1//===----------------------------------------------------------------------===//
2//
3// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4// See https://llvm.org/LICENSE.txt for license information.
5// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6//
7//===----------------------------------------------------------------------===//
8
9// UNSUPPORTED: c++03, c++11, c++14, c++17
10// UNSUPPORTED: GCC-ALWAYS_INLINE-FIXME
11
12// This version runs the test when the platform has Unicode support.
13// UNSUPPORTED: libcpp-has-no-unicode
14
15// TODO FMT This test should not require std::to_chars(floating-point)
16// XFAIL: availability-fp_to_chars-missing
17
18// <format>
19
20// Tests the Unicode width support of the standard format specifiers.
21// It tests [format.string.std]/8 - 11:
22// - Properly determining the estimated with of a unicode string.
23// - Properly truncating to the wanted maximum width.
24
25// More specific extended grapheme cluster boundary rules are tested in
26// test/libcxx/utilities/format/format.string/format.string.std/extended_grapheme_cluster.pass.cpp
27// this test is based on test data provided by the Unicode Consortium.
28
29#include <format>
30#include <cassert>
31#include <vector>
32
33#include "make_string.h"
34#include "test_macros.h"
35#include "string_literal.h"
36#include "test_format_string.h"
37
38#ifndef TEST_HAS_NO_LOCALIZATION
39# include <iostream>
40# include <type_traits>
41#endif
42
43#define SV(S) MAKE_STRING_VIEW(CharT, S)
44
45template < class CharT, class... Args>
46void check(std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
47 std::basic_string<CharT> out = std::format(fmt, std::forward<Args>(args)...);
48#ifndef TEST_HAS_NO_LOCALIZATION
49 if constexpr (std::same_as<CharT, char>)
50 if (out != expected)
51 std::cerr << "\nFormat string " << fmt.get() << "\nExpected output " << expected << "\nActual output " << out
52 << '\n';
53#endif
54 assert(out == expected);
55};
56
57template <class CharT>
58static void test_single_code_point_fill() {
59 //*** 1-byte code points ***
60 check(SV("* *"), SV("{:*^3}"), SV(" "));
61 check(SV("*~*"), SV("{:*^3}"), SV("~"));
62
63 //*** 2-byte code points ***
64 check(SV("*\u00a1*"), SV("{:*^3}"), SV("\u00a1")); // INVERTED EXCLAMATION MARK
65 check(SV("*\u07ff*"), SV("{:*^3}"), SV("\u07ff")); // NKO TAMAN SIGN
66
67 //*** 3-byte code points ***
68 check(SV("*\u0800*"), SV("{:*^3}"), SV("\u0800")); // SAMARITAN LETTER ALAF
69 check(SV("*\ufffd*"), SV("{:*^3}"), SV("\ufffd")); // REPLACEMENT CHARACTER
70
71 // 2 column ranges
72 check(SV("*\u1100*"), SV("{:*^4}"), SV("\u1100")); // HANGUL CHOSEONG KIYEOK
73 check(SV("*\u115f*"), SV("{:*^4}"), SV("\u115f")); // HANGUL CHOSEONG FILLER
74
75 check(SV("*\u2329*"), SV("{:*^4}"), SV("\u2329")); // LEFT-POINTING ANGLE BRACKET
76 check(SV("*\u232a*"), SV("{:*^4}"), SV("\u232a")); // RIGHT-POINTING ANGLE BRACKET
77
78 check(SV("*\u2e80*"), SV("{:*^4}"), SV("\u2e80")); // CJK RADICAL REPEAT
79 check(SV("*\u303e*"), SV("{:*^4}"), SV("\u303e")); // IDEOGRAPHIC VARIATION INDICATOR
80
81 check(SV("*\u3041*"), SV("{:*^4}"), SV("\u3041")); // U+3041 HIRAGANA LETTER SMALL A
82 check(SV("*\ua4d0*"), SV("{:*^3}"), SV("\ua4d0")); // U+A4D0 LISU LETTER BA
83
84 check(SV("*\uac00*"), SV("{:*^4}"), SV("\uac00")); // <Hangul Syllable, First>
85 check(SV("*\ud7a3*"), SV("{:*^4}"), SV("\ud7a3")); // Hangul Syllable Hih
86
87 check(SV("*\uf900*"), SV("{:*^4}"), SV("\uf900")); // CJK COMPATIBILITY IDEOGRAPH-F900
88 check(SV("*\ufaff*"), SV("{:*^4}"), SV("\ufaff")); // U+FB00 LATIN SMALL LIGATURE FF
89
90 check(SV("*\ufe10*"), SV("{:*^4}"), SV("\ufe10")); // PRESENTATION FORM FOR VERTICAL COMMA
91 check(SV("*\ufe19*"), SV("{:*^4}"), SV("\ufe19")); // PRESENTATION FORM FOR VERTICAL HORIZONTAL ELLIPSIS
92
93 check(SV("*\ufe30*"), SV("{:*^4}"), SV("\ufe30")); // PRESENTATION FORM FOR VERTICAL TWO DOT LEADER
94 check(SV("*\ufe70*"), SV("{:*^3}"), SV("\ufe70")); // U+FE70 ARABIC FATHATAN ISOLATED FORM
95
96 check(SV("*\uff01*"), SV("{:*^4}"), SV("\uff01")); // U+FF01 FULLWIDTH EXCLAMATION MARK
97 check(SV("*\uff60*"), SV("{:*^4}"), SV("\uff60")); // FULLWIDTH RIGHT WHITE PARENTHESIS
98
99 check(SV("*\uffe0*"), SV("{:*^4}"), SV("\uffe0")); // FULLWIDTH CENT SIGN
100 check(SV("*\uffe6*"), SV("{:*^4}"), SV("\uffe6")); // FULLWIDTH WON SIGN
101
102 //*** 4-byte code points ***
103 check(SV("*\U00010000*"), SV("{:*^3}"), SV("\U00010000")); // LINEAR B SYLLABLE B008 A
104 check(SV("*\U0010FFFF*"), SV("{:*^3}"), SV("\U0010FFFF")); // Undefined Character
105
106 // 2 column ranges
107 check(SV("*\U0001f300*"), SV("{:*^4}"), SV("\U0001f300")); // CYCLONE
108 check(SV("*\U0001f64f*"), SV("{:*^4}"), SV("\U0001f64f")); // PERSON WITH FOLDED HANDS
109 check(SV("*\U0001f900*"), SV("{:*^4}"), SV("\U0001f900")); // CIRCLED CROSS FORMEE WITH FOUR DOTS
110 check(SV("*\U0001f9ff*"), SV("{:*^4}"), SV("\U0001f9ff")); // NAZAR AMULET
111 check(SV("*\U00020000*"), SV("{:*^4}"), SV("\U00020000")); // <CJK Ideograph Extension B, First>
112 check(SV("*\U0002fffd*"), SV("{:*^4}"), SV("\U0002fffd")); // Undefined Character
113 check(SV("*\U00030000*"), SV("{:*^4}"), SV("\U00030000")); // <CJK Ideograph Extension G, First>
114 check(SV("*\U0003fffd*"), SV("{:*^4}"), SV("\U0003fffd")); // Undefined Character
115}
116
117// One column output is unaffected.
118// Two column output is removed, thus the result is only the fill character.
119template <class CharT>
120static void test_single_code_point_truncate() {
121 //*** 1-byte code points ***
122 check(SV("* *"), SV("{:*^3.1}"), SV(" "));
123 check(SV("*~*"), SV("{:*^3.1}"), SV("~"));
124
125 //*** 2-byte code points ***
126 check(SV("*\u00a1*"), SV("{:*^3.1}"), SV("\u00a1")); // INVERTED EXCLAMATION MARK
127 check(SV("*\u07ff*"), SV("{:*^3.1}"), SV("\u07ff")); // NKO TAMAN SIGN
128
129 //*** 3.1-byte code points ***
130 check(SV("*\u0800*"), SV("{:*^3.1}"), SV("\u0800")); // SAMARITAN LETTER ALAF
131 check(SV("*\ufffd*"), SV("{:*^3.1}"), SV("\ufffd")); // REPLACEMENT CHARACTER
132
133 // 2 column ranges
134 check(SV("***"), SV("{:*^3.1}"), SV("\u1100")); // HANGUL CHOSEONG KIYEOK
135 check(SV("***"), SV("{:*^3.1}"), SV("\u115f")); // HANGUL CHOSEONG FILLER
136
137 check(SV("***"), SV("{:*^3.1}"), SV("\u2329")); // LEFT-POINTING ANGLE BRACKET
138 check(SV("***"), SV("{:*^3.1}"), SV("\u232a")); // RIGHT-POINTING ANGLE BRACKET
139
140 check(SV("***"), SV("{:*^3.1}"), SV("\u2e80")); // CJK RADICAL REPEAT
141 check(SV("***"), SV("{:*^3.1}"), SV("\u303e")); // IDEOGRAPHIC VARIATION INDICATOR
142
143 check(SV("***"), SV("{:*^3.1}"), SV("\u3041")); // U+3041 HIRAGANA LETTER SMALL A
144 check(SV("*\ua4d0*"), SV("{:*^3.1}"), SV("\ua4d0")); // U+A4D0 LISU LETTER BA
145
146 check(SV("***"), SV("{:*^3.1}"), SV("\uac00")); // <Hangul Syllable, First>
147 check(SV("***"), SV("{:*^3.1}"), SV("\ud7a3")); // Hangul Syllable Hih
148
149 check(SV("***"), SV("{:*^3.1}"), SV("\uf900")); // CJK COMPATIBILITY IDEOGRAPH-F900
150 check(SV("***"), SV("{:*^3.1}"), SV("\ufaff")); // U+FB00 LATIN SMALL LIGATURE FF
151
152 check(SV("***"), SV("{:*^3.1}"), SV("\ufe10")); // PRESENTATION FORM FOR VERTICAL COMMA
153 check(SV("***"), SV("{:*^3.1}"), SV("\ufe19")); // PRESENTATION FORM FOR VERTICAL HORIZONTAL ELLIPSIS
154
155 check(SV("***"), SV("{:*^3.1}"), SV("\ufe30")); // PRESENTATION FORM FOR VERTICAL TWO DOT LEADER
156 check(SV("*\ufe70*"), SV("{:*^3.1}"), SV("\ufe70")); // U+FE70 ARABIC FATHATAN ISOLATED FORM
157
158 check(SV("***"), SV("{:*^3.1}"), SV("\uff01")); // U+FF01 FULLWIDTH EXCLAMATION MARK
159 check(SV("***"), SV("{:*^3.1}"), SV("\uff60")); // FULLWIDTH RIGHT WHITE PARENTHESIS
160
161 check(SV("***"), SV("{:*^3.1}"), SV("\uffe0")); // FULLWIDTH CENT SIGN
162 check(SV("***"), SV("{:*^3.1}"), SV("\uffe6")); // FULLWIDTH WON SIGN
163
164 //*** 3.1-byte code points ***
165 check(SV("*\U00010000*"), SV("{:*^3.1}"), SV("\U00010000")); // LINEAR B SYLLABLE B008 A
166 check(SV("*\U0010FFFF*"), SV("{:*^3.1}"), SV("\U0010FFFF")); // Undefined Character
167
168 // 2 column ranges
169 check(SV("***"), SV("{:*^3.1}"), SV("\U0001f300")); // CYCLONE
170 check(SV("***"), SV("{:*^3.1}"), SV("\U0001f64f")); // PERSON WITH FOLDED HANDS
171 check(SV("***"), SV("{:*^3.1}"), SV("\U0001f900")); // CIRCLED CROSS FORMEE WITH FOUR DOTS
172 check(SV("***"), SV("{:*^3.1}"), SV("\U0001f9ff")); // NAZAR AMULET
173 check(SV("***"), SV("{:*^3.1}"), SV("\U00020000")); // <CJK Ideograph Extension B, First>
174 check(SV("***"), SV("{:*^3.1}"), SV("\U0002fffd")); // Undefined Character
175 check(SV("***"), SV("{:*^3.1}"), SV("\U00030000")); // <CJK Ideograph Extension G, First>
176 check(SV("***"), SV("{:*^3.1}"), SV("\U0003fffd")); // Undefined Character
177}
178
179// The examples used in that paper.
180template <class CharT>
181static void test_P1868() {
182 // Fill
183 check(SV("*\u0041*"), SV("{:*^3}"), SV("\u0041")); // { LATIN CAPITAL LETTER A }
184 check(SV("*\u00c1*"), SV("{:*^3}"), SV("\u00c1")); // { LATIN CAPITAL LETTER A WITH ACUTE }
185 check(SV("*\u0041\u0301*"),
186 SV("{:*^3}"),
187 SV("\u0041\u0301")); // { LATIN CAPITAL LETTER A } { COMBINING ACUTE ACCENT }
188 check(SV("*\u0132*"), SV("{:*^3}"), SV("\u0132")); // { LATIN CAPITAL LIGATURE IJ }
189 check(SV("*\u0394*"), SV("{:*^3}"), SV("\u0394")); // { GREEK CAPITAL LETTER DELTA }
190
191 check(SV("*\u0429*"), SV("{:*^3}"), SV("\u0429")); // { CYRILLIC CAPITAL LETTER SHCHA }
192 check(SV("*\u05d0*"), SV("{:*^3}"), SV("\u05d0")); // { HEBREW LETTER ALEF }
193 check(SV("*\u0634*"), SV("{:*^3}"), SV("\u0634")); // { ARABIC LETTER SHEEN }
194 check(SV("*\u3009*"), SV("{:*^4}"), SV("\u3009")); // { RIGHT-POINTING ANGLE BRACKET }
195 check(SV("*\u754c*"), SV("{:*^4}"), SV("\u754c")); // { CJK Unified Ideograph-754C }
196 check(SV("*\U0001f921*"), SV("{:*^4}"), SV("\U0001f921")); // { UNICORN FACE }
197 check(SV("*\U0001f468\u200d\U0001F469\u200d\U0001F467\u200d\U0001F466*"),
198 SV("{:*^4}"),
199 SV("\U0001f468\u200d\U0001F469\u200d\U0001F467\u200d\U0001F466")); // { Family: Man, Woman, Girl, Boy }
200
201 // Truncate to 1 column: 1 column grapheme clusters are kept together.
202 check(SV("*\u0041*"), SV("{:*^3.1}"), SV("\u0041")); // { LATIN CAPITAL LETTER A }
203 check(SV("*\u00c1*"), SV("{:*^3.1}"), SV("\u00c1")); // { LATIN CAPITAL LETTER A WITH ACUTE }
204 check(SV("*\u0041\u0301*"),
205 SV("{:*^3.1}"),
206 SV("\u0041\u0301")); // { LATIN CAPITAL LETTER A } { COMBINING ACUTE ACCENT }
207 check(SV("*\u0132*"), SV("{:*^3.1}"), SV("\u0132")); // { LATIN CAPITAL LIGATURE IJ }
208 check(SV("*\u0394*"), SV("{:*^3.1}"), SV("\u0394")); // { GREEK CAPITAL LETTER DELTA }
209
210 check(SV("*\u0429*"), SV("{:*^3.1}"), SV("\u0429")); // { CYRILLIC CAPITAL LETTER SHCHA }
211 check(SV("*\u05d0*"), SV("{:*^3.1}"), SV("\u05d0")); // { HEBREW LETTER ALEF }
212 check(SV("*\u0634*"), SV("{:*^3.1}"), SV("\u0634")); // { ARABIC LETTER SHEEN }
213 check(SV("***"), SV("{:*^3.1}"), SV("\u3009")); // { RIGHT-POINTING ANGLE BRACKET }
214 check(SV("***"), SV("{:*^3.1}"), SV("\u754c")); // { CJK Unified Ideograph-754C }
215 check(SV("***"), SV("{:*^3.1}"), SV("\U0001f921")); // { UNICORN FACE }
216 check(SV("***"),
217 SV("{:*^3.1}"),
218 SV("\U0001f468\u200d\U0001F469\u200d\U0001F467\u200d\U0001F466")); // { Family: Man, Woman, Girl, Boy }
219
220 // Truncate to 2 column: 2 column grapheme clusters are kept together.
221 check(SV("*\u0041*"), SV("{:*^3.2}"), SV("\u0041")); // { LATIN CAPITAL LETTER A }
222 check(SV("*\u00c1*"), SV("{:*^3.2}"), SV("\u00c1")); // { LATIN CAPITAL LETTER A WITH ACUTE }
223 check(SV("*\u0041\u0301*"),
224 SV("{:*^3.2}"),
225 SV("\u0041\u0301")); // { LATIN CAPITAL LETTER A } { COMBINING ACUTE ACCENT }
226 check(SV("*\u0132*"), SV("{:*^3.2}"), SV("\u0132")); // { LATIN CAPITAL LIGATURE IJ }
227 check(SV("*\u0394*"), SV("{:*^3.2}"), SV("\u0394")); // { GREEK CAPITAL LETTER DELTA }
228
229 check(SV("*\u0429*"), SV("{:*^3.2}"), SV("\u0429")); // { CYRILLIC CAPITAL LETTER SHCHA }
230 check(SV("*\u05d0*"), SV("{:*^3.2}"), SV("\u05d0")); // { HEBREW LETTER ALEF }
231 check(SV("*\u0634*"), SV("{:*^3.2}"), SV("\u0634")); // { ARABIC LETTER SHEEN }
232 check(SV("*\u3009*"), SV("{:*^4.2}"), SV("\u3009")); // { RIGHT-POINTING ANGLE BRACKET }
233 check(SV("*\u754c*"), SV("{:*^4.2}"), SV("\u754c")); // { CJK Unified Ideograph-754C }
234 check(SV("*\U0001f921*"), SV("{:*^4.2}"), SV("\U0001f921")); // { UNICORN FACE }
235 check(SV("*\U0001f468\u200d\U0001F469\u200d\U0001F467\u200d\U0001F466*"),
236 SV("{:*^4.2}"),
237 SV("\U0001f468\u200d\U0001F469\u200d\U0001F467\u200d\U0001F466")); // { Family: Man, Woman, Girl, Boy }
238}
239
240#ifdef _LIBCPP_VERSION
241// Tests the libc++ specific behaviour for malformed UTF-sequences. The
242// Standard doesn't specify how to handle this.
243template <class CharT>
244static void test_malformed_code_point() {
245 if constexpr (sizeof(CharT) == 1) {
246 // Malformed at end.
247 check(SV("*ZZZZ\x8f*"), SV("{:*^7}"), SV("ZZZZ\x8f"));
248 check(SV("*ZZZZ\xcf*"), SV("{:*^7}"), SV("ZZZZ\xcf"));
249 check(SV("*ZZZZ\xef*"), SV("{:*^7}"), SV("ZZZZ\xef"));
250 check(SV("*ZZZZ\xff*"), SV("{:*^7}"), SV("ZZZZ\xff"));
251
252 // Malformed in middle, no continuation
253 check(SV("*ZZZZ\x8fZ*"), SV("{:*^8}"), SV("ZZZZ\x8fZ"));
254 check(SV("*ZZZZ\xcfZ*"), SV("{:*^8}"), SV("ZZZZ\xcfZ"));
255 check(SV("*ZZZZ\xefZ*"), SV("{:*^8}"), SV("ZZZZ\xefZ"));
256 check(SV("*ZZZZ\xffZ*"), SV("{:*^8}"), SV("ZZZZ\xffZ"));
257
258 check(SV("*ZZZZ\x8fZZ*"), SV("{:*^9}"), SV("ZZZZ\x8fZZ"));
259 check(SV("*ZZZZ\xcfZZ*"), SV("{:*^9}"), SV("ZZZZ\xcfZZ"));
260 check(SV("*ZZZZ\xefZZ*"), SV("{:*^9}"), SV("ZZZZ\xefZZ"));
261 check(SV("*ZZZZ\xffZZ*"), SV("{:*^9}"), SV("ZZZZ\xffZZ"));
262
263 check(SV("*ZZZZ\x8fZZZ*"), SV("{:*^10}"), SV("ZZZZ\x8fZZZ"));
264 check(SV("*ZZZZ\xcfZZZ*"), SV("{:*^10}"), SV("ZZZZ\xcfZZZ"));
265 check(SV("*ZZZZ\xefZZZ*"), SV("{:*^10}"), SV("ZZZZ\xefZZZ"));
266 check(SV("*ZZZZ\xffZZZ*"), SV("{:*^10}"), SV("ZZZZ\xffZZZ"));
267
268 check(SV("*ZZZZ\x8fZZZZ*"), SV("{:*^11}"), SV("ZZZZ\x8fZZZZ"));
269 check(SV("*ZZZZ\xcfZZZZ*"), SV("{:*^11}"), SV("ZZZZ\xcfZZZZ"));
270 check(SV("*ZZZZ\xefZZZZ*"), SV("{:*^11}"), SV("ZZZZ\xefZZZZ"));
271 check(SV("*ZZZZ\xffZZZZ*"), SV("{:*^11}"), SV("ZZZZ\xffZZZZ"));
272
273 // Invalid continuations
274 check(SV("\xc2\x00"), SV("{}"), SV("\xc2\x00"));
275 check(SV("\xc2\x40"), SV("{}"), SV("\xc2\x40"));
276 check(SV("\xc2\xc0"), SV("{}"), SV("\xc2\xc0"));
277
278 check(SV("\xe0\x00\x80"), SV("{}"), SV("\xe0\x00\x80"));
279 check(SV("\xe0\x40\x80"), SV("{}"), SV("\xe0\x40\x80"));
280 check(SV("\xe0\xc0\x80"), SV("{}"), SV("\xe0\xc0\x80"));
281
282 check(SV("\xe0\x80\x00"), SV("{}"), SV("\xe0\x80\x00"));
283 check(SV("\xe0\x80\x40"), SV("{}"), SV("\xe0\x80\x40"));
284 check(SV("\xe0\x80\xc0"), SV("{}"), SV("\xe0\x80\xc0"));
285
286 check(SV("\xf0\x80\x80\x00"), SV("{}"), SV("\xf0\x80\x80\x00"));
287 check(SV("\xf0\x80\x80\x40"), SV("{}"), SV("\xf0\x80\x80\x40"));
288 check(SV("\xf0\x80\x80\xc0"), SV("{}"), SV("\xf0\x80\x80\xc0"));
289
290 check(SV("\xf0\x80\x00\x80"), SV("{}"), SV("\xf0\x80\x00\x80"));
291 check(SV("\xf0\x80\x40\x80"), SV("{}"), SV("\xf0\x80\x40\x80"));
292 check(SV("\xf0\x80\xc0\x80"), SV("{}"), SV("\xf0\x80\xc0\x80"));
293
294 check(SV("\xf0\x00\x80\x80"), SV("{}"), SV("\xf0\x00\x80\x80"));
295 check(SV("\xf0\x40\x80\x80"), SV("{}"), SV("\xf0\x40\x80\x80"));
296 check(SV("\xf0\xc0\x80\x80"), SV("{}"), SV("\xf0\xc0\x80\x80"));
297
298 // Premature end.
299 check(SV("*ZZZZ\xef\xf5*"), SV("{:*^8}"), SV("ZZZZ\xef\xf5"));
300 check(SV("*ZZZZ\xef\xf5ZZZZ*"), SV("{:*^12}"), SV("ZZZZ\xef\xf5ZZZZ"));
301 check(SV("*ZZZZ\xff\xf5\xf5*"), SV("{:*^9}"), SV("ZZZZ\xff\xf5\xf5"));
302 check(SV("*ZZZZ\xff\xf5\xf5ZZZZ*"), SV("{:*^13}"), SV("ZZZZ\xff\xf5\xf5ZZZZ"));
303
304 } else if constexpr (sizeof(CharT) == 2) {
305 // TODO FMT Add these tests.
306 }
307 // UTF-32 doesn't combine characters, thus no corruption tests.
308}
309#endif
310
311template <class CharT>
312static void test() {
313 test_single_code_point_fill<CharT>();
314 test_single_code_point_truncate<CharT>();
315 test_P1868<CharT>();
316
317#ifdef _LIBCPP_VERSION
318 test_malformed_code_point<CharT>();
319#endif
320}
321
322int main(int, char**) {
323 test<char>();
324
325#ifndef TEST_HAS_NO_WIDE_CHARACTERS
326 test<wchar_t>();
327#endif
328
329 return 0;
330}
331

source code of libcxx/test/std/utilities/format/format.functions/unicode.pass.cpp