1// -*- C++ -*-
2//===----------------------------------------------------------------------===//
3//
4// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
5// See https://llvm.org/LICENSE.txt for license information.
6// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
7//
8//===----------------------------------------------------------------------===//
9
10#ifndef _LIBCPP___FORMAT_UNICODE_H
11#define _LIBCPP___FORMAT_UNICODE_H
12
13#include <__assert>
14#include <__bit/countl.h>
15#include <__concepts/same_as.h>
16#include <__config>
17#include <__format/extended_grapheme_cluster_table.h>
18#include <__format/indic_conjunct_break_table.h>
19#include <__iterator/concepts.h>
20#include <__iterator/readable_traits.h> // iter_value_t
21#include <__utility/unreachable.h>
22#include <string_view>
23
24#if !defined(_LIBCPP_HAS_NO_PRAGMA_SYSTEM_HEADER)
25# pragma GCC system_header
26#endif
27
28_LIBCPP_BEGIN_NAMESPACE_STD
29
30#if _LIBCPP_STD_VER >= 20
31
32namespace __unicode {
33
34// Helper struct for the result of a consume operation.
35//
36// The status value for a correct code point is 0. This allows a valid value to
37// be used without masking.
38// When the decoding fails it know the number of code units affected. For the
39// current use-cases that value is not needed, therefore it is not stored.
40// The escape routine needs the number of code units for both a valid and
41// invalid character and keeps track of it itself. Doing it in this result
42// unconditionally would give some overhead when the value is unneeded.
43struct __consume_result {
44 // When __status == __ok it contains the decoded code point.
45 // Else it contains the replacement character U+FFFD
46 char32_t __code_point : 31;
47
48 enum : char32_t {
49 // Consumed a well-formed code point.
50 __ok = 0,
51 // Encountered invalid UTF-8
52 __error = 1
53 } __status : 1 {__ok};
54};
55static_assert(sizeof(__consume_result) == sizeof(char32_t));
56
57# if _LIBCPP_HAS_UNICODE
58
59/// Implements the grapheme cluster boundary rules
60///
61/// These rules are used to implement format's width estimation as stated in
62/// [format.string.std]/11
63///
64/// The Standard refers to UAX \#29 for Unicode 12.0.0
65/// https://www.unicode.org/reports/tr29/#Grapheme_Cluster_Boundary_Rules
66///
67/// The data tables used are
68/// https://www.unicode.org/Public/UCD/latest/ucd/auxiliary/GraphemeBreakProperty.txt
69/// https://www.unicode.org/Public/UCD/latest/ucd/emoji/emoji-data.txt
70/// https://www.unicode.org/Public/UCD/latest/ucd/auxiliary/GraphemeBreakTest.txt (for testing only)
71
72inline constexpr char32_t __replacement_character = U'\ufffd';
73
74// The error of a consume operation.
75//
76// This sets the code point to the replacement character. This code point does
77// not participate in the grapheme clustering, so grapheme clustering code can
78// ignore the error status and always use the code point.
79inline constexpr __consume_result __consume_result_error{__replacement_character, __consume_result::__error};
80
81[[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool __is_high_surrogate(char32_t __value) {
82 return __value >= 0xd800 && __value <= 0xdbff;
83}
84
85[[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool __is_low_surrogate(char32_t __value) {
86 return __value >= 0xdc00 && __value <= 0xdfff;
87}
88
89// https://www.unicode.org/glossary/#surrogate_code_point
90[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline constexpr bool __is_surrogate(char32_t __value) {
91 return __value >= 0xd800 && __value <= 0xdfff;
92}
93
94// https://www.unicode.org/glossary/#code_point
95[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline constexpr bool __is_code_point(char32_t __value) {
96 return __value <= 0x10ffff;
97}
98
99// https://www.unicode.org/glossary/#unicode_scalar_value
100[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline constexpr bool __is_scalar_value(char32_t __value) {
101 return __unicode::__is_code_point(__value) && !__unicode::__is_surrogate(__value);
102}
103
104template <contiguous_iterator _Iterator>
105 requires same_as<iter_value_t<_Iterator>, char>
106_LIBCPP_HIDE_FROM_ABI constexpr bool __is_continuation(_Iterator __char, int __count) {
107 do {
108 if ((*__char & 0b1100'0000) != 0b1000'0000)
109 return false;
110 --__count;
111 ++__char;
112 } while (__count);
113 return true;
114}
115
116/// Helper class to extract a code unit from a Unicode character range.
117///
118/// The stored range is a view. There are multiple specialization for different
119/// character types.
120template <class _CharT>
121class __code_point_view;
122
123/// UTF-8 specialization.
124template <>
125class __code_point_view<char> {
126 using _Iterator _LIBCPP_NODEBUG = basic_string_view<char>::const_iterator;
127
128public:
129 _LIBCPP_HIDE_FROM_ABI constexpr explicit __code_point_view(_Iterator __first, _Iterator __last)
130 : __first_(__first), __last_(__last) {}
131
132 _LIBCPP_HIDE_FROM_ABI constexpr bool __at_end() const noexcept { return __first_ == __last_; }
133 _LIBCPP_HIDE_FROM_ABI constexpr _Iterator __position() const noexcept { return __first_; }
134
135 // https://www.unicode.org/versions/latest/ch03.pdf#G7404
136 // Based on Table 3-7, Well-Formed UTF-8 Byte Sequences
137 //
138 // Code Points First Byte Second Byte Third Byte Fourth Byte Remarks
139 // U+0000..U+007F 00..7F U+0000..U+007F 1 code unit range
140 // C0..C1 80..BF invalid overlong encoding
141 // U+0080..U+07FF C2..DF 80..BF U+0080..U+07FF 2 code unit range
142 // E0 80..9F 80..BF invalid overlong encoding
143 // U+0800..U+0FFF E0 A0..BF 80..BF U+0800..U+FFFF 3 code unit range
144 // U+1000..U+CFFF E1..EC 80..BF 80..BF
145 // U+D000..U+D7FF ED 80..9F 80..BF
146 // U+D800..U+DFFF ED A0..BF 80..BF invalid encoding of surrogate code point
147 // U+E000..U+FFFF EE..EF 80..BF 80..BF
148 // F0 80..8F 80..BF 80..BF invalid overlong encoding
149 // U+10000..U+3FFFF F0 90..BF 80..BF 80..BF U+10000..U+10FFFF 4 code unit range
150 // U+40000..U+FFFFF F1..F3 80..BF 80..BF 80..BF
151 // U+100000..U+10FFFF F4 80..8F 80..BF 80..BF
152 // F4 90..BF 80..BF 80..BF U+110000.. invalid code point range
153 //
154 // Unlike other parsers, these invalid entries are tested after decoding.
155 // - The parser always needs to consume these code units
156 // - The code is optimized for well-formed UTF-8
157 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr __consume_result __consume() noexcept {
158 _LIBCPP_ASSERT_INTERNAL(__first_ != __last_, "can't move beyond the end of input");
159
160 // Based on the number of leading 1 bits the number of code units in the
161 // code point can be determined. See
162 // https://en.wikipedia.org/wiki/UTF-8#Encoding
163 switch (std::countl_one(t: static_cast<unsigned char>(*__first_))) {
164 case 0:
165 return {static_cast<unsigned char>(*__first_++)};
166
167 case 2: {
168 if (__last_ - __first_ < 2 || !__unicode::__is_continuation(char: __first_ + 1, count: 1)) [[unlikely]]
169 break;
170
171 char32_t __value = static_cast<unsigned char>(*__first_++) & 0x1f;
172 __value <<= 6;
173 __value |= static_cast<unsigned char>(*__first_++) & 0x3f;
174
175 // These values should be encoded in 1 UTF-8 code unit.
176 if (__value < 0x0080) [[unlikely]]
177 return __consume_result_error;
178
179 return {__value};
180 }
181
182 case 3: {
183 if (__last_ - __first_ < 3 || !__unicode::__is_continuation(char: __first_ + 1, count: 2)) [[unlikely]]
184 break;
185
186 char32_t __value = static_cast<unsigned char>(*__first_++) & 0x0f;
187 __value <<= 6;
188 __value |= static_cast<unsigned char>(*__first_++) & 0x3f;
189 __value <<= 6;
190 __value |= static_cast<unsigned char>(*__first_++) & 0x3f;
191
192 // These values should be encoded in 1 or 2 UTF-8 code units.
193 if (__value < 0x0800) [[unlikely]]
194 return __consume_result_error;
195
196 // A surrogate value is always encoded in 3 UTF-8 code units.
197 if (__unicode::__is_surrogate(__value)) [[unlikely]]
198 return __consume_result_error;
199
200 return {__value};
201 }
202
203 case 4: {
204 if (__last_ - __first_ < 4 || !__unicode::__is_continuation(char: __first_ + 1, count: 3)) [[unlikely]]
205 break;
206
207 char32_t __value = static_cast<unsigned char>(*__first_++) & 0x07;
208 __value <<= 6;
209 __value |= static_cast<unsigned char>(*__first_++) & 0x3f;
210 __value <<= 6;
211 __value |= static_cast<unsigned char>(*__first_++) & 0x3f;
212 __value <<= 6;
213 __value |= static_cast<unsigned char>(*__first_++) & 0x3f;
214
215 // These values should be encoded in 1, 2, or 3 UTF-8 code units.
216 if (__value < 0x10000) [[unlikely]]
217 return __consume_result_error;
218
219 // A value too large is always encoded in 4 UTF-8 code units.
220 if (!__unicode::__is_code_point(__value)) [[unlikely]]
221 return __consume_result_error;
222
223 return {__value};
224 }
225 }
226 // An invalid number of leading ones can be garbage or a code unit in the
227 // middle of a code point. By consuming one code unit the parser may get
228 // "in sync" after a few code units.
229 ++__first_;
230 return __consume_result_error;
231 }
232
233private:
234 _Iterator __first_;
235 _Iterator __last_;
236};
237
238# if _LIBCPP_HAS_WIDE_CHARACTERS
239_LIBCPP_HIDE_FROM_ABI constexpr bool __is_surrogate_pair_high(wchar_t __value) {
240 return __value >= 0xd800 && __value <= 0xdbff;
241}
242
243_LIBCPP_HIDE_FROM_ABI constexpr bool __is_surrogate_pair_low(wchar_t __value) {
244 return __value >= 0xdc00 && __value <= 0xdfff;
245}
246
247/// This specialization depends on the size of wchar_t
248/// - 2 UTF-16 (for example Windows and AIX)
249/// - 4 UTF-32 (for example Linux)
250template <>
251class __code_point_view<wchar_t> {
252 using _Iterator _LIBCPP_NODEBUG = typename basic_string_view<wchar_t>::const_iterator;
253
254public:
255 static_assert(sizeof(wchar_t) == 2 || sizeof(wchar_t) == 4, "sizeof(wchar_t) has a not implemented value");
256
257 _LIBCPP_HIDE_FROM_ABI constexpr explicit __code_point_view(_Iterator __first, _Iterator __last)
258 : __first_(__first), __last_(__last) {}
259
260 _LIBCPP_HIDE_FROM_ABI constexpr _Iterator __position() const noexcept { return __first_; }
261 _LIBCPP_HIDE_FROM_ABI constexpr bool __at_end() const noexcept { return __first_ == __last_; }
262
263 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr __consume_result __consume() noexcept {
264 _LIBCPP_ASSERT_INTERNAL(__first_ != __last_, "can't move beyond the end of input");
265
266 char32_t __value = static_cast<char32_t>(*__first_++);
267 if constexpr (sizeof(wchar_t) == 2) {
268 if (__unicode::__is_low_surrogate(__value)) [[unlikely]]
269 return __consume_result_error;
270
271 if (__unicode::__is_high_surrogate(__value)) {
272 if (__first_ == __last_ || !__unicode::__is_low_surrogate(value: static_cast<char32_t>(*__first_))) [[unlikely]]
273 return __consume_result_error;
274
275 __value -= 0xd800;
276 __value <<= 10;
277 __value += static_cast<char32_t>(*__first_++) - 0xdc00;
278 __value += 0x10000;
279
280 if (!__unicode::__is_code_point(__value)) [[unlikely]]
281 return __consume_result_error;
282 }
283 } else {
284 if (!__unicode::__is_scalar_value(__value)) [[unlikely]]
285 return __consume_result_error;
286 }
287
288 return {__value};
289 }
290
291private:
292 _Iterator __first_;
293 _Iterator __last_;
294};
295# endif // _LIBCPP_HAS_WIDE_CHARACTERS
296
297// State machine to implement the Extended Grapheme Cluster Boundary
298//
299// The exact rules may change between Unicode versions.
300// This implements the extended rules see
301// https://www.unicode.org/reports/tr29/#Grapheme_Cluster_Boundaries
302class __extended_grapheme_cluster_break {
303 using __EGC_property _LIBCPP_NODEBUG = __extended_grapheme_custer_property_boundary::__property;
304 using __inCB_property _LIBCPP_NODEBUG = __indic_conjunct_break::__property;
305
306public:
307 _LIBCPP_HIDE_FROM_ABI constexpr explicit __extended_grapheme_cluster_break(char32_t __first_code_point)
308 : __prev_code_point_(__first_code_point),
309 __prev_property_(__extended_grapheme_custer_property_boundary::__get_property(code_point: __first_code_point)) {
310 // Initializes the active rule.
311 if (__prev_property_ == __EGC_property::__Extended_Pictographic)
312 __active_rule_ = __rule::__GB11_emoji;
313 else if (__prev_property_ == __EGC_property::__Regional_Indicator)
314 __active_rule_ = __rule::__GB12_GB13_regional_indicator;
315 else if (__indic_conjunct_break::__get_property(code_point: __first_code_point) == __inCB_property::__Consonant)
316 __active_rule_ = __rule::__GB9c_indic_conjunct_break;
317 }
318
319 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool operator()(char32_t __next_code_point) {
320 __EGC_property __next_property = __extended_grapheme_custer_property_boundary::__get_property(code_point: __next_code_point);
321 bool __result = __evaluate(__next_code_point, __next_property);
322 __prev_code_point_ = __next_code_point;
323 __prev_property_ = __next_property;
324 return __result;
325 }
326
327 // The code point whose break propery are considered during the next
328 // evaluation cyle.
329 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr char32_t __current_code_point() const { return __prev_code_point_; }
330
331private:
332 // The naming of the identifiers matches the Unicode standard.
333 // NOLINTBEGIN(readability-identifier-naming)
334
335 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool
336 __evaluate(char32_t __next_code_point, __EGC_property __next_property) {
337 switch (__active_rule_) {
338 case __rule::__none:
339 return __evaluate_none(__next_code_point, __next_property);
340 case __rule::__GB9c_indic_conjunct_break:
341 return __evaluate_GB9c_indic_conjunct_break(__next_code_point, __next_property);
342 case __rule::__GB11_emoji:
343 return __evaluate_GB11_emoji(__next_code_point, __next_property);
344 case __rule::__GB12_GB13_regional_indicator:
345 return __evaluate_GB12_GB13_regional_indicator(__next_code_point, __next_property);
346 }
347 __libcpp_unreachable();
348 }
349
350 _LIBCPP_HIDE_FROM_ABI constexpr bool __evaluate_none(char32_t __next_code_point, __EGC_property __next_property) {
351 // *** Break at the start and end of text, unless the text is empty. ***
352
353 _LIBCPP_ASSERT_INTERNAL(__prev_property_ != __EGC_property::__sot, "should be handled in the constructor"); // GB1
354 _LIBCPP_ASSERT_INTERNAL(__prev_property_ != __EGC_property::__eot, "should be handled by our caller"); // GB2
355
356 // *** Do not break between a CR and LF. Otherwise, break before and after controls. ***
357 if (__prev_property_ == __EGC_property::__CR && __next_property == __EGC_property::__LF) // GB3
358 return false;
359
360 if (__prev_property_ == __EGC_property::__Control || __prev_property_ == __EGC_property::__CR ||
361 __prev_property_ == __EGC_property::__LF) // GB4
362 return true;
363
364 if (__next_property == __EGC_property::__Control || __next_property == __EGC_property::__CR ||
365 __next_property == __EGC_property::__LF) // GB5
366 return true;
367
368 // *** Do not break Hangul syllable sequences. ***
369 if (__prev_property_ == __EGC_property::__L &&
370 (__next_property == __EGC_property::__L || __next_property == __EGC_property::__V ||
371 __next_property == __EGC_property::__LV || __next_property == __EGC_property::__LVT)) // GB6
372 return false;
373
374 if ((__prev_property_ == __EGC_property::__LV || __prev_property_ == __EGC_property::__V) &&
375 (__next_property == __EGC_property::__V || __next_property == __EGC_property::__T)) // GB7
376 return false;
377
378 if ((__prev_property_ == __EGC_property::__LVT || __prev_property_ == __EGC_property::__T) &&
379 __next_property == __EGC_property::__T) // GB8
380 return false;
381
382 // *** Do not break before extending characters or ZWJ. ***
383 if (__next_property == __EGC_property::__Extend || __next_property == __EGC_property::__ZWJ)
384 return false; // GB9
385
386 // *** Do not break before SpacingMarks, or after Prepend characters. ***
387 if (__next_property == __EGC_property::__SpacingMark) // GB9a
388 return false;
389
390 if (__prev_property_ == __EGC_property::__Prepend) // GB9b
391 return false;
392
393 // *** Do not break within certain combinations with Indic_Conjunct_Break (InCB)=Linker. ***
394 if (__indic_conjunct_break::__get_property(code_point: __next_code_point) == __inCB_property::__Consonant) {
395 __active_rule_ = __rule::__GB9c_indic_conjunct_break;
396 __GB9c_indic_conjunct_break_state_ = __GB9c_indic_conjunct_break_state::__Consonant;
397 return true;
398 }
399
400 // *** Do not break within emoji modifier sequences or emoji zwj sequences. ***
401 if (__next_property == __EGC_property::__Extended_Pictographic) {
402 __active_rule_ = __rule::__GB11_emoji;
403 __GB11_emoji_state_ = __GB11_emoji_state::__Extended_Pictographic;
404 return true;
405 }
406
407 // *** Do not break within emoji flag sequences ***
408
409 // That is, do not break between regional indicator (RI) symbols if there
410 // is an odd number of RI characters before the break point.
411 if (__next_property == __EGC_property::__Regional_Indicator) { // GB12 + GB13
412 __active_rule_ = __rule::__GB12_GB13_regional_indicator;
413 return true;
414 }
415
416 // *** Otherwise, break everywhere. ***
417 return true; // GB999
418 }
419
420 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool
421 __evaluate_GB9c_indic_conjunct_break(char32_t __next_code_point, __EGC_property __next_property) {
422 __inCB_property __break = __indic_conjunct_break::__get_property(code_point: __next_code_point);
423 if (__break == __inCB_property::__none) {
424 __active_rule_ = __rule::__none;
425 return __evaluate_none(__next_code_point, __next_property);
426 }
427
428 switch (__GB9c_indic_conjunct_break_state_) {
429 case __GB9c_indic_conjunct_break_state::__Consonant:
430 if (__break == __inCB_property::__Extend) {
431 return false;
432 }
433 if (__break == __inCB_property::__Linker) {
434 __GB9c_indic_conjunct_break_state_ = __GB9c_indic_conjunct_break_state::__Linker;
435 return false;
436 }
437 __active_rule_ = __rule::__none;
438 return __evaluate_none(__next_code_point, __next_property);
439
440 case __GB9c_indic_conjunct_break_state::__Linker:
441 if (__break == __inCB_property::__Extend) {
442 return false;
443 }
444 if (__break == __inCB_property::__Linker) {
445 return false;
446 }
447 if (__break == __inCB_property::__Consonant) {
448 __GB9c_indic_conjunct_break_state_ = __GB9c_indic_conjunct_break_state::__Consonant;
449 return false;
450 }
451 __active_rule_ = __rule::__none;
452 return __evaluate_none(__next_code_point, __next_property);
453 }
454 __libcpp_unreachable();
455 }
456
457 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool
458 __evaluate_GB11_emoji(char32_t __next_code_point, __EGC_property __next_property) {
459 switch (__GB11_emoji_state_) {
460 case __GB11_emoji_state::__Extended_Pictographic:
461 if (__next_property == __EGC_property::__Extend) {
462 __GB11_emoji_state_ = __GB11_emoji_state::__Extend;
463 return false;
464 }
465 [[fallthrough]];
466 case __GB11_emoji_state::__Extend:
467 if (__next_property == __EGC_property::__ZWJ) {
468 __GB11_emoji_state_ = __GB11_emoji_state::__ZWJ;
469 return false;
470 }
471 if (__next_property == __EGC_property::__Extend)
472 return false;
473 __active_rule_ = __rule::__none;
474 return __evaluate_none(__next_code_point, __next_property);
475
476 case __GB11_emoji_state::__ZWJ:
477 if (__next_property == __EGC_property::__Extended_Pictographic) {
478 __GB11_emoji_state_ = __GB11_emoji_state::__Extended_Pictographic;
479 return false;
480 }
481 __active_rule_ = __rule::__none;
482 return __evaluate_none(__next_code_point, __next_property);
483 }
484 __libcpp_unreachable();
485 }
486
487 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool
488 __evaluate_GB12_GB13_regional_indicator(char32_t __next_code_point, __EGC_property __next_property) {
489 __active_rule_ = __rule::__none;
490 if (__next_property == __EGC_property::__Regional_Indicator)
491 return false;
492 return __evaluate_none(__next_code_point, __next_property);
493 }
494
495 char32_t __prev_code_point_;
496 __EGC_property __prev_property_;
497
498 enum class __rule {
499 __none,
500 __GB9c_indic_conjunct_break,
501 __GB11_emoji,
502 __GB12_GB13_regional_indicator,
503 };
504 __rule __active_rule_ = __rule::__none;
505
506 enum class __GB11_emoji_state {
507 __Extended_Pictographic,
508 __Extend,
509 __ZWJ,
510 };
511 __GB11_emoji_state __GB11_emoji_state_ = __GB11_emoji_state::__Extended_Pictographic;
512
513 enum class __GB9c_indic_conjunct_break_state {
514 __Consonant,
515 __Linker,
516 };
517
518 __GB9c_indic_conjunct_break_state __GB9c_indic_conjunct_break_state_ = __GB9c_indic_conjunct_break_state::__Consonant;
519
520 // NOLINTEND(readability-identifier-naming)
521};
522
523/// Helper class to extract an extended grapheme cluster from a Unicode character range.
524///
525/// This function is used to determine the column width of an extended grapheme
526/// cluster. In order to do that only the first code point is evaluated.
527/// Therefore only this code point is extracted.
528template <class _CharT>
529class __extended_grapheme_cluster_view {
530 using _Iterator _LIBCPP_NODEBUG = typename basic_string_view<_CharT>::const_iterator;
531
532public:
533 _LIBCPP_HIDE_FROM_ABI constexpr explicit __extended_grapheme_cluster_view(_Iterator __first, _Iterator __last)
534 : __code_point_view_(__first, __last), __at_break_(__code_point_view_.__consume().__code_point) {}
535
536 struct __cluster {
537 /// The first code point of the extended grapheme cluster.
538 ///
539 /// The first code point is used to estimate the width of the extended
540 /// grapheme cluster.
541 char32_t __code_point_;
542
543 /// Points one beyond the last code unit in the extended grapheme cluster.
544 ///
545 /// It's expected the caller has the start position and thus can determine
546 /// the code unit range of the extended grapheme cluster.
547 _Iterator __last_;
548 };
549
550 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr __cluster __consume() {
551 char32_t __code_point = __at_break_.__current_code_point();
552 _Iterator __position = __code_point_view_.__position();
553 while (!__code_point_view_.__at_end()) {
554 if (__at_break_(__code_point_view_.__consume().__code_point))
555 break;
556 __position = __code_point_view_.__position();
557 }
558 return {__code_point, __position};
559 }
560
561private:
562 __code_point_view<_CharT> __code_point_view_;
563 __extended_grapheme_cluster_break __at_break_;
564};
565
566template <contiguous_iterator _Iterator>
567__extended_grapheme_cluster_view(_Iterator, _Iterator) -> __extended_grapheme_cluster_view<iter_value_t<_Iterator>>;
568
569# else // _LIBCPP_HAS_UNICODE
570
571// For ASCII every character is a "code point".
572// This makes it easier to write code agnostic of the _LIBCPP_HAS_UNICODE define.
573template <class _CharT>
574class __code_point_view {
575 using _Iterator _LIBCPP_NODEBUG = typename basic_string_view<_CharT>::const_iterator;
576
577public:
578 _LIBCPP_HIDE_FROM_ABI constexpr explicit __code_point_view(_Iterator __first, _Iterator __last)
579 : __first_(__first), __last_(__last) {}
580
581 _LIBCPP_HIDE_FROM_ABI constexpr bool __at_end() const noexcept { return __first_ == __last_; }
582 _LIBCPP_HIDE_FROM_ABI constexpr _Iterator __position() const noexcept { return __first_; }
583
584 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr __consume_result __consume() noexcept {
585 _LIBCPP_ASSERT_INTERNAL(__first_ != __last_, "can't move beyond the end of input");
586 return {static_cast<char32_t>(*__first_++)};
587 }
588
589private:
590 _Iterator __first_;
591 _Iterator __last_;
592};
593
594# endif // _LIBCPP_HAS_UNICODE
595
596} // namespace __unicode
597
598#endif // _LIBCPP_STD_VER >= 20
599
600_LIBCPP_END_NAMESPACE_STD
601
602#endif // _LIBCPP___FORMAT_UNICODE_H
603