xref: /llvm-project/libcxx/test/std/utilities/format/format.functions/escaped_output.unicode.pass.cpp (revision f5832bab6f5024cabe32a9f668b7f44e6b7cfef5)
1 //===----------------------------------------------------------------------===//
2 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
3 // See https://llvm.org/LICENSE.txt for license information.
4 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
5 //
6 //===----------------------------------------------------------------------===//
7 
8 // UNSUPPORTED: c++03, c++11, c++14, c++17, c++20
9 // UNSUPPORTED: GCC-ALWAYS_INLINE-FIXME
10 
11 // This version runs the test when the platform has Unicode support.
12 // UNSUPPORTED: libcpp-has-no-unicode
13 
14 // TODO FMT Investigate Windows issues.
15 // UNSUPPORTED: msvc, target={{.+}}-windows-gnu
16 
17 // TODO FMT This test should not require std::to_chars(floating-point)
18 // XFAIL: availability-fp_to_chars-missing
19 
20 // <format>
21 
22 // This test the debug string type for the formatter specializations for char
23 // and string types. This tests Unicode strings.
24 
25 #include <format>
26 
27 #include <cassert>
28 #include <concepts>
29 #include <iterator>
30 #include <list>
31 #include <vector>
32 
33 #include "test_macros.h"
34 #include "make_string.h"
35 #include "test_format_string.h"
36 #include "assert_macros.h"
37 #include "concat_macros.h"
38 
39 #ifndef TEST_HAS_NO_LOCALIZATION
40 #  include <iostream>
41 #endif
42 
43 #define SV(S) MAKE_STRING_VIEW(CharT, S)
44 
45 auto test_format = []<class CharT, class... Args>(
46                        std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
47   {
48     std::basic_string<CharT> out = std::format(fmt, std::forward<Args>(args)...);
49     TEST_REQUIRE(out == expected,
50                  TEST_WRITE_CONCATENATED(
51                      "\nFormat string   ", fmt.get(), "\nExpected output ", expected, "\nActual output   ", out, '\n'));
52   }
53 #ifndef TEST_HAS_NO_LOCALIZATION
54   {
55     std::basic_string<CharT> out = std::format(std::locale(), fmt, std::forward<Args>(args)...);
56     assert(out == expected);
57   }
58 #endif // TEST_HAS_NO_LOCALIZATION
59 };
60 
61 auto test_format_to =
62     []<class CharT, class... Args>(
63         std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
64       {
65         std::basic_string<CharT> out(expected.size(), CharT(' '));
66         auto it = std::format_to(out.begin(), fmt, std::forward<Args>(args)...);
67         assert(it == out.end());
68         assert(out == expected);
69       }
70 #ifndef TEST_HAS_NO_LOCALIZATION
71       {
72         std::basic_string<CharT> out(expected.size(), CharT(' '));
73         auto it = std::format_to(out.begin(), std::locale(), fmt, std::forward<Args>(args)...);
74         assert(it == out.end());
75         assert(out == expected);
76       }
77 #endif // TEST_HAS_NO_LOCALIZATION
78       {
79         std::list<CharT> out;
80         std::format_to(std::back_inserter(out), fmt, std::forward<Args>(args)...);
81         assert(std::equal(out.begin(), out.end(), expected.begin(), expected.end()));
82       }
83       {
84         std::vector<CharT> out;
85         std::format_to(std::back_inserter(out), fmt, std::forward<Args>(args)...);
86         assert(std::equal(out.begin(), out.end(), expected.begin(), expected.end()));
87       }
88       {
89         assert(expected.size() < 4096 && "Update the size of the buffer.");
90         CharT out[4096];
91         CharT* it = std::format_to(out, fmt, std::forward<Args>(args)...);
92         assert(std::distance(out, it) == int(expected.size()));
93         // Convert to std::string since output contains '\0' for boolean tests.
94         assert(std::basic_string<CharT>(out, it) == expected);
95       }
96     };
97 
98 auto test_formatted_size =
99     []<class CharT, class... Args>(
100         std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
101       {
102         std::size_t size = std::formatted_size(fmt, std::forward<Args>(args)...);
103         assert(size == expected.size());
104       }
105 #ifndef TEST_HAS_NO_LOCALIZATION
106       {
107         std::size_t size = std::formatted_size(std::locale(), fmt, std::forward<Args>(args)...);
108         assert(size == expected.size());
109       }
110 #endif // TEST_HAS_NO_LOCALIZATION
111     };
112 
113 auto test_format_to_n =
114     []<class CharT, class... Args>(
115         std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
116       {
117         std::size_t n = expected.size();
118         std::basic_string<CharT> out(n, CharT(' '));
119         std::format_to_n_result result = std::format_to_n(out.begin(), n, fmt, std::forward<Args>(args)...);
120         assert(result.size == static_cast<std::ptrdiff_t>(expected.size()));
121         assert(result.out == out.end());
122         assert(out == expected);
123       }
124 #ifndef TEST_HAS_NO_LOCALIZATION
125       {
126         std::size_t n = expected.size();
127         std::basic_string<CharT> out(n, CharT(' '));
128         std::format_to_n_result result =
129             std::format_to_n(out.begin(), n, std::locale(), fmt, std::forward<Args>(args)...);
130         assert(result.size == static_cast<std::ptrdiff_t>(expected.size()));
131         assert(result.out == out.end());
132         assert(out == expected);
133       }
134 #endif // TEST_HAS_NO_LOCALIZATION
135       {
136         std::ptrdiff_t n = 0;
137         std::basic_string<CharT> out;
138         std::format_to_n_result result = std::format_to_n(out.begin(), n, fmt, std::forward<Args>(args)...);
139         assert(result.size == static_cast<std::ptrdiff_t>(expected.size()));
140         assert(result.out == out.end());
141         assert(out.empty());
142       }
143       {
144         std::ptrdiff_t n = expected.size() / 2;
145         std::basic_string<CharT> out(n, CharT(' '));
146         std::format_to_n_result result = std::format_to_n(out.begin(), n, fmt, std::forward<Args>(args)...);
147         assert(result.size == static_cast<std::ptrdiff_t>(expected.size()));
148         assert(result.out == out.end());
149         assert(out == expected.substr(0, n));
150       }
151     };
152 
153 template <class CharT>
154 void test_char() {
155   // *** P2286 examples ***
156   test_format(SV("['\\'', '\"']"), SV("[{:?}, {:?}]"), CharT('\''), CharT('"'));
157 
158   // *** Special cases ***
159   test_format(SV("'\\t'"), SV("{:?}"), CharT('\t'));
160   test_format(SV("'\\n'"), SV("{:?}"), CharT('\n'));
161   test_format(SV("'\\r'"), SV("{:?}"), CharT('\r'));
162   test_format(SV("'\\\\'"), SV("{:?}"), CharT('\\'));
163 
164   test_format(SV("'\\\''"), SV("{:?}"), CharT('\''));
165   test_format(SV("'\"'"), SV("{:?}"), CharT('"')); // only special for string
166 
167   test_format(SV("' '"), SV("{:?}"), CharT(' '));
168 
169   // *** Printable ***
170   test_format(SV("'a'"), SV("{:?}"), CharT('a'));
171   test_format(SV("'b'"), SV("{:?}"), CharT('b'));
172   test_format(SV("'c'"), SV("{:?}"), CharT('c'));
173 
174   // *** Non-printable ***
175 
176   // Control
177   test_format(SV("'\\u{0}'"), SV("{:?}"), CharT('\0'));
178   test_format(SV("'\\u{1f}'"), SV("{:?}"), CharT('\x1f'));
179 
180   // Ill-formed
181   if constexpr (sizeof(CharT) == 1)
182     test_format(SV("'\\x{80}'"), SV("{:?}"), CharT('\x80'));
183 
184 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
185   if constexpr (sizeof(CharT) > 1) {
186     using V = std::basic_string_view<CharT>;
187 
188     // Unicode fitting in a 16-bit wchar_t
189 
190     // *** Non-printable ***
191 
192     // Space_Separator
193     test_format(V{L"'\\u{a0}'"}, L"{:?}", L'\xa0');     // NO-BREAK SPACE
194     test_format(V{L"'\\u{3000}'"}, L"{:?}", L'\x3000'); // IDEOGRAPHIC SPACE
195 
196     // Line_Separator
197     test_format(V{L"'\\u{2028}'"}, L"{:?}", L'\x2028'); // LINE SEPARATOR
198 
199     // Paragraph_Separator
200     test_format(V{L"'\\u{2029}'"}, L"{:?}", L'\x2029'); // PARAGRAPH SEPARATOR
201 
202     // Format
203     test_format(V{L"'\\u{ad}'"}, L"{:?}", L'\xad');     // SOFT HYPHEN
204     test_format(V{L"'\\u{600}'"}, L"{:?}", L'\x600');   // ARABIC NUMBER SIGN
205     test_format(V{L"'\\u{feff}'"}, L"{:?}", L'\xfeff'); // ZERO WIDTH NO-BREAK SPACE
206 
207     // Incomplete surrogate pair in UTF-16
208     test_format(V{L"'\\x{d800}'"}, L"{:?}", L'\xd800'); // <surrogate-D800>
209     test_format(V{L"'\\x{dfff}'"}, L"{:?}", L'\xdfff'); // <surrogate-DFFF>
210 
211     // Private_Use
212     test_format(V{L"'\\u{e000}'"}, L"{:?}", L'\xe000'); // <private-use-E000>
213     test_format(V{L"'\\u{f8ff}'"}, L"{:?}", L'\xf8ff'); // <private-use-F8FF>
214 
215     // Unassigned
216     test_format(V{L"'\\u{378}'"}, L"{:?}", L'\x378');   // <reserved-0378>
217     test_format(V{L"'\\u{1774}'"}, L"{:?}", L'\x1774'); // <reserved-1774>
218     test_format(V{L"'\\u{ffff}'"}, L"{:?}", L'\xffff'); // <noncharacter-FFFF>
219 
220     // Grapheme Extended
221     test_format(V{L"'\\u{300}'"}, L"{:?}", L'\x300');   // COMBINING GRAVE ACCENT
222     test_format(V{L"'\\u{fe20}'"}, L"{:?}", L'\xfe20'); // VARIATION SELECTOR-1
223   }
224 #  ifndef TEST_SHORT_WCHAR
225   if constexpr (sizeof(CharT) > 2) {
226     static_assert(sizeof(CharT) == 4, "add support for unexpected size");
227     // Unicode fitting in a 32-bit wchar_t
228 
229     constexpr wchar_t x  = 0x1ffff;
230     constexpr std::uint32_t y = 0x1ffff;
231     static_assert(x == y);
232 
233     using V = std::basic_string_view<CharT>;
234 
235     // *** Non-printable ***
236     // Format
237     test_format(V{L"'\\u{110bd}'"}, L"{:?}", L'\x110bd'); // KAITHI NUMBER SIGN
238     test_format(V{L"'\\u{e007f}'"}, L"{:?}", L'\xe007f'); // CANCEL TAG
239 
240     // Private_Use
241     test_format(V{L"'\\u{f0000}'"}, L"{:?}", L'\xf0000'); // <private-use-F0000>
242     test_format(V{L"'\\u{ffffd}'"}, L"{:?}", L'\xffffd'); // <private-use-FFFFD>
243 
244     test_format(V{L"'\\u{100000}'"}, L"{:?}", L'\x100000'); // <private-use-100000>
245     test_format(V{L"'\\u{10fffd}'"}, L"{:?}", L'\x10fffd'); // <private-use-10FFFD>
246 
247     // Unassigned
248     test_format(V{L"'\\u{1000c}'"}, L"{:?}", L'\x1000c');   // <reserved-1000c>
249     test_format(V{L"'\\u{fffff}'"}, L"{:?}", L'\xfffff');   // <noncharacter-FFFFF>
250     test_format(V{L"'\\u{10fffe}'"}, L"{:?}", L'\x10fffe'); // <noncharacter-10FFFE>
251 
252     // Grapheme Extended
253     test_format(V{L"'\\u{101fd}'"}, L"{:?}", L'\x101fd'); // COMBINING OLD PERMIC LETTER AN
254     test_format(V{L"'\\u{e0100}'"}, L"{:?}", L'\xe0100'); // VARIATION SELECTOR-17
255 
256     // Ill-formed
257     test_format(V{L"'\\x{110000}'"}, L"{:?}", L'\x110000');
258     test_format(V{L"'\\x{ffffffff}'"}, L"{:?}", L'\xffffffff');
259   }
260 #  endif // TEST_SHORT_WCHAR
261 #endif   // TEST_HAS_NO_WIDE_CHARACTERS
262 }
263 
264 template <class CharT>
265 void test_string() {
266   // *** P2286 examples ***
267   test_format(SV("[h\tllo]"), SV("[{}]"), SV("h\tllo"));
268   test_format(SV(R"(["h\tllo"])"), SV("[{:?}]"), SV("h\tllo"));
269   test_format(SV(R"(["Спасибо, Виктор ♥!"])"), SV("[{:?}]"), SV("Спасибо, Виктор ♥!"));
270 
271   test_format(SV(R"(["\u{0} \n \t \u{2} \u{1b}"])"), SV("[{:?}]"), SV("\0 \n \t \x02 \x1b"));
272 
273   if constexpr (sizeof(CharT) == 1) {
274     // Ill-formend UTF-8
275     test_format(SV(R"(["\x{c3}"])"), SV("[{:?}]"), "\xc3");
276     test_format(SV(R"(["\x{c3}("])"), SV("[{:?}]"), "\xc3\x28");
277 
278     /* U+0000..U+0007F 1 code unit range, encoded in 2 code units. */
279     test_format(SV(R"(["\x{c0}\x{80}"])"), SV("[{:?}]"), "\xc0\x80"); // U+0000
280     test_format(SV(R"(["\x{c1}\x{bf}"])"), SV("[{:?}]"), "\xc1\xbf"); // U+007F
281     test_format(SV(R"(["\u{80}"])"), SV("[{:?}]"), "\xc2\x80");       // U+0080 first valid (General_Category=Control)
282 
283     /* U+0000..U+07FFF 1 and 2 code unit range, encoded in 3 code units. */
284     test_format(SV(R"(["\x{e0}\x{80}\x{80}"])"), SV("[{:?}]"), "\xe0\x80\x80"); // U+0000
285     test_format(SV(R"(["\x{e0}\x{81}\x{bf}"])"), SV("[{:?}]"), "\xe0\x81\xbf"); // U+007F
286     test_format(SV(R"(["\x{e0}\x{82}\x{80}"])"), SV("[{:?}]"), "\xe0\x82\x80"); // U+0080
287     test_format(SV(R"(["\x{e0}\x{9f}\x{bf}"])"), SV("[{:?}]"), "\xe0\x9f\xbf"); // U+07FF
288     test_format(SV("[\"\u0800\"]"), SV("[{:?}]"), "\xe0\xa0\x80");              // U+0800 first valid
289 
290 #if 0
291 	// This code point is in the Hangul Jamo Extended-B block and at the time of writing
292 	// it's unassigned. When it comes defined, this branch might become true.
293     test_format(SV("[\"\ud7ff\"]"), SV("[{:?}]"), "\xed\x9f\xbf");              // U+D7FF last valid
294 #else
295     /* U+D800..D+DFFFF surrogate range */
296     test_format(SV(R"(["\u{d7ff}"])"), SV("[{:?}]"), "\xed\x9f\xbf");           // U+D7FF last valid
297 #endif
298     test_format(SV(R"(["\x{ed}\x{a0}\x{80}"])"), SV("[{:?}]"), "\xed\xa0\x80"); // U+D800
299     test_format(SV(R"(["\x{ed}\x{af}\x{bf}"])"), SV("[{:?}]"), "\xed\xaf\xbf"); // U+DBFF
300     test_format(SV(R"(["\x{ed}\x{bf}\x{80}"])"), SV("[{:?}]"), "\xed\xbf\x80"); // U+DC00
301     test_format(SV(R"(["\x{ed}\x{bf}\x{bf}"])"), SV("[{:?}]"), "\xed\xbf\xbf"); // U+DFFF
302     test_format(SV(R"(["\u{e000}"])"), SV("[{:?}]"), "\xee\x80\x80");           // U+E000 first valid
303                                                                                 // (in the Private Use Area block)
304 
305     /* U+0000..U+FFFF 1, 2, and 3 code unit range */
306     test_format(SV(R"(["\x{f0}\x{80}\x{80}\x{80}"])"), SV("[{:?}]"), "\xf0\x80\x80\x80"); // U+0000
307     test_format(SV(R"(["\x{f0}\x{80}\x{81}\x{bf}"])"), SV("[{:?}]"), "\xf0\x80\x81\xbf"); // U+007F
308     test_format(SV(R"(["\x{f0}\x{80}\x{82}\x{80}"])"), SV("[{:?}]"), "\xf0\x80\x82\x80"); // U+0080
309     test_format(SV(R"(["\x{f0}\x{80}\x{9f}\x{bf}"])"), SV("[{:?}]"), "\xf0\x80\x9f\xbf"); // U+07FF
310     test_format(SV(R"(["\x{f0}\x{80}\x{a0}\x{80}"])"), SV("[{:?}]"), "\xf0\x80\xa0\x80"); // U+0800
311     test_format(SV(R"(["\x{f0}\x{8f}\x{bf}\x{bf}"])"), SV("[{:?}]"), "\xf0\x8f\xbf\xbf"); // U+FFFF
312     test_format(SV("[\"\U00010000\"]"), SV("[{:?}]"), "\xf0\x90\x80\x80");                // U+10000 first valid
313 
314     /* U+10FFFF..U+1FFFFF invalid range */
315     test_format(SV(R"(["\u{10ffff}"])"), SV("[{:?}]"), "\xf4\x8f\xbf\xbf"); // U+10FFFF last valid
316                                                                             // (in Supplementary Private Use Area-B)
317     test_format(SV(R"(["\x{f4}\x{90}\x{80}\x{80}"])"), SV("[{:?}]"), "\xf4\x90\x80\x80"); // U+110000
318     test_format(SV(R"(["\x{f4}\x{bf}\x{bf}\x{bf}"])"), SV("[{:?}]"), "\xf4\xbf\xbf\xbf"); // U+11FFFF
319   } else {
320     // Valid UTF-16 and UTF-32
321     test_format(SV("[\"\u00c3\"]"), SV("[{:?}]"), L"\xc3"); // LATIN CAPITAL LETTER A WITH TILDE
322     test_format(SV("[\"\u00c3(\"]"), SV("[{:?}]"), L"\xc3\x28");
323   }
324 
325   test_format(SV(R"(["����\u{200d}♂\u{fe0f}"])"), SV("[{:?}]"), SV("����‍♂️"));
326 
327   // *** Special cases ***
328   test_format(SV(R"("\t\n\r\\'\" ")"), SV("{:?}"), SV("\t\n\r\\'\" "));
329 
330   // *** Printable ***
331   test_format(SV(R"("abcdefg")"), SV("{:?}"), SV("abcdefg"));
332 
333   // *** Non-printable ***
334 
335   // Control
336   test_format(SV(R"("\u{0}\u{1f}")"), SV("{:?}"), SV("\0\x1f"));
337 
338   // Ill-formed
339   if constexpr (sizeof(CharT) == 1)
340     test_format(SV(R"("\x{80}")"), SV("{:?}"), SV("\x80"));
341 
342 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
343   if constexpr (sizeof(CharT) > 1) {
344     using V = std::basic_string_view<CharT>;
345 
346     // Unicode fitting in a 16-bit wchar_t
347 
348     // *** Non-printable ***
349 
350     // Space_Separator
351     test_format(V{LR"("\u{a0}\u{3000}")"}, L"{:?}", L"\xa0\x3000");
352 
353     // Line_Separator
354     test_format(V{LR"("\u{2028}")"}, L"{:?}", L"\x2028"); // LINE SEPARATOR
355 
356     // Paragraph_Separator
357     test_format(V{LR"("\u{2029}")"}, L"{:?}", L"\x2029"); // PARAGRAPH SEPARATOR
358 
359     // Format
360     test_format(V{LR"("\u{ad}\u{600}\u{feff}")"}, L"{:?}", L"\xad\x600\xfeff");
361 
362     // Incomplete surrogate pair in UTF-16
363     test_format(V{LR"("\x{d800}")"}, L"{:?}", L"\xd800");
364 
365     // Private_Use
366     test_format(V{LR"("\u{e000}\u{f8ff}")"}, L"{:?}", L"\xe000\xf8ff");
367 
368     // Unassigned
369     test_format(V{LR"("\u{378}\u{1774}\u{ffff}")"}, L"{:?}", L"\x378\x1774\xffff");
370 
371     // Grapheme Extended
372     test_format(V{LR"("\u{300}\u{fe20}")"}, L"{:?}", L"\x300\xfe20");
373   }
374 #  ifndef TEST_SHORT_WCHAR
375   if constexpr (sizeof(CharT) > 2) {
376     static_assert(sizeof(CharT) == 4, "add support for unexpected size");
377     // Unicode fitting in a 32-bit wchar_t
378 
379     constexpr wchar_t x  = 0x1ffff;
380     constexpr std::uint32_t y = 0x1ffff;
381     static_assert(x == y);
382 
383     using V = std::basic_string_view<CharT>;
384 
385     // *** Non-printable ***
386     // Format
387     test_format(V{LR"("\u{110bd}\u{e007f}")"}, L"{:?}", L"\x110bd\xe007f");
388 
389     // Private_Use
390     test_format(V{LR"("\u{f0000}\u{ffffd}\u{100000}\u{10fffd}")"}, L"{:?}", L"\xf0000\xffffd\x100000\x10fffd");
391 
392     // Unassigned
393     test_format(V{LR"("\u{1000c}\u{fffff}\u{10fffe}")"}, L"{:?}", L"\x1000c\xfffff\x10fffe");
394 
395     // Grapheme Extended
396     test_format(V{LR"("\u{101fd}\u{e0100}")"}, L"{:?}", L"\x101fd\xe0100");
397 
398     // Ill-formed
399     test_format(V{LR"("\x{110000}\x{ffffffff}")"}, L"{:?}", L"\x110000\xffffffff");
400   }
401 #  endif // TEST_SHORT_WCHAR
402 #endif   // TEST_HAS_NO_WIDE_CHARACTERS
403 }
404 
405 template <class CharT, class TestFunction>
406 void test_format_functions(TestFunction check) {
407   // *** align-fill & width ***
408   check(SV(R"(***"hellö")"), SV("{:*>10?}"), SV("hellö")); // ö is LATIN SMALL LETTER O WITH DIAERESIS
409   check(SV(R"(*"hellö"**)"), SV("{:*^10?}"), SV("hellö"));
410   check(SV(R"("hellö"***)"), SV("{:*<10?}"), SV("hellö"));
411 
412   check(SV(R"("hello\u{308}")"), SV("{:*>10?}"), SV("hello\u0308"));
413   check(SV(R"(***"hello\u{308}")"), SV("{:*>17?}"), SV("hello\u0308"));
414   check(SV(R"(*"hello\u{308}"**)"), SV("{:*^17?}"), SV("hello\u0308"));
415   check(SV(R"("hello\u{308}"***)"), SV("{:*<17?}"), SV("hello\u0308"));
416 
417   check(SV(R"("hello ����\u{200d}♂\u{fe0f}")"), SV("{:*>10?}"), SV("hello ����‍♂️"));
418   check(SV(R"(***"hello ����\u{200d}♂\u{fe0f}")"), SV("{:*>30?}"), SV("hello ����‍♂️"));
419   check(SV(R"(*"hello ����\u{200d}♂\u{fe0f}"**)"), SV("{:*^30?}"), SV("hello ����‍♂️"));
420   check(SV(R"("hello ����\u{200d}♂\u{fe0f}"***)"), SV("{:*<30?}"), SV("hello ����‍♂️"));
421 
422   // *** width ***
423   check(SV(R"("hellö"   )"), SV("{:10?}"), SV("hellö"));
424   check(SV(R"("hello\u{308}"   )"), SV("{:17?}"), SV("hello\u0308"));
425   check(SV(R"("hello ����\u{200d}♂\u{fe0f}"   )"), SV("{:30?}"), SV("hello ����‍♂️"));
426 
427   // *** precision ***
428   check(SV(R"("hell)"), SV("{:.5?}"), SV("hellö"));
429   check(SV(R"("hellö)"), SV("{:.6?}"), SV("hellö"));
430   check(SV(R"("hellö")"), SV("{:.7?}"), SV("hellö"));
431 
432   check(SV(R"("hello )"), SV("{:.7?}"), SV("hello ����‍♂️"));
433   check(SV(R"("hello )"), SV("{:.8?}"), SV("hello ����‍♂️")); // shrug is two columns
434   check(SV(R"("hello ����)"), SV("{:.9?}"), SV("hello ����‍♂️"));
435   check(SV(R"("hello ����\)"), SV("{:.10?}"), SV("hello ����‍♂️"));
436   check(SV(R"("hello ����\u{200d})"), SV("{:.17?}"), SV("hello ����‍♂️"));
437   check(SV(R"("hello ����\u{200d}♂)"), SV("{:.18?}"), SV("hello ����‍♂️"));
438   check(SV(R"("hello ����\u{200d}♂\)"), SV("{:.19?}"), SV("hello ����‍♂️"));
439   check(SV(R"("hello ����\u{200d}♂\u{fe0f}")"), SV("{:.28?}"), SV("hello ����‍♂️"));
440 
441   // *** width & precision ***
442   check(SV(R"("hell#########################)"), SV("{:#<30.5?}"), SV("hellö"));
443   check(SV(R"("hellö########################)"), SV("{:#<30.6?}"), SV("hellö"));
444   check(SV(R"("hellö"#######################)"), SV("{:#<30.7?}"), SV("hellö"));
445 
446   check(SV(R"("hello #######################)"), SV("{:#<30.7?}"), SV("hello ����‍♂️"));
447   check(SV(R"("hello #######################)"), SV("{:#<30.8?}"), SV("hello ����‍♂️"));
448   check(SV(R"("hello ����#####################)"), SV("{:#<30.9?}"), SV("hello ����‍♂️"));
449   check(SV(R"("hello ����\####################)"), SV("{:#<30.10?}"), SV("hello ����‍♂️"));
450   check(SV(R"("hello ����\u{200d}#############)"), SV("{:#<30.17?}"), SV("hello ����‍♂️"));
451   check(SV(R"("hello ����\u{200d}♂############)"), SV("{:#<30.18?}"), SV("hello ����‍♂️"));
452   check(SV(R"("hello ����\u{200d}♂\###########)"), SV("{:#<30.19?}"), SV("hello ����‍♂️"));
453   check(SV(R"("hello ����\u{200d}♂\u{fe0f}"###)"), SV("{:#<30.28?}"), SV("hello ����‍♂️"));
454 }
455 
456 template <class CharT>
457 void test() {
458   test_char<CharT>();
459   test_string<CharT>();
460 
461   test_format_functions<CharT>(test_format);
462   test_format_functions<CharT>(test_format_to);
463   test_format_functions<CharT>(test_formatted_size);
464   test_format_functions<CharT>(test_format_to_n);
465 }
466 
467 static void test_ill_formed_utf8() {
468   using namespace std::literals;
469 
470   // Too few code units
471   test_format(R"("\x{df}")"sv, "{:?}", "\xdf");
472   test_format(R"("\x{ef}")"sv, "{:?}", "\xef");
473   test_format(R"("\x{ef}\x{bf}")"sv, "{:?}", "\xef\xbf");
474   test_format(R"("\x{f7}")"sv, "{:?}", "\xf7");
475   test_format(R"("\x{f7}\x{bf}")"sv, "{:?}", "\xf7\xbf");
476   test_format(R"("\x{f7}\x{bf}\x{bf}")"sv, "{:?}", "\xf7\xbf\xbf");
477 
478   // Invalid continuation byte
479   test_format(R"("\x{df}a")"sv,
480               "{:?}",
481               "\xdf"
482               "a");
483   test_format(R"("\x{ef}a")"sv,
484               "{:?}",
485               "\xef"
486               "a");
487   test_format(R"("\x{ef}\x{bf}a")"sv,
488               "{:?}",
489               "\xef\xbf"
490               "a");
491   test_format(R"("\x{f7}a")"sv,
492               "{:?}",
493               "\xf7"
494               "a");
495   test_format(R"("\x{f7}\x{bf}a")"sv,
496               "{:?}",
497               "\xf7\xbf"
498               "a");
499   test_format(R"("\x{f7}\x{bf}\x{bf}a")"sv,
500               "{:?}",
501               "\xf7\xbf\xbf"
502               "a");
503 
504   test_format(R"("a\x{f1}\x{80}\x{80}\x{e1}\x{80}\x{c2}b")"sv,
505               "{:?}",
506               "a"
507               "\xf1\x80\x80\xe1\x80\xc2"
508               "b");
509 
510   // Code unit out of range
511   test_format(R"("\u{10ffff}")"sv, "{:?}", "\xf4\x8f\xbf\xbf");               // last valid code point
512   test_format(R"("\x{f4}\x{90}\x{80}\x{80}")"sv, "{:?}", "\xf4\x90\x80\x80"); // first invalid code point
513   test_format(R"("\x{f5}\x{b1}\x{b2}\x{b3}")"sv, "{:?}", "\xf5\xb1\xb2\xb3");
514   test_format(R"("\x{f7}\x{bf}\x{bf}\x{bf}")"sv, "{:?}", "\xf7\xbf\xbf\xbf"); // largest encoded code point
515 }
516 
517 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
518 #  ifdef _LIBCPP_SHORT_WCHAR
519 static void test_ill_formed_utf16() {
520   using namespace std::literals;
521 
522   // Too few code units
523   test_format(LR"("\x{d800}")"sv, L"{:?}", L"\xd800");
524   test_format(LR"("\x{dbff}")"sv, L"{:?}", L"\xdbff");
525 
526   // Start with low surrogate pair
527   test_format(LR"("\x{dc00}a")"sv,
528               L"{:?}",
529               L"\xdc00"
530               "a");
531   test_format(LR"("\x{dfff}a")"sv,
532               L"{:?}",
533               L"\xdfff"
534               "a");
535 
536   // Only high surrogate pair
537   test_format(LR"("\x{d800}a")"sv,
538               L"{:?}",
539               L"\xd800"
540               "a");
541   test_format(LR"("\x{dbff}a")"sv,
542               L"{:?}",
543               L"\xdbff"
544               "a");
545 }
546 #  else // _LIBCPP_SHORT_WCHAR
547 static void test_ill_formed_utf32() {
548   using namespace std::literals;
549 
550   test_format(LR"("\u{10ffff}")"sv, L"{:?}", L"\x10ffff");     // last valid code point
551   test_format(LR"("\x{110000}")"sv, L"{:?}", L"\x110000");     // first invalid code point
552   test_format(LR"("\x{ffffffff}")"sv, L"{:?}", L"\xffffffff"); // largest encoded code point
553 }
554 
555 #  endif // _LIBCPP_SHORT_WCHAR
556 #endif   // TEST_HAS_NO_WIDE_CHARACTERS
557 
558 int main(int, char**) {
559   test<char>();
560 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
561   test<wchar_t>();
562 #endif
563 
564   test_ill_formed_utf8();
565 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
566 #  ifdef _LIBCPP_SHORT_WCHAR
567   test_ill_formed_utf16();
568 #  else  // _LIBCPP_SHORT_WCHAR
569   test_ill_formed_utf32();
570 #  endif // _LIBCPP_SHORT_WCHAR
571 #endif   // TEST_HAS_NO_WIDE_CHARACTERS
572 
573   return 0;
574 }
575