xref: /llvm-project/libcxx/test/std/utilities/format/format.functions/escaped_output.unicode.pass.cpp (revision c866855b42eb3e8aa7578aadb26e4431d1d71efd)
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: libcpp-has-no-incomplete-format
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 and AIX issues.
15 // UNSUPPORTED: msvc, target={{.+}}-windows-gnu
16 // UNSUPPORTED: LIBCXX-AIX-FIXME
17 
18 // <format>
19 
20 // This test the debug string type for the formatter specializations for char
21 // and string types. This tests Unicode strings.
22 
23 #include <format>
24 
25 #include <cassert>
26 #include <concepts>
27 #include <iterator>
28 #include <list>
29 #include <vector>
30 
31 #include "test_macros.h"
32 #include "make_string.h"
33 #include "test_format_string.h"
34 #include "assert_macros.h"
35 #include "concat_macros.h"
36 
37 #ifndef TEST_HAS_NO_LOCALIZATION
38 #  include <iostream>
39 #endif
40 
41 #define SV(S) MAKE_STRING_VIEW(CharT, S)
42 
43 auto test_format = []<class CharT, class... Args>(
44                        std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
45   {
46     std::basic_string<CharT> out = std::format(fmt, std::forward<Args>(args)...);
47     TEST_REQUIRE(out == expected,
48                  TEST_WRITE_CONCATENATED(
49                      "\nFormat string   ", fmt.get(), "\nExpected output ", expected, "\nActual output   ", out, '\n'));
50   }
51 #ifndef TEST_HAS_NO_LOCALIZATION
52   {
53     std::basic_string<CharT> out = std::format(std::locale(), fmt, std::forward<Args>(args)...);
54     assert(out == expected);
55   }
56 #endif // TEST_HAS_NO_LOCALIZATION
57 };
58 
59 auto test_format_to =
60     []<class CharT, class... Args>(
61         std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
62       {
63         std::basic_string<CharT> out(expected.size(), CharT(' '));
64         auto it = std::format_to(out.begin(), fmt, std::forward<Args>(args)...);
65         assert(it == out.end());
66         assert(out == expected);
67       }
68 #ifndef TEST_HAS_NO_LOCALIZATION
69       {
70         std::basic_string<CharT> out(expected.size(), CharT(' '));
71         auto it = std::format_to(out.begin(), std::locale(), fmt, std::forward<Args>(args)...);
72         assert(it == out.end());
73         assert(out == expected);
74       }
75 #endif // TEST_HAS_NO_LOCALIZATION
76       {
77         std::list<CharT> out;
78         std::format_to(std::back_inserter(out), fmt, std::forward<Args>(args)...);
79         assert(std::equal(out.begin(), out.end(), expected.begin(), expected.end()));
80       }
81       {
82         std::vector<CharT> out;
83         std::format_to(std::back_inserter(out), fmt, std::forward<Args>(args)...);
84         assert(std::equal(out.begin(), out.end(), expected.begin(), expected.end()));
85       }
86       {
87         assert(expected.size() < 4096 && "Update the size of the buffer.");
88         CharT out[4096];
89         CharT* it = std::format_to(out, fmt, std::forward<Args>(args)...);
90         assert(std::distance(out, it) == int(expected.size()));
91         // Convert to std::string since output contains '\0' for boolean tests.
92         assert(std::basic_string<CharT>(out, it) == expected);
93       }
94     };
95 
96 auto test_formatted_size =
97     []<class CharT, class... Args>(
98         std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
99       {
100         size_t size = std::formatted_size(fmt, std::forward<Args>(args)...);
101         assert(size == expected.size());
102       }
103 #ifndef TEST_HAS_NO_LOCALIZATION
104       {
105         size_t size = std::formatted_size(std::locale(), fmt, std::forward<Args>(args)...);
106         assert(size == expected.size());
107       }
108 #endif // TEST_HAS_NO_LOCALIZATION
109     };
110 
111 auto test_format_to_n =
112     []<class CharT, class... Args>(
113         std::basic_string_view<CharT> expected, test_format_string<CharT, Args...> fmt, Args&&... args) {
114       {
115         size_t n = expected.size();
116         std::basic_string<CharT> out(n, CharT(' '));
117         std::format_to_n_result result = std::format_to_n(out.begin(), n, fmt, std::forward<Args>(args)...);
118         assert(result.size == static_cast<ptrdiff_t>(expected.size()));
119         assert(result.out == out.end());
120         assert(out == expected);
121       }
122 #ifndef TEST_HAS_NO_LOCALIZATION
123       {
124         size_t n = expected.size();
125         std::basic_string<CharT> out(n, CharT(' '));
126         std::format_to_n_result result =
127             std::format_to_n(out.begin(), n, std::locale(), fmt, std::forward<Args>(args)...);
128         assert(result.size == static_cast<ptrdiff_t>(expected.size()));
129         assert(result.out == out.end());
130         assert(out == expected);
131       }
132 #endif // TEST_HAS_NO_LOCALIZATION
133       {
134         ptrdiff_t n = 0;
135         std::basic_string<CharT> out;
136         std::format_to_n_result result = std::format_to_n(out.begin(), n, fmt, std::forward<Args>(args)...);
137         assert(result.size == static_cast<ptrdiff_t>(expected.size()));
138         assert(result.out == out.end());
139         assert(out.empty());
140       }
141       {
142         ptrdiff_t n = expected.size() / 2;
143         std::basic_string<CharT> out(n, CharT(' '));
144         std::format_to_n_result result = std::format_to_n(out.begin(), n, fmt, std::forward<Args>(args)...);
145         assert(result.size == static_cast<ptrdiff_t>(expected.size()));
146         assert(result.out == out.end());
147         assert(out == expected.substr(0, n));
148       }
149     };
150 
151 template <class CharT>
152 void test_char() {
153   // *** P2286 examples ***
154   test_format(SV("['\\'', '\"']"), SV("[{:?}, {:?}]"), CharT('\''), CharT('"'));
155 
156   // *** Specical cases ***
157   test_format(SV("'\\t'"), SV("{:?}"), CharT('\t'));
158   test_format(SV("'\\n'"), SV("{:?}"), CharT('\n'));
159   test_format(SV("'\\r'"), SV("{:?}"), CharT('\r'));
160   test_format(SV("'\\\\'"), SV("{:?}"), CharT('\\'));
161 
162   test_format(SV("'\\\''"), SV("{:?}"), CharT('\''));
163   test_format(SV("'\"'"), SV("{:?}"), CharT('"')); // only special for string
164 
165   test_format(SV("' '"), SV("{:?}"), CharT(' '));
166 
167   // *** Printable ***
168   test_format(SV("'a'"), SV("{:?}"), CharT('a'));
169   test_format(SV("'b'"), SV("{:?}"), CharT('b'));
170   test_format(SV("'c'"), SV("{:?}"), CharT('c'));
171 
172   // *** Non-printable ***
173 
174   // Control
175   test_format(SV("'\\u{0}'"), SV("{:?}"), CharT('\0'));
176   test_format(SV("'\\u{1f}'"), SV("{:?}"), CharT('\x1f'));
177 
178   // Ill-formed
179   if constexpr (sizeof(CharT) == 1)
180     test_format(SV("'\\x{80}'"), SV("{:?}"), CharT('\x80'));
181 
182 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
183   if constexpr (sizeof(CharT) > 1) {
184     using V = std::basic_string_view<CharT>;
185 
186     // Unicode fitting in a 16-bit wchar_t
187 
188     // *** Non-printable ***
189 
190     // Space_Separator
191     test_format(V{L"'\\u{a0}'"}, L"{:?}", L'\xa0');     // NO-BREAK SPACE
192     test_format(V{L"'\\u{3000}'"}, L"{:?}", L'\x3000'); // IDEOGRAPHIC SPACE
193 
194     // Line_Separator
195     test_format(V{L"'\\u{2028}'"}, L"{:?}", L'\x2028'); // LINE SEPARATOR
196 
197     // Paragraph_Separator
198     test_format(V{L"'\\u{2029}'"}, L"{:?}", L'\x2029'); // PARAGRAPH SEPARATOR
199 
200     // Format
201     test_format(V{L"'\\u{ad}'"}, L"{:?}", L'\xad');     // SOFT HYPHEN
202     test_format(V{L"'\\u{600}'"}, L"{:?}", L'\x600');   // ARABIC NUMBER SIGN
203     test_format(V{L"'\\u{feff}'"}, L"{:?}", L'\xfeff'); // ZERO WIDTH NO-BREAK SPACE
204 
205     // Incomplete surrogate pair in UTF-16
206     test_format(V{L"'\\x{d800}'"}, L"{:?}", L'\xd800'); // <surrogate-D800>
207     test_format(V{L"'\\x{dfff}'"}, L"{:?}", L'\xdfff'); // <surrogate-DFFF>
208 
209     // Private_Use
210     test_format(V{L"'\\u{e000}'"}, L"{:?}", L'\xe000'); // <private-use-E000>
211     test_format(V{L"'\\u{f8ff}'"}, L"{:?}", L'\xf8ff'); // <private-use-F8FF>
212 
213     // Unassigned
214     test_format(V{L"'\\u{378}'"}, L"{:?}", L'\x378');   // <reserved-0378>
215     test_format(V{L"'\\u{1774}'"}, L"{:?}", L'\x1774'); // <reserved-1774>
216     test_format(V{L"'\\u{ffff}'"}, L"{:?}", L'\xffff'); // <noncharacter-FFFF>
217 
218     // Grapheme Extended
219     test_format(V{L"'\\u{300}'"}, L"{:?}", L'\x300');   // COMBINING GRAVE ACCENT
220     test_format(V{L"'\\u{fe20}'"}, L"{:?}", L'\xfe20'); // VARIATION SELECTOR-1
221   }
222 #  ifndef TEST_SHORT_WCHAR
223   if constexpr (sizeof(CharT) > 2) {
224     static_assert(sizeof(CharT) == 4, "add support for unexpected size");
225     // Unicode fitting in a 32-bit wchar_t
226 
227     constexpr wchar_t x  = 0x1ffff;
228     constexpr std::uint32_t y = 0x1ffff;
229     static_assert(x == y);
230 
231     using V = std::basic_string_view<CharT>;
232 
233     // *** Non-printable ***
234     // Format
235     test_format(V{L"'\\u{110bd}'"}, L"{:?}", L'\x110bd'); // KAITHI NUMBER SIGN
236     test_format(V{L"'\\u{e007f}'"}, L"{:?}", L'\xe007f'); // CANCEL TAG
237 
238     // Private_Use
239     test_format(V{L"'\\u{f0000}'"}, L"{:?}", L'\xf0000'); // <private-use-F0000>
240     test_format(V{L"'\\u{ffffd}'"}, L"{:?}", L'\xffffd'); // <private-use-FFFFD>
241 
242     test_format(V{L"'\\u{100000}'"}, L"{:?}", L'\x100000'); // <private-use-100000>
243     test_format(V{L"'\\u{10fffd}'"}, L"{:?}", L'\x10fffd'); // <private-use-10FFFD>
244 
245     // Unassigned
246     test_format(V{L"'\\u{1000c}'"}, L"{:?}", L'\x1000c');   // <reserved-1000c>
247     test_format(V{L"'\\u{fffff}'"}, L"{:?}", L'\xfffff');   // <noncharacter-FFFFF>
248     test_format(V{L"'\\u{10fffe}'"}, L"{:?}", L'\x10fffe'); // <noncharacter-10FFFE>
249 
250     // Grapheme Extended
251     test_format(V{L"'\\u{101fd}'"}, L"{:?}", L'\x101fd'); // COMBINING OLD PERMIC LETTER AN
252     test_format(V{L"'\\u{e0100}'"}, L"{:?}", L'\xe0100'); // VARIATION SELECTOR-17
253 
254     // Ill-formed
255     test_format(V{L"'\\x{110000}'"}, L"{:?}", L'\x110000');
256     test_format(V{L"'\\x{ffffffff}'"}, L"{:?}", L'\xffffffff');
257   }
258 #  endif // TEST_SHORT_WCHAR
259 #endif   // TEST_HAS_NO_WIDE_CHARACTERS
260 }
261 
262 template <class CharT>
263 void test_string() {
264   // *** P2286 examples ***
265   test_format(SV("[h\tllo]"), SV("[{}]"), SV("h\tllo"));
266   test_format(SV(R"(["h\tllo"])"), SV("[{:?}]"), SV("h\tllo"));
267   test_format(SV(R"(["Спасибо, Виктор ♥!"])"), SV("[{:?}]"), SV("Спасибо, Виктор ♥!"));
268 
269   test_format(SV(R"(["\u{0} \n \t \u{2} \u{1b}"])"), SV("[{:?}]"), SV("\0 \n \t \x02 \x1b"));
270 
271   if constexpr (sizeof(CharT) == 1) {
272     // Ill-formend UTF-8
273     test_format(SV(R"(["\x{c3}"])"), SV("[{:?}]"), "\xc3");
274     test_format(SV(R"(["\x{c3}("])"), SV("[{:?}]"), "\xc3\x28");
275 
276     /* U+0000..U+0007F 1 code unit range, encoded in 2 code units. */
277     test_format(SV(R"(["\x{c0}\x{80}"])"), SV("[{:?}]"), "\xc0\x80"); // U+0000
278     test_format(SV(R"(["\x{c1}\x{bf}"])"), SV("[{:?}]"), "\xc1\xbf"); // U+007F
279     test_format(SV(R"(["\u{80}"])"), SV("[{:?}]"), "\xc2\x80");       // U+0080 first valid (General_Category=Control)
280 
281     /* U+0000..U+07FFF 1 and 2 code unit range, encoded in 3 code units. */
282     test_format(SV(R"(["\x{e0}\x{80}\x{80}"])"), SV("[{:?}]"), "\xe0\x80\x80"); // U+0000
283     test_format(SV(R"(["\x{e0}\x{81}\x{bf}"])"), SV("[{:?}]"), "\xe0\x81\xbf"); // U+007F
284     test_format(SV(R"(["\x{e0}\x{82}\x{80}"])"), SV("[{:?}]"), "\xe0\x82\x80"); // U+0080
285     test_format(SV(R"(["\x{e0}\x{9f}\x{bf}"])"), SV("[{:?}]"), "\xe0\x9f\xbf"); // U+07FF
286     test_format(SV("[\"\u0800\"]"), SV("[{:?}]"), "\xe0\xa0\x80");              // U+0800 first valid
287 
288 #if 0
289 	// This code point is in the Hangul Jamo Extended-B block and at the time of writing
290 	// it's unassigned. When it comes defined, this branch might become true.
291     test_format(SV("[\"\ud7ff\"]"), SV("[{:?}]"), "\xed\x9f\xbf");              // U+D7FF last valid
292 #else
293     /* U+D800..D+DFFFF surrogate range */
294     test_format(SV(R"(["\u{d7ff}"])"), SV("[{:?}]"), "\xed\x9f\xbf");           // U+D7FF last valid
295 #endif
296     test_format(SV(R"(["\x{ed}\x{a0}\x{80}"])"), SV("[{:?}]"), "\xed\xa0\x80"); // U+D800
297     test_format(SV(R"(["\x{ed}\x{af}\x{bf}"])"), SV("[{:?}]"), "\xed\xaf\xbf"); // U+DBFF
298     test_format(SV(R"(["\x{ed}\x{bf}\x{80}"])"), SV("[{:?}]"), "\xed\xbf\x80"); // U+DC00
299     test_format(SV(R"(["\x{ed}\x{bf}\x{bf}"])"), SV("[{:?}]"), "\xed\xbf\xbf"); // U+DFFF
300     test_format(SV(R"(["\u{e000}"])"), SV("[{:?}]"), "\xee\x80\x80");           // U+E000 first valid
301                                                                                 // (in the Private Use Area block)
302 
303     /* U+0000..U+FFFF 1, 2, and 3 code unit range */
304     test_format(SV(R"(["\x{f0}\x{80}\x{80}\x{80}"])"), SV("[{:?}]"), "\xf0\x80\x80\x80"); // U+0000
305     test_format(SV(R"(["\x{f0}\x{80}\x{81}\x{bf}"])"), SV("[{:?}]"), "\xf0\x80\x81\xbf"); // U+007F
306     test_format(SV(R"(["\x{f0}\x{80}\x{82}\x{80}"])"), SV("[{:?}]"), "\xf0\x80\x82\x80"); // U+0080
307     test_format(SV(R"(["\x{f0}\x{80}\x{9f}\x{bf}"])"), SV("[{:?}]"), "\xf0\x80\x9f\xbf"); // U+07FF
308     test_format(SV(R"(["\x{f0}\x{80}\x{a0}\x{80}"])"), SV("[{:?}]"), "\xf0\x80\xa0\x80"); // U+0800
309     test_format(SV(R"(["\x{f0}\x{8f}\x{bf}\x{bf}"])"), SV("[{:?}]"), "\xf0\x8f\xbf\xbf"); // U+FFFF
310     test_format(SV("[\"\U00010000\"]"), SV("[{:?}]"), "\xf0\x90\x80\x80");                // U+10000 first valid
311 
312     /* U+10FFFF..U+1FFFFF invalid range */
313     test_format(SV(R"(["\u{10ffff}"])"), SV("[{:?}]"), "\xf4\x8f\xbf\xbf"); // U+10FFFF last valid
314                                                                             // (in Supplementary Private Use Area-B)
315     test_format(SV(R"(["\x{f4}\x{90}\x{80}\x{80}"])"), SV("[{:?}]"), "\xf4\x90\x80\x80"); // U+110000
316     test_format(SV(R"(["\x{f4}\x{bf}\x{bf}\x{bf}"])"), SV("[{:?}]"), "\xf4\xbf\xbf\xbf"); // U+11FFFF
317   } else {
318     // Valid UTF-16 and UTF-32
319     test_format(SV("[\"\u00c3\"]"), SV("[{:?}]"), L"\xc3"); // LATIN CAPITAL LETTER A WITH TILDE
320     test_format(SV("[\"\u00c3(\"]"), SV("[{:?}]"), L"\xc3\x28");
321   }
322 
323   test_format(SV(R"(["����\u{200d}♂\u{fe0f}"])"), SV("[{:?}]"), SV("����‍♂️"));
324 
325   // *** Specical cases ***
326   test_format(SV(R"("\t\n\r\\'\" ")"), SV("{:?}"), SV("\t\n\r\\'\" "));
327 
328   // *** Printable ***
329   test_format(SV(R"("abcdefg")"), SV("{:?}"), SV("abcdefg"));
330 
331   // *** Non-printable ***
332 
333   // Control
334   test_format(SV(R"("\u{0}\u{1f}")"), SV("{:?}"), SV("\0\x1f"));
335 
336   // Ill-formed
337   if constexpr (sizeof(CharT) == 1)
338     test_format(SV(R"("\x{80}")"), SV("{:?}"), SV("\x80"));
339 
340 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
341   if constexpr (sizeof(CharT) > 1) {
342     using V = std::basic_string_view<CharT>;
343 
344     // Unicode fitting in a 16-bit wchar_t
345 
346     // *** Non-printable ***
347 
348     // Space_Separator
349     test_format(V{LR"("\u{a0}\u{3000}")"}, L"{:?}", L"\xa0\x3000");
350 
351     // Line_Separator
352     test_format(V{LR"("\u{2028}")"}, L"{:?}", L"\x2028"); // LINE SEPARATOR
353 
354     // Paragraph_Separator
355     test_format(V{LR"("\u{2029}")"}, L"{:?}", L"\x2029"); // PARAGRAPH SEPARATOR
356 
357     // Format
358     test_format(V{LR"("\u{ad}\u{600}\u{feff}")"}, L"{:?}", L"\xad\x600\xfeff");
359 
360     // Incomplete surrogate pair in UTF-16
361     test_format(V{LR"("\x{d800}")"}, L"{:?}", L"\xd800");
362 
363     // Private_Use
364     test_format(V{LR"("\u{e000}\u{f8ff}")"}, L"{:?}", L"\xe000\xf8ff");
365 
366     // Unassigned
367     test_format(V{LR"("\u{378}\u{1774}\u{ffff}")"}, L"{:?}", L"\x378\x1774\xffff");
368 
369     // Grapheme Extended
370     test_format(V{LR"("\u{300}\u{fe20}")"}, L"{:?}", L"\x300\xfe20");
371   }
372 #  ifndef TEST_SHORT_WCHAR
373   if constexpr (sizeof(CharT) > 2) {
374     static_assert(sizeof(CharT) == 4, "add support for unexpected size");
375     // Unicode fitting in a 32-bit wchar_t
376 
377     constexpr wchar_t x  = 0x1ffff;
378     constexpr std::uint32_t y = 0x1ffff;
379     static_assert(x == y);
380 
381     using V = std::basic_string_view<CharT>;
382 
383     // *** Non-printable ***
384     // Format
385     test_format(V{LR"("\u{110bd}\u{e007f}")"}, L"{:?}", L"\x110bd\xe007f");
386 
387     // Private_Use
388     test_format(V{LR"("\u{f0000}\u{ffffd}\u{100000}\u{10fffd}")"}, L"{:?}", L"\xf0000\xffffd\x100000\x10fffd");
389 
390     // Unassigned
391     test_format(V{LR"("\u{1000c}\u{fffff}\u{10fffe}")"}, L"{:?}", L"\x1000c\xfffff\x10fffe");
392 
393     // Grapheme Extended
394     test_format(V{LR"("\u{101fd}\u{e0100}")"}, L"{:?}", L"\x101fd\xe0100");
395 
396     // Ill-formed
397     test_format(V{LR"("\x{110000}\x{ffffffff}")"}, L"{:?}", L"\x110000\xffffffff");
398   }
399 #  endif // TEST_SHORT_WCHAR
400 #endif   // TEST_HAS_NO_WIDE_CHARACTERS
401 }
402 
403 template <class CharT, class TestFunction>
404 void test_format_functions(TestFunction check) {
405   // *** align-fill & width ***
406   check(SV(R"(***"hellö")"), SV("{:*>10?}"), SV("hellö")); // ö is LATIN SMALL LETTER O WITH DIAERESIS
407   check(SV(R"(*"hellö"**)"), SV("{:*^10?}"), SV("hellö"));
408   check(SV(R"("hellö"***)"), SV("{:*<10?}"), SV("hellö"));
409 
410   check(SV(R"("hello\u{308}")"), SV("{:*>10?}"), SV("hello\u0308"));
411   check(SV(R"(***"hello\u{308}")"), SV("{:*>17?}"), SV("hello\u0308"));
412   check(SV(R"(*"hello\u{308}"**)"), SV("{:*^17?}"), SV("hello\u0308"));
413   check(SV(R"("hello\u{308}"***)"), SV("{:*<17?}"), SV("hello\u0308"));
414 
415   check(SV(R"("hello ����\u{200d}♂\u{fe0f}")"), SV("{:*>10?}"), SV("hello ����‍♂️"));
416   check(SV(R"(***"hello ����\u{200d}♂\u{fe0f}")"), SV("{:*>30?}"), SV("hello ����‍♂️"));
417   check(SV(R"(*"hello ����\u{200d}♂\u{fe0f}"**)"), SV("{:*^30?}"), SV("hello ����‍♂️"));
418   check(SV(R"("hello ����\u{200d}♂\u{fe0f}"***)"), SV("{:*<30?}"), SV("hello ����‍♂️"));
419 
420   // *** width ***
421   check(SV(R"("hellö"   )"), SV("{:10?}"), SV("hellö"));
422   check(SV(R"("hello\u{308}"   )"), SV("{:17?}"), SV("hello\u0308"));
423   check(SV(R"("hello ����\u{200d}♂\u{fe0f}"   )"), SV("{:30?}"), SV("hello ����‍♂️"));
424 
425   // *** precision ***
426   check(SV(R"("hell)"), SV("{:.5?}"), SV("hellö"));
427   check(SV(R"("hellö)"), SV("{:.6?}"), SV("hellö"));
428   check(SV(R"("hellö")"), SV("{:.7?}"), SV("hellö"));
429 
430   check(SV(R"("hello )"), SV("{:.7?}"), SV("hello ����‍♂️"));
431   check(SV(R"("hello )"), SV("{:.8?}"), SV("hello ����‍♂️")); // shrug is two columns
432   check(SV(R"("hello ����)"), SV("{:.9?}"), SV("hello ����‍♂️"));
433   check(SV(R"("hello ����\)"), SV("{:.10?}"), SV("hello ����‍♂️"));
434   check(SV(R"("hello ����\u{200d})"), SV("{:.17?}"), SV("hello ����‍♂️"));
435   check(SV(R"("hello ����\u{200d}♂)"), SV("{:.18?}"), SV("hello ����‍♂️"));
436   check(SV(R"("hello ����\u{200d}♂\)"), SV("{:.19?}"), SV("hello ����‍♂️"));
437   check(SV(R"("hello ����\u{200d}♂\u{fe0f}")"), SV("{:.28?}"), SV("hello ����‍♂️"));
438 
439   // *** width & precision ***
440   check(SV(R"("hell#########################)"), SV("{:#<30.5?}"), SV("hellö"));
441   check(SV(R"("hellö########################)"), SV("{:#<30.6?}"), SV("hellö"));
442   check(SV(R"("hellö"#######################)"), SV("{:#<30.7?}"), SV("hellö"));
443 
444   check(SV(R"("hello #######################)"), SV("{:#<30.7?}"), SV("hello ����‍♂️"));
445   check(SV(R"("hello #######################)"), SV("{:#<30.8?}"), SV("hello ����‍♂️"));
446   check(SV(R"("hello ����#####################)"), SV("{:#<30.9?}"), SV("hello ����‍♂️"));
447   check(SV(R"("hello ����\####################)"), SV("{:#<30.10?}"), SV("hello ����‍♂️"));
448   check(SV(R"("hello ����\u{200d}#############)"), SV("{:#<30.17?}"), SV("hello ����‍♂️"));
449   check(SV(R"("hello ����\u{200d}♂############)"), SV("{:#<30.18?}"), SV("hello ����‍♂️"));
450   check(SV(R"("hello ����\u{200d}♂\###########)"), SV("{:#<30.19?}"), SV("hello ����‍♂️"));
451   check(SV(R"("hello ����\u{200d}♂\u{fe0f}"###)"), SV("{:#<30.28?}"), SV("hello ����‍♂️"));
452 }
453 
454 template <class CharT>
455 void test() {
456   test_char<CharT>();
457   test_string<CharT>();
458 
459   test_format_functions<CharT>(test_format);
460   test_format_functions<CharT>(test_format_to);
461   test_format_functions<CharT>(test_formatted_size);
462   test_format_functions<CharT>(test_format_to_n);
463 }
464 
465 static void test_ill_formed_utf8() {
466   using namespace std::literals;
467 
468   // Too few code units
469   test_format(R"("\x{df}")"sv, "{:?}", "\xdf");
470   test_format(R"("\x{ef}")"sv, "{:?}", "\xef");
471   test_format(R"("\x{ef}\x{bf}")"sv, "{:?}", "\xef\xbf");
472   test_format(R"("\x{f7}")"sv, "{:?}", "\xf7");
473   test_format(R"("\x{f7}\x{bf}")"sv, "{:?}", "\xf7\xbf");
474   test_format(R"("\x{f7}\x{bf}\x{bf}")"sv, "{:?}", "\xf7\xbf\xbf");
475 
476   // Invalid continuation byte
477   test_format(R"("\x{df}a")"sv,
478               "{:?}",
479               "\xdf"
480               "a");
481   test_format(R"("\x{ef}a")"sv,
482               "{:?}",
483               "\xef"
484               "a");
485   test_format(R"("\x{ef}\x{bf}a")"sv,
486               "{:?}",
487               "\xef\xbf"
488               "a");
489   test_format(R"("\x{f7}a")"sv,
490               "{:?}",
491               "\xf7"
492               "a");
493   test_format(R"("\x{f7}\x{bf}a")"sv,
494               "{:?}",
495               "\xf7\xbf"
496               "a");
497   test_format(R"("\x{f7}\x{bf}\x{bf}a")"sv,
498               "{:?}",
499               "\xf7\xbf\xbf"
500               "a");
501 
502   // Code unit out of range
503   test_format(R"("\u{10ffff}")"sv, "{:?}", "\xf4\x8f\xbf\xbf");               // last valid code point
504   test_format(R"("\x{f4}\x{90}\x{80}\x{80}")"sv, "{:?}", "\xf4\x90\x80\x80"); // first invalid code point
505   test_format(R"("\x{f5}\x{b1}\x{b2}\x{b3}")"sv, "{:?}", "\xf5\xb1\xb2\xb3");
506   test_format(R"("\x{f7}\x{bf}\x{bf}\x{bf}")"sv, "{:?}", "\xf7\xbf\xbf\xbf"); // largest encoded code point
507 }
508 
509 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
510 #  ifdef _LIBCPP_SHORT_WCHAR
511 static void test_ill_formed_utf16() {
512   using namespace std::literals;
513 
514   // Too few code units
515   test_format(LR"("\x{d800}")"sv, L"{:?}", L"\xd800");
516   test_format(LR"("\x{dbff}")"sv, L"{:?}", L"\xdbff");
517 
518   // Start with low surrogate pair
519   test_format(LR"("\x{dc00}a")"sv,
520               L"{:?}",
521               L"\xdc00"
522               "a");
523   test_format(LR"("\x{dfff}a")"sv,
524               L"{:?}",
525               L"\xdfff"
526               "a");
527 
528   // Only high surrogate pair
529   test_format(LR"("\x{d800}a")"sv,
530               L"{:?}",
531               L"\xd800"
532               "a");
533   test_format(LR"("\x{dbff}a")"sv,
534               L"{:?}",
535               L"\xdbff"
536               "a");
537 }
538 #  else // _LIBCPP_SHORT_WCHAR
539 static void test_ill_formed_utf32() {
540   using namespace std::literals;
541 
542   test_format(LR"("\u{10ffff}")"sv, L"{:?}", L"\x10ffff");     // last valid code point
543   test_format(LR"("\x{110000}")"sv, L"{:?}", L"\x110000");     // first invalid code point
544   test_format(LR"("\x{ffffffff}")"sv, L"{:?}", L"\xffffffff"); // largest encoded code point
545 }
546 
547 #  endif // _LIBCPP_SHORT_WCHAR
548 #endif   // TEST_HAS_NO_WIDE_CHARACTERS
549 
550 int main(int, char**) {
551   test<char>();
552 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
553   test<wchar_t>();
554 #endif
555 
556   test_ill_formed_utf8();
557 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
558 #  ifdef _LIBCPP_SHORT_WCHAR
559   test_ill_formed_utf16();
560   assert(false);
561 #  else  // _LIBCPP_SHORT_WCHAR
562   test_ill_formed_utf32();
563 #  endif // _LIBCPP_SHORT_WCHAR
564 #endif   // TEST_HAS_NO_WIDE_CHARACTERS
565 
566   return 0;
567 }
568