1 // SPDX-License-Identifier: GPL-2.0+
3 * Unit tests for Unicode functions
5 * Copyright (c) 2018 Heinrich Schuchardt <xypron.glpk@gmx.de>
13 #include <test/test.h>
14 #include <test/suites.h>
17 /* Linker list entry for a Unicode test */
18 #define UNICODE_TEST(_name) UNIT_TEST(_name, 0, unicode_test)
20 /* Constants c1-c4 and d1-d4 encode the same letters */
22 /* Six characters translating to one utf-8 byte each. */
23 static const u16 c1[] = {0x55, 0x2d, 0x42, 0x6f, 0x6f, 0x74, 0x00};
24 /* One character translating to two utf-8 bytes */
25 static const u16 c2[] = {0x6b, 0x61, 0x66, 0x62, 0xe1, 0x74, 0x75, 0x72, 0x00};
26 /* Three characters translating to three utf-8 bytes each */
27 static const u16 c3[] = {0x6f5c, 0x6c34, 0x8266, 0x00};
28 /* Three letters translating to four utf-8 bytes each */
29 static const u16 c4[] = {0xd801, 0xdc8d, 0xd801, 0xdc96, 0xd801, 0xdc87,
32 /* Illegal utf-16 strings */
33 static const u16 i1[] = {0x69, 0x31, 0xdc87, 0x6c, 0x00};
34 static const u16 i2[] = {0x69, 0x32, 0xd801, 0xd801, 0x6c, 0x00};
35 static const u16 i3[] = {0x69, 0x33, 0xd801, 0x00};
37 /* Six characters translating to one utf-16 word each. */
38 static const char d1[] = {0x55, 0x2d, 0x42, 0x6f, 0x6f, 0x74, 0x00};
39 /* Eight characters translating to one utf-16 word each */
40 static const char d2[] = {0x6b, 0x61, 0x66, 0x62, 0xc3, 0xa1, 0x74, 0x75,
42 /* Three characters translating to one utf-16 word each */
43 static const char d3[] = {0xe6, 0xbd, 0x9c, 0xe6, 0xb0, 0xb4, 0xe8, 0x89,
45 /* Three letters translating to two utf-16 word each */
46 static const char d4[] = {0xf0, 0x90, 0x92, 0x8d, 0xf0, 0x90, 0x92, 0x96,
47 0xf0, 0x90, 0x92, 0x87, 0x00};
49 /* Illegal utf-8 strings */
50 static const char j1[] = {0x6a, 0x31, 0xa1, 0x6c, 0x00};
51 static const char j2[] = {0x6a, 0x32, 0xc3, 0xc3, 0x6c, 0x00};
52 static const char j3[] = {0x6a, 0x33, 0xf0, 0x90, 0xf0, 0x00};
54 static int unicode_test_u16_strlen(struct unit_test_state *uts)
56 ut_asserteq(6, u16_strlen(c1));
57 ut_asserteq(8, u16_strlen(c2));
58 ut_asserteq(3, u16_strlen(c3));
59 ut_asserteq(6, u16_strlen(c4));
62 UNICODE_TEST(unicode_test_u16_strlen);
64 static int unicode_test_u16_strdup(struct unit_test_state *uts)
66 u16 *copy = u16_strdup(c4);
68 ut_assert(copy != c4);
69 ut_assert(!memcmp(copy, c4, sizeof(c4)));
73 UNICODE_TEST(unicode_test_u16_strdup);
75 static int unicode_test_u16_strcpy(struct unit_test_state *uts)
80 r = u16_strcpy(copy, c1);
82 ut_assert(!memcmp(copy, c1, sizeof(c1)));
85 UNICODE_TEST(unicode_test_u16_strcpy);
87 /* U-Boot uses UTF-16 strings in the EFI context only. */
88 #if CONFIG_IS_ENABLED(EFI_LOADER) && !defined(API_BUILD)
89 static int unicode_test_string16(struct unit_test_state *uts)
93 /* Test length and precision */
94 memset(buf, 0xff, sizeof(buf));
95 sprintf(buf, "%8.6ls", c2);
96 ut_asserteq(' ', buf[1]);
97 ut_assert(!strncmp(&buf[2], d2, 7));
100 memset(buf, 0xff, sizeof(buf));
101 sprintf(buf, "%8.6ls", c4);
102 ut_asserteq(' ', buf[4]);
103 ut_assert(!strncmp(&buf[5], d4, 12));
106 memset(buf, 0xff, sizeof(buf));
107 sprintf(buf, "%-8.2ls", c4);
108 ut_asserteq(' ', buf[8]);
109 ut_assert(!strncmp(buf, d4, 8));
112 /* Test handling of illegal utf-16 sequences */
113 memset(buf, 0xff, sizeof(buf));
114 sprintf(buf, "%ls", i1);
115 ut_asserteq_str("i1?l", buf);
117 memset(buf, 0xff, sizeof(buf));
118 sprintf(buf, "%ls", i2);
119 ut_asserteq_str("i2?l", buf);
121 memset(buf, 0xff, sizeof(buf));
122 sprintf(buf, "%ls", i3);
123 ut_asserteq_str("i3?", buf);
127 UNICODE_TEST(unicode_test_string16);
130 static int unicode_test_utf8_get(struct unit_test_state *uts)
136 /* Check characters less than 0x800 */
138 for (i = 0; i < 8; ++i) {
139 code = utf8_get((const char **)&s);
140 /* c2 is the utf-8 encoding of d2 */
141 ut_asserteq(c2[i], code);
145 ut_asserteq_ptr(s, d2 + 9)
147 /* Check characters less than 0x10000 */
149 for (i = 0; i < 4; ++i) {
150 code = utf8_get((const char **)&s);
151 /* c3 is the utf-8 encoding of d3 */
152 ut_asserteq(c3[i], code);
156 ut_asserteq_ptr(s, d3 + 9)
158 /* Check character greater 0xffff */
160 code = utf8_get((const char **)&s);
161 ut_asserteq(0x0001048d, code);
162 ut_asserteq_ptr(s, d4 + 4);
166 UNICODE_TEST(unicode_test_utf8_get);
168 static int unicode_test_utf8_put(struct unit_test_state *uts)
170 char buffer[8] = { 0, };
173 /* Commercial at, translates to one character */
175 ut_assert(!utf8_put('@', &pos))
176 ut_asserteq(1, pos - buffer);
177 ut_asserteq('@', buffer[0]);
178 ut_assert(!buffer[1]);
180 /* Latin letter G with acute, translates to two charactes */
182 ut_assert(!utf8_put(0x1f4, &pos));
183 ut_asserteq(2, pos - buffer);
184 ut_asserteq_str("\xc7\xb4", buffer);
186 /* Tagalog letter i, translates to three characters */
188 ut_assert(!utf8_put(0x1701, &pos));
189 ut_asserteq(3, pos - buffer);
190 ut_asserteq_str("\xe1\x9c\x81", buffer);
192 /* Hamster face, translates to four characters */
194 ut_assert(!utf8_put(0x1f439, &pos));
195 ut_asserteq(4, pos - buffer);
196 ut_asserteq_str("\xf0\x9f\x90\xb9", buffer);
200 ut_asserteq(-1, utf8_put(0xd888, &pos));
204 UNICODE_TEST(unicode_test_utf8_put);
206 static int unicode_test_utf8_utf16_strlen(struct unit_test_state *uts)
208 ut_asserteq(6, utf8_utf16_strlen(d1));
209 ut_asserteq(8, utf8_utf16_strlen(d2));
210 ut_asserteq(3, utf8_utf16_strlen(d3));
211 ut_asserteq(6, utf8_utf16_strlen(d4));
213 /* illegal utf-8 sequences */
214 ut_asserteq(4, utf8_utf16_strlen(j1));
215 ut_asserteq(4, utf8_utf16_strlen(j2));
216 ut_asserteq(3, utf8_utf16_strlen(j3));
220 UNICODE_TEST(unicode_test_utf8_utf16_strlen);
222 static int unicode_test_utf8_utf16_strnlen(struct unit_test_state *uts)
224 ut_asserteq(3, utf8_utf16_strnlen(d1, 3));
225 ut_asserteq(6, utf8_utf16_strnlen(d1, 13));
226 ut_asserteq(6, utf8_utf16_strnlen(d2, 6));
227 ut_asserteq(2, utf8_utf16_strnlen(d3, 2));
228 ut_asserteq(4, utf8_utf16_strnlen(d4, 2));
229 ut_asserteq(6, utf8_utf16_strnlen(d4, 3));
231 /* illegal utf-8 sequences */
232 ut_asserteq(4, utf8_utf16_strnlen(j1, 16));
233 ut_asserteq(4, utf8_utf16_strnlen(j2, 16));
234 ut_asserteq(3, utf8_utf16_strnlen(j3, 16));
238 UNICODE_TEST(unicode_test_utf8_utf16_strnlen);
241 * ut_u16_strcmp() - Compare to u16 strings.
245 * @count: number of u16 to compare
246 * Return: -1 if a1 < a2, 0 if a1 == a2, 1 if a1 > a2
248 static int unicode_test_u16_strcmp(const u16 *a1, const u16 *a2, size_t count)
250 for (; (*a1 || *a2) && count; ++a1, ++a2, --count) {
259 static int unicode_test_utf8_utf16_strcpy(struct unit_test_state *uts)
265 utf8_utf16_strcpy(&pos, d1);
266 ut_asserteq(6, pos - buf);
267 ut_assert(!unicode_test_u16_strcmp(buf, c1, SIZE_MAX));
270 utf8_utf16_strcpy(&pos, d2);
271 ut_asserteq(8, pos - buf);
272 ut_assert(!unicode_test_u16_strcmp(buf, c2, SIZE_MAX));
275 utf8_utf16_strcpy(&pos, d3);
276 ut_asserteq(3, pos - buf);
277 ut_assert(!unicode_test_u16_strcmp(buf, c3, SIZE_MAX));
280 utf8_utf16_strcpy(&pos, d4);
281 ut_asserteq(6, pos - buf);
282 ut_assert(!unicode_test_u16_strcmp(buf, c4, SIZE_MAX));
284 /* Illegal utf-8 strings */
286 utf8_utf16_strcpy(&pos, j1);
287 ut_asserteq(4, pos - buf);
288 ut_assert(!unicode_test_u16_strcmp(buf, L"j1?l", SIZE_MAX));
291 utf8_utf16_strcpy(&pos, j2);
292 ut_asserteq(4, pos - buf);
293 ut_assert(!unicode_test_u16_strcmp(buf, L"j2?l", SIZE_MAX));
296 utf8_utf16_strcpy(&pos, j3);
297 ut_asserteq(3, pos - buf);
298 ut_assert(!unicode_test_u16_strcmp(buf, L"j3?", SIZE_MAX));
302 UNICODE_TEST(unicode_test_utf8_utf16_strcpy);
304 static int unicode_test_utf8_utf16_strncpy(struct unit_test_state *uts)
310 memset(buf, 0, sizeof(buf));
311 utf8_utf16_strncpy(&pos, d1, 4);
312 ut_asserteq(4, pos - buf);
314 ut_assert(!unicode_test_u16_strcmp(buf, c1, 4));
317 memset(buf, 0, sizeof(buf));
318 utf8_utf16_strncpy(&pos, d2, 10);
319 ut_asserteq(8, pos - buf);
321 ut_assert(!unicode_test_u16_strcmp(buf, c2, SIZE_MAX));
324 memset(buf, 0, sizeof(buf));
325 utf8_utf16_strncpy(&pos, d3, 2);
326 ut_asserteq(2, pos - buf);
328 ut_assert(!unicode_test_u16_strcmp(buf, c3, 2));
331 memset(buf, 0, sizeof(buf));
332 utf8_utf16_strncpy(&pos, d4, 2);
333 ut_asserteq(4, pos - buf);
335 ut_assert(!unicode_test_u16_strcmp(buf, c4, 4));
338 memset(buf, 0, sizeof(buf));
339 utf8_utf16_strncpy(&pos, d4, 10);
340 ut_asserteq(6, pos - buf);
342 ut_assert(!unicode_test_u16_strcmp(buf, c4, SIZE_MAX));
346 UNICODE_TEST(unicode_test_utf8_utf16_strncpy);
348 static int unicode_test_utf16_get(struct unit_test_state *uts)
354 /* Check characters less than 0x10000 */
356 for (i = 0; i < 9; ++i) {
357 code = utf16_get((const u16 **)&s);
358 ut_asserteq(c2[i], code);
362 ut_asserteq_ptr(c2 + 8, s);
364 /* Check character greater 0xffff */
366 code = utf16_get((const u16 **)&s);
367 ut_asserteq(0x0001048d, code);
368 ut_asserteq_ptr(c4 + 2, s);
372 UNICODE_TEST(unicode_test_utf16_get);
374 static int unicode_test_utf16_put(struct unit_test_state *uts)
376 u16 buffer[4] = { 0, };
379 /* Commercial at, translates to one word */
381 ut_assert(!utf16_put('@', &pos));
382 ut_asserteq(1, pos - buffer);
383 ut_asserteq((u16)'@', buffer[0]);
384 ut_assert(!buffer[1]);
386 /* Hamster face, translates to two words */
388 ut_assert(!utf16_put(0x1f439, &pos));
389 ut_asserteq(2, pos - buffer);
390 ut_asserteq((u16)0xd83d, buffer[0]);
391 ut_asserteq((u16)0xdc39, buffer[1]);
392 ut_assert(!buffer[2]);
396 ut_asserteq(-1, utf16_put(0xd888, &pos));
400 UNICODE_TEST(unicode_test_utf16_put);
402 static int unicode_test_utf16_strnlen(struct unit_test_state *uts)
404 ut_asserteq(3, utf16_strnlen(c1, 3));
405 ut_asserteq(6, utf16_strnlen(c1, 13));
406 ut_asserteq(6, utf16_strnlen(c2, 6));
407 ut_asserteq(2, utf16_strnlen(c3, 2));
408 ut_asserteq(2, utf16_strnlen(c4, 2));
409 ut_asserteq(3, utf16_strnlen(c4, 3));
411 /* illegal utf-16 word sequences */
412 ut_asserteq(4, utf16_strnlen(i1, 16));
413 ut_asserteq(4, utf16_strnlen(i2, 16));
414 ut_asserteq(3, utf16_strnlen(i3, 16));
418 UNICODE_TEST(unicode_test_utf16_strnlen);
420 static int unicode_test_utf16_utf8_strlen(struct unit_test_state *uts)
422 ut_asserteq(6, utf16_utf8_strlen(c1));
423 ut_asserteq(9, utf16_utf8_strlen(c2));
424 ut_asserteq(9, utf16_utf8_strlen(c3));
425 ut_asserteq(12, utf16_utf8_strlen(c4));
427 /* illegal utf-16 word sequences */
428 ut_asserteq(4, utf16_utf8_strlen(i1));
429 ut_asserteq(4, utf16_utf8_strlen(i2));
430 ut_asserteq(3, utf16_utf8_strlen(i3));
434 UNICODE_TEST(unicode_test_utf16_utf8_strlen);
436 static int unicode_test_utf16_utf8_strnlen(struct unit_test_state *uts)
438 ut_asserteq(3, utf16_utf8_strnlen(c1, 3));
439 ut_asserteq(6, utf16_utf8_strnlen(c1, 13));
440 ut_asserteq(7, utf16_utf8_strnlen(c2, 6));
441 ut_asserteq(6, utf16_utf8_strnlen(c3, 2));
442 ut_asserteq(8, utf16_utf8_strnlen(c4, 2));
443 ut_asserteq(12, utf16_utf8_strnlen(c4, 3));
446 UNICODE_TEST(unicode_test_utf16_utf8_strnlen);
448 static int unicode_test_utf16_utf8_strcpy(struct unit_test_state *uts)
454 utf16_utf8_strcpy(&pos, c1);
455 ut_asserteq(6, pos - buf);
456 ut_asserteq_str(d1, buf);
459 utf16_utf8_strcpy(&pos, c2);
460 ut_asserteq(9, pos - buf);
461 ut_asserteq_str(d2, buf);
464 utf16_utf8_strcpy(&pos, c3);
465 ut_asserteq(9, pos - buf);
466 ut_asserteq_str(d3, buf);
469 utf16_utf8_strcpy(&pos, c4);
470 ut_asserteq(12, pos - buf);
471 ut_asserteq_str(d4, buf);
473 /* Illegal utf-16 strings */
475 utf16_utf8_strcpy(&pos, i1);
476 ut_asserteq(4, pos - buf);
477 ut_asserteq_str("i1?l", buf);
480 utf16_utf8_strcpy(&pos, i2);
481 ut_asserteq(4, pos - buf);
482 ut_asserteq_str("i2?l", buf);
485 utf16_utf8_strcpy(&pos, i3);
486 ut_asserteq(3, pos - buf);
487 ut_asserteq_str("i3?", buf);
491 UNICODE_TEST(unicode_test_utf16_utf8_strcpy);
493 static int unicode_test_utf16_utf8_strncpy(struct unit_test_state *uts)
499 memset(buf, 0, sizeof(buf));
500 utf16_utf8_strncpy(&pos, c1, 4);
501 ut_asserteq(4, pos - buf);
503 ut_assert(!strncmp(buf, d1, 4));
506 memset(buf, 0, sizeof(buf));
507 utf16_utf8_strncpy(&pos, c2, 10);
508 ut_asserteq(9, pos - buf);
510 ut_assert(!strncmp(buf, d2, SIZE_MAX));
513 memset(buf, 0, sizeof(buf));
514 utf16_utf8_strncpy(&pos, c3, 2);
515 ut_asserteq(6, pos - buf);
517 ut_assert(!strncmp(buf, d3, 6));
520 memset(buf, 0, sizeof(buf));
521 utf16_utf8_strncpy(&pos, c4, 2);
522 ut_asserteq(8, pos - buf);
524 ut_assert(!strncmp(buf, d4, 8));
527 memset(buf, 0, sizeof(buf));
528 utf16_utf8_strncpy(&pos, c4, 10);
529 ut_asserteq(12, pos - buf);
531 ut_assert(!strncmp(buf, d4, SIZE_MAX));
535 UNICODE_TEST(unicode_test_utf16_utf8_strncpy);
537 static int unicode_test_utf_to_lower(struct unit_test_state *uts)
539 ut_asserteq('@', utf_to_lower('@'));
540 ut_asserteq('a', utf_to_lower('A'));
541 ut_asserteq('z', utf_to_lower('Z'));
542 ut_asserteq('[', utf_to_lower('['));
543 ut_asserteq('m', utf_to_lower('m'));
544 /* Latin letter O with diaresis (umlaut) */
545 ut_asserteq(0x00f6, utf_to_lower(0x00d6));
546 #ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
547 /* Cyrillic letter I*/
548 ut_asserteq(0x0438, utf_to_lower(0x0418));
552 UNICODE_TEST(unicode_test_utf_to_lower);
554 static int unicode_test_utf_to_upper(struct unit_test_state *uts)
556 ut_asserteq('`', utf_to_upper('`'));
557 ut_asserteq('A', utf_to_upper('a'));
558 ut_asserteq('Z', utf_to_upper('z'));
559 ut_asserteq('{', utf_to_upper('{'));
560 ut_asserteq('M', utf_to_upper('M'));
561 /* Latin letter O with diaresis (umlaut) */
562 ut_asserteq(0x00d6, utf_to_upper(0x00f6));
563 #ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
564 /* Cyrillic letter I */
565 ut_asserteq(0x0418, utf_to_upper(0x0438));
569 UNICODE_TEST(unicode_test_utf_to_upper);
571 static int unicode_test_u16_strncmp(struct unit_test_state *uts)
573 ut_assert(u16_strncmp(L"abc", L"abc", 3) == 0);
574 ut_assert(u16_strncmp(L"abcdef", L"abcghi", 3) == 0);
575 ut_assert(u16_strncmp(L"abcdef", L"abcghi", 6) < 0);
576 ut_assert(u16_strncmp(L"abcghi", L"abcdef", 6) > 0);
577 ut_assert(u16_strcmp(L"abc", L"abc") == 0);
578 ut_assert(u16_strcmp(L"abcdef", L"deghi") < 0);
579 ut_assert(u16_strcmp(L"deghi", L"abcdef") > 0);
582 UNICODE_TEST(unicode_test_u16_strncmp);
584 int do_ut_unicode(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
586 struct unit_test *tests = ll_entry_start(struct unit_test, unicode_test);
587 const int n_ents = ll_entry_count(struct unit_test, unicode_test);
589 return cmd_ut_category("Unicode", "unicode_test_",
590 tests, n_ents, argc, argv);