar C-family
536	/// languages. The exact rules are:
537	///
538	/// Tab is escaped as `\t`.*
539	/// Carriage return is escaped as `\r`.*
540	/// Line feed is escaped as `\n`.*
541	/// Single quote is escaped as `\'`.*
542	/// Double quote is escaped as `\"`.*
543	/// Backslash is escaped as `\\`.*
544	/// Any character in the 'printable ASCII' range `0x20` .. `0x7e`*
545	/// inclusive is not escaped.
546	/// All other characters are given hexadecimal Unicode escapes; see*
547	/// [`escape_unicode`].
548	///
549	/// [`escape_unicode`]: #method.escape_unicode
550	///
551	/// # Examples
552	///
553	/// As an iterator:
554	///
555	/// ```
556	/// for c in '"'.escape_default() {
557	/// print!("{c}");
558	/// }
559	/// println!();
560	/// ```
561	///
562	/// Using `println!` directly:
563	///
564	/// ```
565	/// println!("{}", '"'.escape_default());
566	/// ```
567	///
568	/// Both are equivalent to:
569	///
570	/// ```
571	/// println!("`\\\"`");
572	/// ```
573	///
574	/// Using [`to_string`](../std/string/trait.ToString.html#tymethod.to_string):
575	///
576	/// ```
577	/// assert_eq!('"'.escape_default().to_string(), "`\\\"`");
578	/// ```
579	#[must_use = "this returns the escaped char as an iterator, \
580	without modifying the original"]
581	#[stable(feature = "rust1", since = "1.0.0")]
582	#[inline]
583	pub fn escape_default(self) -> EscapeDefault {
584	match self {
585	'`\t`' => EscapeDefault::backslash(ascii::Char::SmallT),
586	'`\r`' => EscapeDefault::backslash(ascii::Char::SmallR),
587	'`\n`' => EscapeDefault::backslash(ascii::Char::SmallN),
588	'`\\`' \| '`\'`' \| '`\"`' => EscapeDefault::backslash(self.as_ascii().unwrap()),
589	'`\x20`'..='`\x7e`' => EscapeDefault::printable(self.as_ascii().unwrap()),
590	_ => EscapeDefault::unicode(self),
591	}
592	}
593
594	/// Returns the number of bytes this `char` would need if encoded in UTF-8.
595	///
596	/// That number of bytes is always between 1 and 4, inclusive.
597	///
598	/// # Examples
599	///
600	/// Basic usage:
601	///
602	/// ```
603	/// let len = 'A'.len_utf8();
604	/// assert_eq!(len, `1`);
605	///
606	/// let len = 'ß'.len_utf8();
607	/// assert_eq!(len, `2`);
608	///
609	/// let len = 'ℝ'.len_utf8();
610	/// assert_eq!(len, `3`);
611	///
612	/// let len = '💣'.len_utf8();
613	/// assert_eq!(len, `4`);
614	/// ```
615	///
616	/// The `&str` type guarantees that its contents are UTF-8, and so we can compare the length it
617	/// would take if each code point was represented as a `char` vs in the `&str` itself:
618	///
619	/// ```
620	/// // as chars
621	/// let eastern = '東';
622	/// let capital = '京';
623	///
624	/// // both can be represented as three bytes
625	/// assert_eq!(`3`, eastern.len_utf8());
626	/// assert_eq!(`3`, capital.len_utf8());
627	///
628	/// // as a &str, these two are encoded in UTF-8
629	/// let tokyo = "東京";
630	///
631	/// let len = eastern.len_utf8() + capital.len_utf8();
632	///
633	/// // we can see that they take six bytes total...
634	/// assert_eq!(`6`, tokyo.len());
635	///
636	/// // ... just like the &str
637	/// assert_eq!(len, tokyo.len());
638	/// ```
639	#[stable(feature = "rust1", since = "1.0.0")]
640	#[rustc_const_stable(feature = "const_char_len_utf", since = "1.52.0")]
641	#[inline]
642	#[must_use]
643	pub const fn len_utf8(self) -> usize {
644	len_utf8(self as u32)
645	}
646
647	/// Returns the number of 16-bit code units this `char` would need if
648	/// encoded in UTF-16.
649	///
650	/// That number of code units is always either 1 or 2, for unicode scalar values in
651	/// the [basic multilingual plane] or [supplementary planes] respectively.
652	///
653	/// See the documentation for [`len_utf8()`] for more explanation of this
654	/// concept. This function is a mirror, but for UTF-16 instead of UTF-8.
655	///
656	/// [basic multilingual plane]: http://www.unicode.org/glossary/#basic_multilingual_plane
657	/// [supplementary planes]: http://www.unicode.org/glossary/#supplementary_planes
658	/// [`len_utf8()`]: #method.len_utf8
659	///
660	/// # Examples
661	///
662	/// Basic usage:
663	///
664	/// ```
665	/// let n = 'ß'.len_utf16();
666	/// assert_eq!(n, `1`);
667	///
668	/// let len = '💣'.len_utf16();
669	/// assert_eq!(len, `2`);
670	/// ```
671	#[stable(feature = "rust1", since = "1.0.0")]
672	#[rustc_const_stable(feature = "const_char_len_utf", since = "1.52.0")]
673	#[inline]
674	#[must_use]
675	pub const fn len_utf16(self) -> usize {
676	len_utf16(self as u32)
677	}
678
679	/// Encodes this character as UTF-8 into the provided byte buffer,
680	/// and then returns the subslice of the buffer that contains the encoded character.
681	///
682	/// # Panics
683	///
684	/// Panics if the buffer is not large enough.
685	/// A buffer of length four is large enough to encode any `char`.
686	///
687	/// # Examples
688	///
689	/// In both of these examples, 'ß' takes two bytes to encode.
690	///
691	/// ```
692	/// let mut b = [`0`; `2`];
693	///
694	/// let result = 'ß'.encode_utf8(&mut b);
695	///
696	/// assert_eq!(result, "ß");
697	///
698	/// assert_eq!(result.len(), `2`);
699	/// ```
700	///
701	/// A buffer that's too small:
702	///
703	/// ```should_panic
704	/// let mut b = [`0`; `1`];
705	///
706	/// // this panics
707	/// 'ß'.encode_utf8(&mut b);
708	/// ```
709	#[stable(feature = "unicode_encode_char", since = "1.15.0")]
710	#[rustc_const_stable(feature = "const_char_encode_utf8", since = "1.83.0")]
711	#[inline]
712	pub const fn encode_utf8(self, dst: &mut [u8]) -> &mut str {
713	// SAFETY: `char` is not a surrogate, so this is valid UTF-8.
714	unsafe { from_utf8_unchecked_mut(encode_utf8_raw(self as u32, dst)) }
715	}
716
717	/// Encodes this character as native endian UTF-16 into the provided `u16` buffer,
718	/// and then returns the subslice of the buffer that contains the encoded character.
719	///
720	/// # Panics
721	///
722	/// Panics if the buffer is not large enough.
723	/// A buffer of length 2 is large enough to encode any `char`.
724	///
725	/// # Examples
726	///
727	/// In both of these examples, '𝕊' takes two `u16`s to encode.
728	///
729	/// ```
730	/// let mut b = [`0`; `2`];
731	///
732	/// let result = '𝕊'.encode_utf16(&mut b);
733	///
734	/// assert_eq!(result.len(), `2`);
735	/// ```
736	///
737	/// A buffer that's too small:
738	///
739	/// ```should_panic
740	/// let mut b = [`0`; `1`];
741	///
742	/// // this panics
743	/// '𝕊'.encode_utf16(&mut b);
744	/// ```
745	#[stable(feature = "unicode_encode_char", since = "1.15.0")]
746	#[rustc_const_stable(feature = "const_char_encode_utf16", since = "1.84.0")]
747	#[inline]
748	pub const fn encode_utf16(self, dst: &mut [u16]) -> &mut [u16] {
749	encode_utf16_raw(self as u32, dst)
750	}
751
752	/// Returns `true` if this `char` has the `Alphabetic` property.
753	///
754	/// `Alphabetic` is described in Chapter 4 (Character Properties) of the [Unicode Standard] and
755	/// specified in the [Unicode Character Database][ucd] [`DerivedCoreProperties.txt`].
756	///
757	/// [Unicode Standard]: https://www.unicode.org/versions/latest/
758	/// [ucd]: https://www.unicode.org/reports/tr44/
759	/// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
760	///
761	/// # Examples
762	///
763	/// Basic usage:
764	///
765	/// ```
766	/// assert!('a'.is_alphabetic());
767	/// assert!('京'.is_alphabetic());
768	///
769	/// let c = '💝';
770	/// // love is many things, but it is not alphabetic
771	/// assert!(!c.is_alphabetic());
772	/// ```
773	#[must_use]
774	#[stable(feature = "rust1", since = "1.0.0")]
775	#[inline]
776	pub fn is_alphabetic(self) -> bool {
777	match self {
778	'a'..='z' \| 'A'..='Z' => `true`,
779	c => c > '`\x7f`' && unicode::Alphabetic(c),
780	}
781	}
782
783	/// Returns `true` if this `char` has the `Lowercase` property.
784	///
785	/// `Lowercase` is described in Chapter 4 (Character Properties) of the [Unicode Standard] and
786	/// specified in the [Unicode Character Database][ucd] [`DerivedCoreProperties.txt`].
787	///
788	/// [Unicode Standard]: https://www.unicode.org/versions/latest/
789	/// [ucd]: https://www.unicode.org/reports/tr44/
790	/// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
791	///
792	/// # Examples
793	///
794	/// Basic usage:
795	///
796	/// ```
797	/// assert!('a'.is_lowercase());
798	/// assert!('δ'.is_lowercase());
799	/// assert!(!'A'.is_lowercase());
800	/// assert!(!'Δ'.is_lowercase());
801	///
802	/// // The various Chinese scripts and punctuation do not have case, and so:
803	/// assert!(!'中'.is_lowercase());
804	/// assert!(!' '.is_lowercase());
805	/// ```
806	///
807	/// In a const context:
808	///
809	/// ```
810	/// const CAPITAL_DELTA_IS_LOWERCASE: bool = 'Δ'.is_lowercase();
811	/// assert!(!CAPITAL_DELTA_IS_LOWERCASE);
812	/// ```
813	#[must_use]
814	#[stable(feature = "rust1", since = "1.0.0")]
815	#[rustc_const_stable(feature = "const_unicode_case_lookup", since = "1.84.0")]
816	#[inline]
817	pub const fn is_lowercase(self) -> bool {
818	match self {
819	'a'..='z' => `true`,
820	c => c > '`\x7f`' && unicode::Lowercase(c),
821	}
822	}
823
824	/// Returns `true` if this `char` has the `Uppercase` property.
825	///
826	/// `Uppercase` is described in Chapter 4 (Character Properties) of the [Unicode Standard] and
827	/// specified in the [Unicode Character Database][ucd] [`DerivedCoreProperties.txt`].
828	///
829	/// [Unicode Standard]: https://www.unicode.org/versions/latest/
830	/// [ucd]: https://www.unicode.org/reports/tr44/
831	/// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
832	///
833	/// # Examples
834	///
835	/// Basic usage:
836	///
837	/// ```
838	/// assert!(!'a'.is_uppercase());
839	/// assert!(!'δ'.is_uppercase());
840	/// assert!('A'.is_uppercase());
841	/// assert!('Δ'.is_uppercase());
842	///
843	/// // The various Chinese scripts and punctuation do not have case, and so:
844	/// assert!(!'中'.is_uppercase());
845	/// assert!(!' '.is_uppercase());
846	/// ```
847	///
848	/// In a const context:
849	///
850	/// ```
851	/// const CAPITAL_DELTA_IS_UPPERCASE: bool = 'Δ'.is_uppercase();
852	/// assert!(CAPITAL_DELTA_IS_UPPERCASE);
853	/// ```
854	#[must_use]
855	#[stable(feature = "rust1", since = "1.0.0")]
856	#[rustc_const_stable(feature = "const_unicode_case_lookup", since = "1.84.0")]
857	#[inline]
858	pub const fn is_uppercase(self) -> bool {
859	match self {
860	'A'..='Z' => `true`,
861	c => c > '`\x7f`' && unicode::Uppercase(c),
862	}
863	}
864
865	/// Returns `true` if this `char` has the `White_Space` property.
866	///
867	/// `White_Space` is specified in the [Unicode Character Database][ucd] [`PropList.txt`].
868	///
869	/// [ucd]: https://www.unicode.org/reports/tr44/
870	/// [`PropList.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/PropList.txt
871	///
872	/// # Examples
873	///
874	/// Basic usage:
875	///
876	/// ```
877	/// assert!(' '.is_whitespace());
878	///
879	/// // line break
880	/// assert!('`\n`'.is_whitespace());
881	///
882	/// // a non-breaking space
883	/// assert!('`\u{A0}`'.is_whitespace());
884	///
885	/// assert!(!'越'.is_whitespace());
886	/// ```
887	#[must_use]
888	#[stable(feature = "rust1", since = "1.0.0")]
889	#[rustc_const_stable(feature = "const_char_classify", since = "1.87.0")]
890	#[inline]
891	pub const fn is_whitespace(self) -> bool {
892	match self {
893	' ' \| '`\x09`'..='`\x0d`' => `true`,
894	c => c > '`\x7f`' && unicode::White_Space(c),
895	}
896	}
897
898	/// Returns `true` if this `char` satisfies either [`is_alphabetic()`] or [`is_numeric()`].
899	///
900	/// [`is_alphabetic()`]: #method.is_alphabetic
901	/// [`is_numeric()`]: #method.is_numeric
902	///
903	/// # Examples
904	///
905	/// Basic usage:
906	///
907	/// ```
908	/// assert!('٣'.is_alphanumeric());
909	/// assert!('7'.is_alphanumeric());
910	/// assert!('৬'.is_alphanumeric());
911	/// assert!('¾'.is_alphanumeric());
912	/// assert!('①'.is_alphanumeric());
913	/// assert!('K'.is_alphanumeric());
914	/// assert!('و'.is_alphanumeric());
915	/// assert!('藏'.is_alphanumeric());
916	/// ```
917	#[must_use]
918	#[stable(feature = "rust1", since = "1.0.0")]
919	#[inline]
920	pub fn is_alphanumeric(self) -> bool {
921	self.is_alphabetic() \|\| self.is_numeric()
922	}
923
924	/// Returns `true` if this `char` has the general category for control codes.
925	///
926	/// Control codes (code points with the general category of `Cc`) are described in Chapter 4
927	/// (Character Properties) of the [Unicode Standard] and specified in the [Unicode Character
928	/// Database][ucd] [`UnicodeData.txt`].
929	///
930	/// [Unicode Standard]: https://www.unicode.org/versions/latest/
931	/// [ucd]: https://www.unicode.org/reports/tr44/
932	/// [`UnicodeData.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
933	///
934	/// # Examples
935	///
936	/// Basic usage:
937	///
938	/// ```
939	/// // U+009C, STRING TERMINATOR
940	/// assert!(''.is_control());
941	/// assert!(!'q'.is_control());
942	/// ```
943	#[must_use]
944	#[stable(feature = "rust1", since = "1.0.0")]
945	#[inline]
946	pub fn is_control(self) -> bool {
947	unicode::Cc(self)
948	}
949
950	/// Returns `true` if this `char` has the `Grapheme_Extend` property.
951	///
952	/// `Grapheme_Extend` is described in [Unicode Standard Annex #29 (Unicode Text
953	/// Segmentation)][uax29] and specified in the [Unicode Character Database][ucd]
954	/// [`DerivedCoreProperties.txt`].
955	///
956	/// [uax29]: https://www.unicode.org/reports/tr29/
957	/// [ucd]: https://www.unicode.org/reports/tr44/
958	/// [`DerivedCoreProperties.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
959	#[must_use]
960	#[inline]
961	pub(crate) fn is_grapheme_extended(self) -> bool {
962	unicode::Grapheme_Extend(self)
963	}
964
965	/// Returns `true` if this `char` has one of the general categories for numbers.
966	///
967	/// The general categories for numbers (`Nd` for decimal digits, `Nl` for letter-like numeric
968	/// characters, and `No` for other numeric characters) are specified in the [Unicode Character
969	/// Database][ucd] [`UnicodeData.txt`].
970	///
971	/// This method doesn't cover everything that could be considered a number, e.g. ideographic numbers like '三'.
972	/// If you want everything including characters with overlapping purposes then you might want to use
973	/// a unicode or language-processing library that exposes the appropriate character properties instead
974	/// of looking at the unicode categories.
975	///
976	/// If you want to parse ASCII decimal digits (0-9) or ASCII base-N, use
977	/// `is_ascii_digit` or `is_digit` instead.
978	///
979	/// [Unicode Standard]: https://www.unicode.org/versions/latest/
980	/// [ucd]: https://www.unicode.org/reports/tr44/
981	/// [`UnicodeData.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
982	///
983	/// # Examples
984	///
985	/// Basic usage:
986	///
987	/// ```
988	/// assert!('٣'.is_numeric());
989	/// assert!('7'.is_numeric());
990	/// assert!('৬'.is_numeric());
991	/// assert!('¾'.is_numeric());
992	/// assert!('①'.is_numeric());
993	/// assert!(!'K'.is_numeric());
994	/// assert!(!'و'.is_numeric());
995	/// assert!(!'藏'.is_numeric());
996	/// assert!(!'三'.is_numeric());
997	/// ```
998	#[must_use]
999	#[stable(feature = "rust1", since = "1.0.0")]
1000	#[inline]
1001	pub fn is_numeric(self) -> bool {
1002	match self {
1003	'0'..='9' => `true`,
1004	c => c > '`\x7f`' && unicode::N(c),
1005	}
1006	}
1007
1008	/// Returns an iterator that yields the lowercase mapping of this `char` as one or more
1009	/// `char`s.
1010	///
1011	/// If this `char` does not have a lowercase mapping, the iterator yields the same `char`.
1012	///
1013	/// If this `char` has a one-to-one lowercase mapping given by the [Unicode Character
1014	/// Database][ucd] [`UnicodeData.txt`], the iterator yields that `char`.
1015	///
1016	/// [ucd]: https://www.unicode.org/reports/tr44/
1017	/// [`UnicodeData.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
1018	///
1019	/// If this `char` requires special considerations (e.g. multiple `char`s) the iterator yields
1020	/// the `char`(s) given by [`SpecialCasing.txt`].
1021	///
1022	/// [`SpecialCasing.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/SpecialCasing.txt
1023	///
1024	/// This operation performs an unconditional mapping without tailoring. That is, the conversion
1025	/// is independent of context and language.
1026	///
1027	/// In the [Unicode Standard], Chapter 4 (Character Properties) discusses case mapping in
1028	/// general and Chapter 3 (Conformance) discusses the default algorithm for case conversion.
1029	///
1030	/// [Unicode Standard]: https://www.unicode.org/versions/latest/
1031	///
1032	/// # Examples
1033	///
1034	/// As an iterator:
1035	///
1036	/// ```
1037	/// for c in 'İ'.to_lowercase() {
1038	/// print!("{c}");
1039	/// }
1040	/// println!();
1041	/// ```
1042	///
1043	/// Using `println!` directly:
1044	///
1045	/// ```
1046	/// println!("{}", 'İ'.to_lowercase());
1047	/// ```
1048	///
1049	/// Both are equivalent to:
1050	///
1051	/// ```
1052	/// println!("i`\u{307}`");
1053	/// ```
1054	///
1055	/// Using [`to_string`](../std/string/trait.ToString.html#tymethod.to_string):
1056	///
1057	/// ```
1058	/// assert_eq!('C'.to_lowercase().to_string(), "c");
1059	///
1060	/// // Sometimes the result is more than one character:
1061	/// assert_eq!('İ'.to_lowercase().to_string(), "i`\u{307}`");
1062	///
1063	/// // Characters that do not have both uppercase and lowercase
1064	/// // convert into themselves.
1065	/// assert_eq!('山'.to_lowercase().to_string(), "山");
1066	/// ```
1067	#[must_use = "this returns the lowercase character as a new iterator, \
1068	without modifying the original"]
1069	#[stable(feature = "rust1", since = "1.0.0")]
1070	#[inline]
1071	pub fn to_lowercase(self) -> ToLowercase {
1072	ToLowercase(CaseMappingIter::new(conversions::to_lower(self)))
1073	}
1074
1075	/// Returns an iterator that yields the uppercase mapping of this `char` as one or more
1076	/// `char`s.
1077	///
1078	/// If this `char` does not have an uppercase mapping, the iterator yields the same `char`.
1079	///
1080	/// If this `char` has a one-to-one uppercase mapping given by the [Unicode Character
1081	/// Database][ucd] [`UnicodeData.txt`], the iterator yields that `char`.
1082	///
1083	/// [ucd]: https://www.unicode.org/reports/tr44/
1084	/// [`UnicodeData.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
1085	///
1086	/// If this `char` requires special considerations (e.g. multiple `char`s) the iterator yields
1087	/// the `char`(s) given by [`SpecialCasing.txt`].
1088	///
1089	/// [`SpecialCasing.txt`]: https://www.unicode.org/Public/UCD/latest/ucd/SpecialCasing.txt
1090	///
1091	/// This operation performs an unconditional mapping without tailoring. That is, the conversion
1092	/// is independent of context and language.
1093	///
1094	/// In the [Unicode Standard], Chapter 4 (Character Properties) discusses case mapping in
1095	/// general and Chapter 3 (Conformance) discusses the default algorithm for case conversion.
1096	///
1097	/// [Unicode Standard]: https://www.unicode.org/versions/latest/
1098	///
1099	/// # Examples
1100	///
1101	/// As an iterator:
1102	///
1103	/// ```
1104	/// for c in 'ß'.to_uppercase() {
1105	/// print!("{c}");
1106	/// }
1107	/// println!();
1108	/// ```
1109	///
1110	/// Using `println!` directly:
1111	///
1112	/// ```
1113	/// println!("{}", 'ß'.to_uppercase());
1114	/// ```
1115	///
1116	/// Both are equivalent to:
1117	///
1118	/// ```
1119	/// println!("SS");
1120	/// ```
1121	///
1122	/// Using [`to_string`](../std/string/trait.ToString.html#tymethod.to_string):
1123	///
1124	/// ```
1125	/// assert_eq!('c'.to_uppercase().to_string(), "C");
1126	///
1127	/// // Sometimes the result is more than one character:
1128	/// assert_eq!('ß'.to_uppercase().to_string(), "SS");
1129	///
1130	/// // Characters that do not have both uppercase and lowercase
1131	/// // convert into themselves.
1132	/// assert_eq!('山'.to_uppercase().to_string(), "山");
1133	/// ```
1134	///
1135	/// # Note on locale
1136	///
1137	/// In Turkish, the equivalent of 'i' in Latin has five forms instead of two:
1138	///
1139	/// 'Dotless': I / ı, sometimes written ï*
1140	/// 'Dotted': İ / i*
1141	///
1142	/// Note that the lowercase dotted 'i' is the same as the Latin. Therefore:
1143	///
1144	/// ```
1145	/// let upper_i = 'i'.to_uppercase().to_string();
1146	/// ```
1147	///
1148	/// The value of `upper_i` here relies on the language of the text: if we're
1149	/// in `en-US`, it should be `"I"`, but if we're in `tr_TR`, it should
1150	/// be `"İ"`. `to_uppercase()` does not take this into account, and so:
1151	///
1152	/// ```
1153	/// let upper_i = 'i'.to_uppercase().to_string();
1154	///
1155	/// assert_eq!(upper_i, "I");
1156	/// ```
1157	///
1158	/// holds across languages.
1159	#[must_use = "this returns the uppercase character as a new iterator, \
1160	without modifying the original"]
1161	#[stable(feature = "rust1", since = "1.0.0")]
1162	#[inline]
1163	pub fn to_uppercase(self) -> ToUppercase {
1164	ToUppercase(CaseMappingIter::new(conversions::to_upper(self)))
1165	}
1166
1167	/// Checks if the value is within the ASCII range.
1168	///
1169	/// # Examples
1170	///
1171	/// ```
1172	/// let ascii = 'a';
1173	/// let non_ascii = '❤';
1174	///
1175	/// assert!(ascii.is_ascii());
1176	/// assert!(!non_ascii.is_ascii());
1177	/// ```
1178	#[must_use]
1179	#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
1180	#[rustc_const_stable(feature = "const_char_is_ascii", since = "1.32.0")]
1181	#[rustc_diagnostic_item = "char_is_ascii"]
1182	#[inline]
1183	pub const fn is_ascii(&self) -> bool {
1184	self as u32* <= `0x7F`
1185	}
1186
1187	/// Returns `Some` if the value is within the ASCII range,
1188	/// or `None` if it's not.
1189	///
1190	/// This is preferred to [`Self::is_ascii`] when you're passing the value
1191	/// along to something else that can take [`ascii::Char`] rather than
1192	/// needing to check again for itself whether the value is in ASCII.
1193	#[must_use]
1194	#[unstable(feature = "ascii_char", issue = "110998")]
1195	#[inline]
1196	pub const fn as_ascii(&self) -> Option<ascii::Char> {
1197	if self.is_ascii() {
1198	// SAFETY: Just checked that this is ASCII.
1199	Some(unsafe { ascii::Char::from_u8_unchecked(*self as u8) })
1200	} else {
1201	None
1202	}
1203	}
1204
1205	/// Makes a copy of the value in its ASCII upper case equivalent.
1206	///
1207	/// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z',
1208	/// but non-ASCII letters are unchanged.
1209	///
1210	/// To uppercase the value in-place, use [`make_ascii_uppercase()`].
1211	///
1212	/// To uppercase ASCII characters in addition to non-ASCII characters, use
1213	/// [`to_uppercase()`].
1214	///
1215	/// # Examples
1216	///
1217	/// ```
1218	/// let ascii = 'a';
1219	/// let non_ascii = '❤';
1220	///
1221	/// assert_eq!('A', ascii.to_ascii_uppercase());
1222	/// assert_eq!('❤', non_ascii.to_ascii_uppercase());
1223	/// ```
1224	///
1225	/// [`make_ascii_uppercase()`]: #method.make_ascii_uppercase
1226	/// [`to_uppercase()`]: #method.to_uppercase
1227	#[must_use = "to uppercase the value in-place, use `make_ascii_uppercase()`"]
1228	#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
1229	#[rustc_const_stable(feature = "const_ascii_methods_on_intrinsics", since = "1.52.0")]
1230	#[inline]
1231	pub const fn to_ascii_uppercase(&self) -> char {
1232	if self.is_ascii_lowercase() {
1233	(self as u8).ascii_change_case_unchecked() as char*
1234	} else {
1235	*self
1236	}
1237	}
1238
1239	/// Makes a copy of the value in its ASCII lower case equivalent.
1240	///
1241	/// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z',
1242	/// but non-ASCII letters are unchanged.
1243	///
1244	/// To lowercase the value in-place, use [`make_ascii_lowercase()`].
1245	///
1246	/// To lowercase ASCII characters in addition to non-ASCII characters, use
1247	/// [`to_lowercase()`].
1248	///
1249	/// # Examples
1250	///
1251	/// ```
1252	/// let ascii = 'A';
1253	/// let non_ascii = '❤';
1254	///
1255	/// assert_eq!('a', ascii.to_ascii_lowercase());
1256	/// assert_eq!('❤', non_ascii.to_ascii_lowercase());
1257	/// ```
1258	///
1259	/// [`make_ascii_lowercase()`]: #method.make_ascii_lowercase
1260	/// [`to_lowercase()`]: #method.to_lowercase
1261	#[must_use = "to lowercase the value in-place, use `make_ascii_lowercase()`"]
1262	#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
1263	#[rustc_const_stable(feature = "const_ascii_methods_on_intrinsics", since = "1.52.0")]
1264	#[inline]
1265	pub const fn to_ascii_lowercase(&self) -> char {
1266	if self.is_ascii_uppercase() {
1267	(self as u8).ascii_change_case_unchecked() as char*
1268	} else {
1269	*self
1270	}
1271	}
1272
1273	/// Checks that two values are an ASCII case-insensitive match.
1274	///
1275	/// Equivalent to <code>[to_ascii_lowercase]\(a) == [to_ascii_lowercase]\(b)</code>.
1276	///
1277	/// # Examples
1278	///
1279	/// ```
1280	/// let upper_a = 'A';
1281	/// let lower_a = 'a';
1282	/// let lower_z = 'z';
1283	///
1284	/// assert!(upper_a.eq_ignore_ascii_case(&lower_a));
1285	/// assert!(upper_a.eq_ignore_ascii_case(&upper_a));
1286	/// assert!(!upper_a.eq_ignore_ascii_case(&lower_z));
1287	/// ```
1288	///
1289	/// [to_ascii_lowercase]: #method.to_ascii_lowercase
1290	#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
1291	#[rustc_const_stable(feature = "const_ascii_methods_on_intrinsics", since = "1.52.0")]
1292	#[inline]
1293	pub const fn eq_ignore_ascii_case(&self, other: &char) -> bool {
1294	self.to_ascii_lowercase() == other.to_ascii_lowercase()
1295	}
1296
1297	/// Converts this type to its ASCII upper case equivalent in-place.
1298	///
1299	/// ASCII letters 'a' to 'z' are mapped to 'A' to 'Z',
1300	/// but non-ASCII letters are unchanged.
1301	///
1302	/// To return a new uppercased value without modifying the existing one, use
1303	/// [`to_ascii_uppercase()`].
1304	///
1305	/// # Examples
1306	///
1307	/// ```
1308	/// let mut ascii = 'a';
1309	///
1310	/// ascii.make_ascii_uppercase();
1311	///
1312	/// assert_eq!('A', ascii);
1313	/// ```
1314	///
1315	/// [`to_ascii_uppercase()`]: #method.to_ascii_uppercase
1316	#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
1317	#[rustc_const_stable(feature = "const_make_ascii", since = "1.84.0")]
1318	#[inline]
1319	pub const fn make_ascii_uppercase(&mut self) {
1320	*self = self.to_ascii_uppercase();
1321	}
1322
1323	/// Converts this type to its ASCII lower case equivalent in-place.
1324	///
1325	/// ASCII letters 'A' to 'Z' are mapped to 'a' to 'z',
1326	/// but non-ASCII letters are unchanged.
1327	///
1328	/// To return a new lowercased value without modifying the existing one, use
1329	/// [`to_ascii_lowercase()`].
1330	///
1331	/// # Examples
1332	///
1333	/// ```
1334	/// let mut ascii = 'A';
1335	///
1336	/// ascii.make_ascii_lowercase();
1337	///
1338	/// assert_eq!('a', ascii);
1339	/// ```
1340	///
1341	/// [`to_ascii_lowercase()`]: #method.to_ascii_lowercase
1342	#[stable(feature = "ascii_methods_on_intrinsics", since = "1.23.0")]
1343	#[rustc_const_stable(feature = "const_make_ascii", since = "1.84.0")]
1344	#[inline]
1345	pub const fn make_ascii_lowercase(&mut self) {
1346	*self = self.to_ascii_lowercase();
1347	}
1348
1349	/// Checks if the value is an ASCII alphabetic character:
1350	///
1351	/// - U+0041 'A' ..= U+005A 'Z', or
1352	/// - U+0061 'a' ..= U+007A 'z'.
1353	///
1354	/// # Examples
1355	///
1356	/// ```
1357	/// let uppercase_a = 'A';
1358	/// let uppercase_g = 'G';
1359	/// let a = 'a';
1360	/// let g = 'g';
1361	/// let zero = '0';
1362	/// let percent = '%';
1363	/// let space = ' ';
1364	/// let lf = '`\n`';
1365	/// let esc = '`\x1b`';
1366	///
1367	/// assert!(uppercase_a.is_ascii_alphabetic());
1368	/// assert!(uppercase_g.is_ascii_alphabetic());
1369	/// assert!(a.is_ascii_alphabetic());
1370	/// assert!(g.is_ascii_alphabetic());
1371	/// assert!(!zero.is_ascii_alphabetic());
1372	/// assert!(!percent.is_ascii_alphabetic());
1373	/// assert!(!space.is_ascii_alphabetic());
1374	/// assert!(!lf.is_ascii_alphabetic());
1375	/// assert!(!esc.is_ascii_alphabetic());
1376	/// ```
1377	#[must_use]
1378	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1379	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1380	#[inline]
1381	pub const fn is_ascii_alphabetic(&self) -> bool {
1382	matches!(*self, 'A'..='Z' \| 'a'..='z')
1383	}
1384
1385	/// Checks if the value is an ASCII uppercase character:
1386	/// U+0041 'A' ..= U+005A 'Z'.
1387	///
1388	/// # Examples
1389	///
1390	/// ```
1391	/// let uppercase_a = 'A';
1392	/// let uppercase_g = 'G';
1393	/// let a = 'a';
1394	/// let g = 'g';
1395	/// let zero = '0';
1396	/// let percent = '%';
1397	/// let space = ' ';
1398	/// let lf = '`\n`';
1399	/// let esc = '`\x1b`';
1400	///
1401	/// assert!(uppercase_a.is_ascii_uppercase());
1402	/// assert!(uppercase_g.is_ascii_uppercase());
1403	/// assert!(!a.is_ascii_uppercase());
1404	/// assert!(!g.is_ascii_uppercase());
1405	/// assert!(!zero.is_ascii_uppercase());
1406	/// assert!(!percent.is_ascii_uppercase());
1407	/// assert!(!space.is_ascii_uppercase());
1408	/// assert!(!lf.is_ascii_uppercase());
1409	/// assert!(!esc.is_ascii_uppercase());
1410	/// ```
1411	#[must_use]
1412	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1413	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1414	#[inline]
1415	pub const fn is_ascii_uppercase(&self) -> bool {
1416	matches!(*self, 'A'..='Z')
1417	}
1418
1419	/// Checks if the value is an ASCII lowercase character:
1420	/// U+0061 'a' ..= U+007A 'z'.
1421	///
1422	/// # Examples
1423	///
1424	/// ```
1425	/// let uppercase_a = 'A';
1426	/// let uppercase_g = 'G';
1427	/// let a = 'a';
1428	/// let g = 'g';
1429	/// let zero = '0';
1430	/// let percent = '%';
1431	/// let space = ' ';
1432	/// let lf = '`\n`';
1433	/// let esc = '`\x1b`';
1434	///
1435	/// assert!(!uppercase_a.is_ascii_lowercase());
1436	/// assert!(!uppercase_g.is_ascii_lowercase());
1437	/// assert!(a.is_ascii_lowercase());
1438	/// assert!(g.is_ascii_lowercase());
1439	/// assert!(!zero.is_ascii_lowercase());
1440	/// assert!(!percent.is_ascii_lowercase());
1441	/// assert!(!space.is_ascii_lowercase());
1442	/// assert!(!lf.is_ascii_lowercase());
1443	/// assert!(!esc.is_ascii_lowercase());
1444	/// ```
1445	#[must_use]
1446	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1447	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1448	#[inline]
1449	pub const fn is_ascii_lowercase(&self) -> bool {
1450	matches!(*self, 'a'..='z')
1451	}
1452
1453	/// Checks if the value is an ASCII alphanumeric character:
1454	///
1455	/// - U+0041 'A' ..= U+005A 'Z', or
1456	/// - U+0061 'a' ..= U+007A 'z', or
1457	/// - U+0030 '0' ..= U+0039 '9'.
1458	///
1459	/// # Examples
1460	///
1461	/// ```
1462	/// let uppercase_a = 'A';
1463	/// let uppercase_g = 'G';
1464	/// let a = 'a';
1465	/// let g = 'g';
1466	/// let zero = '0';
1467	/// let percent = '%';
1468	/// let space = ' ';
1469	/// let lf = '`\n`';
1470	/// let esc = '`\x1b`';
1471	///
1472	/// assert!(uppercase_a.is_ascii_alphanumeric());
1473	/// assert!(uppercase_g.is_ascii_alphanumeric());
1474	/// assert!(a.is_ascii_alphanumeric());
1475	/// assert!(g.is_ascii_alphanumeric());
1476	/// assert!(zero.is_ascii_alphanumeric());
1477	/// assert!(!percent.is_ascii_alphanumeric());
1478	/// assert!(!space.is_ascii_alphanumeric());
1479	/// assert!(!lf.is_ascii_alphanumeric());
1480	/// assert!(!esc.is_ascii_alphanumeric());
1481	/// ```
1482	#[must_use]
1483	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1484	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1485	#[inline]
1486	pub const fn is_ascii_alphanumeric(&self) -> bool {
1487	matches!(self, '0'..='9') \| matches!(self, 'A'..='Z') \| matches!(*self, 'a'..='z')
1488	}
1489
1490	/// Checks if the value is an ASCII decimal digit:
1491	/// U+0030 '0' ..= U+0039 '9'.
1492	///
1493	/// # Examples
1494	///
1495	/// ```
1496	/// let uppercase_a = 'A';
1497	/// let uppercase_g = 'G';
1498	/// let a = 'a';
1499	/// let g = 'g';
1500	/// let zero = '0';
1501	/// let percent = '%';
1502	/// let space = ' ';
1503	/// let lf = '`\n`';
1504	/// let esc = '`\x1b`';
1505	///
1506	/// assert!(!uppercase_a.is_ascii_digit());
1507	/// assert!(!uppercase_g.is_ascii_digit());
1508	/// assert!(!a.is_ascii_digit());
1509	/// assert!(!g.is_ascii_digit());
1510	/// assert!(zero.is_ascii_digit());
1511	/// assert!(!percent.is_ascii_digit());
1512	/// assert!(!space.is_ascii_digit());
1513	/// assert!(!lf.is_ascii_digit());
1514	/// assert!(!esc.is_ascii_digit());
1515	/// ```
1516	#[must_use]
1517	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1518	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1519	#[inline]
1520	pub const fn is_ascii_digit(&self) -> bool {
1521	matches!(*self, '0'..='9')
1522	}
1523
1524	/// Checks if the value is an ASCII octal digit:
1525	/// U+0030 '0' ..= U+0037 '7'.
1526	///
1527	/// # Examples
1528	///
1529	/// ```
1530	/// #![feature(is_ascii_octdigit)]
1531	///
1532	/// let uppercase_a = 'A';
1533	/// let a = 'a';
1534	/// let zero = '0';
1535	/// let seven = '7';
1536	/// let nine = '9';
1537	/// let percent = '%';
1538	/// let lf = '`\n`';
1539	///
1540	/// assert!(!uppercase_a.is_ascii_octdigit());
1541	/// assert!(!a.is_ascii_octdigit());
1542	/// assert!(zero.is_ascii_octdigit());
1543	/// assert!(seven.is_ascii_octdigit());
1544	/// assert!(!nine.is_ascii_octdigit());
1545	/// assert!(!percent.is_ascii_octdigit());
1546	/// assert!(!lf.is_ascii_octdigit());
1547	/// ```
1548	#[must_use]
1549	#[unstable(feature = "is_ascii_octdigit", issue = "101288")]
1550	#[inline]
1551	pub const fn is_ascii_octdigit(&self) -> bool {
1552	matches!(*self, '0'..='7')
1553	}
1554
1555	/// Checks if the value is an ASCII hexadecimal digit:
1556	///
1557	/// - U+0030 '0' ..= U+0039 '9', or
1558	/// - U+0041 'A' ..= U+0046 'F', or
1559	/// - U+0061 'a' ..= U+0066 'f'.
1560	///
1561	/// # Examples
1562	///
1563	/// ```
1564	/// let uppercase_a = 'A';
1565	/// let uppercase_g = 'G';
1566	/// let a = 'a';
1567	/// let g = 'g';
1568	/// let zero = '0';
1569	/// let percent = '%';
1570	/// let space = ' ';
1571	/// let lf = '`\n`';
1572	/// let esc = '`\x1b`';
1573	///
1574	/// assert!(uppercase_a.is_ascii_hexdigit());
1575	/// assert!(!uppercase_g.is_ascii_hexdigit());
1576	/// assert!(a.is_ascii_hexdigit());
1577	/// assert!(!g.is_ascii_hexdigit());
1578	/// assert!(zero.is_ascii_hexdigit());
1579	/// assert!(!percent.is_ascii_hexdigit());
1580	/// assert!(!space.is_ascii_hexdigit());
1581	/// assert!(!lf.is_ascii_hexdigit());
1582	/// assert!(!esc.is_ascii_hexdigit());
1583	/// ```
1584	#[must_use]
1585	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1586	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1587	#[inline]
1588	pub const fn is_ascii_hexdigit(&self) -> bool {
1589	matches!(self, '0'..='9') \| matches!(self, 'A'..='F') \| matches!(*self, 'a'..='f')
1590	}
1591
1592	/// Checks if the value is an ASCII punctuation character:
1593	///
1594	/// - U+0021 ..= U+002F `! " # $ % & ' ( ) + , - . /`, or*
1595	/// - U+003A ..= U+0040 `: ; < = > ? @`, or
1596	/// - U+005B ..= U+0060 ``[ \ ] ^ _ ` ``, or
1597	/// - U+007B ..= U+007E `{ \| } ~`
1598	///
1599	/// # Examples
1600	///
1601	/// ```
1602	/// let uppercase_a = 'A';
1603	/// let uppercase_g = 'G';
1604	/// let a = 'a';
1605	/// let g = 'g';
1606	/// let zero = '0';
1607	/// let percent = '%';
1608	/// let space = ' ';
1609	/// let lf = '`\n`';
1610	/// let esc = '`\x1b`';
1611	///
1612	/// assert!(!uppercase_a.is_ascii_punctuation());
1613	/// assert!(!uppercase_g.is_ascii_punctuation());
1614	/// assert!(!a.is_ascii_punctuation());
1615	/// assert!(!g.is_ascii_punctuation());
1616	/// assert!(!zero.is_ascii_punctuation());
1617	/// assert!(percent.is_ascii_punctuation());
1618	/// assert!(!space.is_ascii_punctuation());
1619	/// assert!(!lf.is_ascii_punctuation());
1620	/// assert!(!esc.is_ascii_punctuation());
1621	/// ```
1622	#[must_use]
1623	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1624	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1625	#[inline]
1626	pub const fn is_ascii_punctuation(&self) -> bool {
1627	matches!(*self, '!'..='/')
1628	\| matches!(*self, ':'..='@')
1629	\| matches!(*self, '['..='`')
1630	\| matches!(*self, '{'..='~')
1631	}
1632
1633	/// Checks if the value is an ASCII graphic character:
1634	/// U+0021 '!' ..= U+007E '~'.
1635	///
1636	/// # Examples
1637	///
1638	/// ```
1639	/// let uppercase_a = 'A';
1640	/// let uppercase_g = 'G';
1641	/// let a = 'a';
1642	/// let g = 'g';
1643	/// let zero = '0';
1644	/// let percent = '%';
1645	/// let space = ' ';
1646	/// let lf = '`\n`';
1647	/// let esc = '`\x1b`';
1648	///
1649	/// assert!(uppercase_a.is_ascii_graphic());
1650	/// assert!(uppercase_g.is_ascii_graphic());
1651	/// assert!(a.is_ascii_graphic());
1652	/// assert!(g.is_ascii_graphic());
1653	/// assert!(zero.is_ascii_graphic());
1654	/// assert!(percent.is_ascii_graphic());
1655	/// assert!(!space.is_ascii_graphic());
1656	/// assert!(!lf.is_ascii_graphic());
1657	/// assert!(!esc.is_ascii_graphic());
1658	/// ```
1659	#[must_use]
1660	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1661	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1662	#[inline]
1663	pub const fn is_ascii_graphic(&self) -> bool {
1664	matches!(*self, '!'..='~')
1665	}
1666
1667	/// Checks if the value is an ASCII whitespace character:
1668	/// U+0020 SPACE, U+0009 HORIZONTAL TAB, U+000A LINE FEED,
1669	/// U+000C FORM FEED, or U+000D CARRIAGE RETURN.
1670	///
1671	/// Rust uses the WhatWG Infra Standard's [definition of ASCII
1672	/// whitespace][infra-aw]. There are several other definitions in
1673	/// wide use. For instance, [the POSIX locale][pct] includes
1674	/// U+000B VERTICAL TAB as well as all the above characters,
1675	/// but—from the very same specification—[the default rule for
1676	/// "field splitting" in the Bourne shell][bfs] considers only
1677	/// SPACE, HORIZONTAL TAB, and LINE FEED as whitespace.
1678	///
1679	/// If you are writing a program that will process an existing
1680	/// file format, check what that format's definition of whitespace is
1681	/// before using this function.
1682	///
1683	/// [infra-aw]: https://infra.spec.whatwg.org/#ascii-whitespace
1684	/// [pct]: https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap07.html#tag_07_03_01
1685	/// [bfs]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_06_05
1686	///
1687	/// # Examples
1688	///
1689	/// ```
1690	/// let uppercase_a = 'A';
1691	/// let uppercase_g = 'G';
1692	/// let a = 'a';
1693	/// let g = 'g';
1694	/// let zero = '0';
1695	/// let percent = '%';
1696	/// let space = ' ';
1697	/// let lf = '`\n`';
1698	/// let esc = '`\x1b`';
1699	///
1700	/// assert!(!uppercase_a.is_ascii_whitespace());
1701	/// assert!(!uppercase_g.is_ascii_whitespace());
1702	/// assert!(!a.is_ascii_whitespace());
1703	/// assert!(!g.is_ascii_whitespace());
1704	/// assert!(!zero.is_ascii_whitespace());
1705	/// assert!(!percent.is_ascii_whitespace());
1706	/// assert!(space.is_ascii_whitespace());
1707	/// assert!(lf.is_ascii_whitespace());
1708	/// assert!(!esc.is_ascii_whitespace());
1709	/// ```
1710	#[must_use]
1711	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1712	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1713	#[inline]
1714	pub const fn is_ascii_whitespace(&self) -> bool {
1715	matches!(*self, '`\t`' \| '`\n`' \| '`\x0C`' \| '`\r`' \| ' ')
1716	}
1717
1718	/// Checks if the value is an ASCII control character:
1719	/// U+0000 NUL ..= U+001F UNIT SEPARATOR, or U+007F DELETE.
1720	/// Note that most ASCII whitespace characters are control
1721	/// characters, but SPACE is not.
1722	///
1723	/// # Examples
1724	///
1725	/// ```
1726	/// let uppercase_a = 'A';
1727	/// let uppercase_g = 'G';
1728	/// let a = 'a';
1729	/// let g = 'g';
1730	/// let zero = '0';
1731	/// let percent = '%';
1732	/// let space = ' ';
1733	/// let lf = '`\n`';
1734	/// let esc = '`\x1b`';
1735	///
1736	/// assert!(!uppercase_a.is_ascii_control());
1737	/// assert!(!uppercase_g.is_ascii_control());
1738	/// assert!(!a.is_ascii_control());
1739	/// assert!(!g.is_ascii_control());
1740	/// assert!(!zero.is_ascii_control());
1741	/// assert!(!percent.is_ascii_control());
1742	/// assert!(!space.is_ascii_control());
1743	/// assert!(lf.is_ascii_control());
1744	/// assert!(esc.is_ascii_control());
1745	/// ```
1746	#[must_use]
1747	#[stable(feature = "ascii_ctype_on_intrinsics", since = "1.24.0")]
1748	#[rustc_const_stable(feature = "const_ascii_ctype_on_intrinsics", since = "1.47.0")]
1749	#[inline]
1750	pub const fn is_ascii_control(&self) -> bool {
1751	matches!(*self, '`\0`'..='`\x1F`' \| '`\x7F`')
1752	}
1753	}
1754
1755	pub(crate) struct EscapeDebugExtArgs {
1756	/// Escape Extended Grapheme codepoints?
1757	pub(crate) escape_grapheme_extended: bool,
1758
1759	/// Escape single quotes?
1760	pub(crate) escape_single_quote: bool,
1761
1762	/// Escape double quotes?
1763	pub(crate) escape_double_quote: bool,
1764	}
1765
1766	impl EscapeDebugExtArgs {
1767	pub(crate) const ESCAPE_ALL: Self = Self {
1768	escape_grapheme_extended: `true`,
1769	escape_single_quote: `true`,
1770	escape_double_quote: `true`,
1771	};
1772	}
1773
1774	#[inline]
1775	#[must_use]
1776	const fn len_utf8(code: u32) -> usize {
1777	match code {
1778	..MAX_ONE_B => `1`,
1779	..MAX_TWO_B => `2`,
1780	..MAX_THREE_B => `3`,
1781	_ => `4`,
1782	}
1783	}
1784
1785	#[inline]
1786	#[must_use]
1787	const fn len_utf16(code: u32) -> usize {
1788	if (code & `0xFFFF`) == code { `1` } else { `2` }
1789	}
1790
1791	/// Encodes a raw `u32` value as UTF-8 into the provided byte buffer,
1792	/// and then returns the subslice of the buffer that contains the encoded character.
1793	///
1794	/// Unlike `char::encode_utf8`, this method also handles codepoints in the surrogate range.
1795	/// (Creating a `char` in the surrogate range is UB.)
1796	/// The result is valid [generalized UTF-8] but not valid UTF-8.
1797	///
1798	/// [generalized UTF-8]: https://simonsapin.github.io/wtf-8/#generalized-utf8
1799	///
1800	/// # Panics
1801	///
1802	/// Panics if the buffer is not large enough.
1803	/// A buffer of length four is large enough to encode any `char`.
1804	#[unstable(feature = "char_internals", reason = "exposed only for libstd", issue = "none")]
1805	#[doc(hidden)]
1806	#[inline]
1807	pub const fn encode_utf8_raw(code: u32, dst: &mut [u8]) -> &mut [u8] {
1808	let len = len_utf8(code);
1809	match (len, &mut *dst) {
1810	(`1`, [a, ..]) => {
1811	*a = code as u8;
1812	}
1813	(`2`, [a, b, ..]) => {
1814	*a = (code >> `6` & `0x1F`) as u8 \| TAG_TWO_B;
1815	*b = (code & `0x3F`) as u8 \| TAG_CONT;
1816	}
1817	(`3`, [a, b, c, ..]) => {
1818	*a = (code >> `12` & `0x0F`) as u8 \| TAG_THREE_B;
1819	*b = (code >> `6` & `0x3F`) as u8 \| TAG_CONT;
1820	*c = (code & `0x3F`) as u8 \| TAG_CONT;
1821	}
1822	(`4`, [a, b, c, d, ..]) => {
1823	*a = (code >> `18` & `0x07`) as u8 \| TAG_FOUR_B;
1824	*b = (code >> `12` & `0x3F`) as u8 \| TAG_CONT;
1825	*c = (code >> `6` & `0x3F`) as u8 \| TAG_CONT;
1826	*d = (code & `0x3F`) as u8 \| TAG_CONT;
1827	}
1828	_ => {
1829	const_panic!(
1830	"encode_utf8: buffer does not have enough bytes to encode code point",
1831	"encode_utf8: need {len} bytes to encode U+{code:`04`X} but buffer has just {dst_len}",
1832	code: u32 = code,
1833	len: usize = len,
1834	dst_len: usize = dst.len(),
1835	)
1836	}
1837	};
1838	// SAFETY: `<&mut [u8]>::as_mut_ptr` is guaranteed to return a valid pointer and `len` has been tested to be within bounds.
1839	unsafe { slice::from_raw_parts_mut(dst.as_mut_ptr(), len) }
1840	}
1841
1842	/// Encodes a raw `u32` value as native endian UTF-16 into the provided `u16` buffer,
1843	/// and then returns the subslice of the buffer that contains the encoded character.
1844	///
1845	/// Unlike `char::encode_utf16`, this method also handles codepoints in the surrogate range.
1846	/// (Creating a `char` in the surrogate range is UB.)
1847	///
1848	/// # Panics
1849	///
1850	/// Panics if the buffer is not large enough.
1851	/// A buffer of length 2 is large enough to encode any `char`.
1852	#[unstable(feature = "char_internals", reason = "exposed only for libstd", issue = "none")]
1853	#[doc(hidden)]
1854	#[inline]
1855	pub const fn encode_utf16_raw(mut code: u32, dst: &mut [u16]) -> &mut [u16] {
1856	let len: usize = len_utf16(code);
1857	match (len, &mut *dst) {
1858	(`1`, [a: &mut u16, ..]) => {
1859	a = code as u16*;
1860	}
1861	(`2`, [a: &mut u16, b: &mut u16, ..]) => {
1862	code -= `0x1_0000`;
1863	a = (code >> `10`) as u16* \| `0xD800`;
1864	b = (code & `0x3FF`) as u16* \| `0xDC00`;
1865	}
1866	_ => {
1867	const_panic!(
1868	"encode_utf16: buffer does not have enough bytes to encode code point",
1869	"encode_utf16: need {len} bytes to encode U+{code:`04`X} but buffer has just {dst_len}",
1870	code: u32 = code,
1871	len: usize = len,
1872	dst_len: usize = dst.len(),
1873	)
1874	}
1875	};
1876	// SAFETY: `<&mut [u16]>::as_mut_ptr` is guaranteed to return a valid pointer and `len` has been tested to be within bounds.
1877	unsafe { slice::from_raw_parts_mut(data:dst.as_mut_ptr(), len) }
1878	}
1879

Provided by KDAB

Definitions