xref: /openbmc/u-boot/test/unicode_ut.c (revision 3fda0262)
1 // SPDX-License-Identifier: GPL-2.0+
2 /*
3  * Unit tests for Unicode functions
4  *
5  * Copyright (c) 2018 Heinrich Schuchardt <xypron.glpk@gmx.de>
6  */
7 
8 #include <common.h>
9 #include <charset.h>
10 #include <command.h>
11 #include <errno.h>
12 #include <test/test.h>
13 #include <test/suites.h>
14 #include <test/ut.h>
15 
16 /* Linker list entry for a Unicode test */
17 #define UNICODE_TEST(_name) UNIT_TEST(_name, 0, unicode_test)
18 
19 /* Constants c1-c4 and d1-d4 encode the same letters */
20 
21 /* Six characters translating to one utf-8 byte each. */
22 static const u16 c1[] = {0x55, 0x2d, 0x42, 0x6f, 0x6f, 0x74, 0x00};
23 /* One character translating to two utf-8 bytes */
24 static const u16 c2[] = {0x6b, 0x61, 0x66, 0x62, 0xe1, 0x74, 0x75, 0x72, 0x00};
25 /* Three characters translating to three utf-8 bytes each */
26 static const u16 c3[] = {0x6f5c, 0x6c34, 0x8266, 0x00};
27 /* Three letters translating to four utf-8 bytes each */
28 static const u16 c4[] = {0xd801, 0xdc8d, 0xd801, 0xdc96, 0xd801, 0xdc87,
29 			 0x0000};
30 
31 /* Illegal utf-16 strings */
32 static const u16 i1[] = {0x69, 0x31, 0xdc87, 0x6c, 0x00};
33 static const u16 i2[] = {0x69, 0x32, 0xd801, 0xd801, 0x6c, 0x00};
34 static const u16 i3[] = {0x69, 0x33, 0xd801, 0x00};
35 
36 /* Six characters translating to one utf-16 word each. */
37 static const char d1[] = {0x55, 0x2d, 0x42, 0x6f, 0x6f, 0x74, 0x00};
38 /* Eight characters translating to one utf-16 word each */
39 static const char d2[] = {0x6b, 0x61, 0x66, 0x62, 0xc3, 0xa1, 0x74, 0x75,
40 			  0x72, 0x00};
41 /* Three characters translating to one utf-16 word each */
42 static const char d3[] = {0xe6, 0xbd, 0x9c, 0xe6, 0xb0, 0xb4, 0xe8, 0x89,
43 			  0xa6, 0x00};
44 /* Three letters translating to two utf-16 word each */
45 static const char d4[] = {0xf0, 0x90, 0x92, 0x8d, 0xf0, 0x90, 0x92, 0x96,
46 			  0xf0, 0x90, 0x92, 0x87, 0x00};
47 
48 /* Illegal utf-8 strings */
49 static const char j1[] = {0x6a, 0x31, 0xa1, 0x6c, 0x00};
50 static const char j2[] = {0x6a, 0x32, 0xc3, 0xc3, 0x6c, 0x00};
51 static const char j3[] = {0x6a, 0x33, 0xf0, 0x90, 0xf0, 0x00};
52 
53 static int unicode_test_u16_strdup(struct unit_test_state *uts)
54 {
55 	u16 *copy = u16_strdup(c4);
56 
57 	ut_assert(copy != c4);
58 	ut_assert(!memcmp(copy, c4, sizeof(c4)));
59 	free(copy);
60 	return 0;
61 }
62 UNICODE_TEST(unicode_test_u16_strdup);
63 
64 static int unicode_test_u16_strcpy(struct unit_test_state *uts)
65 {
66 	u16 *r;
67 	u16 copy[10];
68 
69 	r = u16_strcpy(copy, c1);
70 	ut_assert(r == copy);
71 	ut_assert(!memcmp(copy, c1, sizeof(c1)));
72 	return 0;
73 }
74 UNICODE_TEST(unicode_test_u16_strcpy);
75 
76 /* U-Boot uses UTF-16 strings in the EFI context only. */
77 #if CONFIG_IS_ENABLED(EFI_LOADER) && !defined(API_BUILD)
78 static int unicode_test_string16(struct unit_test_state *uts)
79 {
80 	char buf[20];
81 
82 	/* Test length and precision */
83 	memset(buf, 0xff, sizeof(buf));
84 	sprintf(buf, "%8.6ls", c2);
85 	ut_asserteq(' ', buf[1]);
86 	ut_assert(!strncmp(&buf[2], d2, 7));
87 	ut_assert(!buf[9]);
88 
89 	memset(buf, 0xff, sizeof(buf));
90 	sprintf(buf, "%8.6ls", c4);
91 	ut_asserteq(' ', buf[4]);
92 	ut_assert(!strncmp(&buf[5], d4, 12));
93 	ut_assert(!buf[17]);
94 
95 	memset(buf, 0xff, sizeof(buf));
96 	sprintf(buf, "%-8.2ls", c4);
97 	ut_asserteq(' ', buf[8]);
98 	ut_assert(!strncmp(buf, d4, 8));
99 	ut_assert(!buf[14]);
100 
101 	/* Test handling of illegal utf-16 sequences */
102 	memset(buf, 0xff, sizeof(buf));
103 	sprintf(buf, "%ls", i1);
104 	ut_asserteq_str("i1?l", buf);
105 
106 	memset(buf, 0xff, sizeof(buf));
107 	sprintf(buf, "%ls", i2);
108 	ut_asserteq_str("i2?l", buf);
109 
110 	memset(buf, 0xff, sizeof(buf));
111 	sprintf(buf, "%ls", i3);
112 	ut_asserteq_str("i3?", buf);
113 
114 	return 0;
115 }
116 UNICODE_TEST(unicode_test_string16);
117 #endif
118 
119 static int unicode_test_utf8_get(struct unit_test_state *uts)
120 {
121 	const char *s;
122 	s32 code;
123 	int i;
124 
125 	/* Check characters less than 0x800 */
126 	s = d2;
127 	for (i = 0; i < 8; ++i) {
128 		code = utf8_get((const char **)&s);
129 		/* c2 is the utf-8 encoding of d2 */
130 		ut_asserteq(c2[i], code);
131 		if (!code)
132 			break;
133 	}
134 	ut_asserteq_ptr(s, d2 + 9)
135 
136 	/* Check characters less than 0x10000 */
137 	s = d3;
138 	for (i = 0; i < 4; ++i) {
139 		code = utf8_get((const char **)&s);
140 		/* c3 is the utf-8 encoding of d3 */
141 		ut_asserteq(c3[i], code);
142 		if (!code)
143 			break;
144 	}
145 	ut_asserteq_ptr(s, d3 + 9)
146 
147 	/* Check character greater 0xffff */
148 	s = d4;
149 	code = utf8_get((const char **)&s);
150 	ut_asserteq(0x0001048d, code);
151 	ut_asserteq_ptr(s, d4 + 4);
152 
153 	return 0;
154 }
155 UNICODE_TEST(unicode_test_utf8_get);
156 
157 static int unicode_test_utf8_put(struct unit_test_state *uts)
158 {
159 	char buffer[8] = { 0, };
160 	char *pos;
161 
162 	/* Commercial at, translates to one character */
163 	pos = buffer;
164 	ut_assert(!utf8_put('@', &pos))
165 	ut_asserteq(1, pos - buffer);
166 	ut_asserteq('@', buffer[0]);
167 	ut_assert(!buffer[1]);
168 
169 	/* Latin letter G with acute, translates to two charactes */
170 	pos = buffer;
171 	ut_assert(!utf8_put(0x1f4, &pos));
172 	ut_asserteq(2, pos - buffer);
173 	ut_asserteq_str("\xc7\xb4", buffer);
174 
175 	/* Tagalog letter i, translates to three characters */
176 	pos = buffer;
177 	ut_assert(!utf8_put(0x1701, &pos));
178 	ut_asserteq(3, pos - buffer);
179 	ut_asserteq_str("\xe1\x9c\x81", buffer);
180 
181 	/* Hamster face, translates to four characters */
182 	pos = buffer;
183 	ut_assert(!utf8_put(0x1f439, &pos));
184 	ut_asserteq(4, pos - buffer);
185 	ut_asserteq_str("\xf0\x9f\x90\xb9", buffer);
186 
187 	/* Illegal code */
188 	pos = buffer;
189 	ut_asserteq(-1, utf8_put(0xd888, &pos));
190 
191 	return 0;
192 }
193 UNICODE_TEST(unicode_test_utf8_put);
194 
195 static int unicode_test_utf8_utf16_strlen(struct unit_test_state *uts)
196 {
197 	ut_asserteq(6, utf8_utf16_strlen(d1));
198 	ut_asserteq(8, utf8_utf16_strlen(d2));
199 	ut_asserteq(3, utf8_utf16_strlen(d3));
200 	ut_asserteq(6, utf8_utf16_strlen(d4));
201 
202 	/* illegal utf-8 sequences */
203 	ut_asserteq(4, utf8_utf16_strlen(j1));
204 	ut_asserteq(4, utf8_utf16_strlen(j2));
205 	ut_asserteq(3, utf8_utf16_strlen(j3));
206 
207 	return 0;
208 }
209 UNICODE_TEST(unicode_test_utf8_utf16_strlen);
210 
211 static int unicode_test_utf8_utf16_strnlen(struct unit_test_state *uts)
212 {
213 	ut_asserteq(3, utf8_utf16_strnlen(d1, 3));
214 	ut_asserteq(6, utf8_utf16_strnlen(d1, 13));
215 	ut_asserteq(6, utf8_utf16_strnlen(d2, 6));
216 	ut_asserteq(2, utf8_utf16_strnlen(d3, 2));
217 	ut_asserteq(4, utf8_utf16_strnlen(d4, 2));
218 	ut_asserteq(6, utf8_utf16_strnlen(d4, 3));
219 
220 	/* illegal utf-8 sequences */
221 	ut_asserteq(4, utf8_utf16_strnlen(j1, 16));
222 	ut_asserteq(4, utf8_utf16_strnlen(j2, 16));
223 	ut_asserteq(3, utf8_utf16_strnlen(j3, 16));
224 
225 	return 0;
226 }
227 UNICODE_TEST(unicode_test_utf8_utf16_strnlen);
228 
229 /**
230  * ut_u16_strcmp() - Compare to u16 strings.
231  *
232  * @a1:		first string
233  * @a2:		second string
234  * @count:	number of u16 to compare
235  * Return:	-1 if a1 < a2, 0 if a1 == a2, 1 if a1 > a2
236  */
237 static int unicode_test_u16_strcmp(const u16 *a1, const u16 *a2, size_t count)
238 {
239 	for (; (*a1 || *a2) && count; ++a1, ++a2, --count) {
240 		if (*a1 < *a2)
241 			return -1;
242 		if (*a1 > *a2)
243 			return 1;
244 	}
245 	return 0;
246 }
247 
248 static int unicode_test_utf8_utf16_strcpy(struct unit_test_state *uts)
249 {
250 	u16 buf[16];
251 	u16 *pos;
252 
253 	pos = buf;
254 	utf8_utf16_strcpy(&pos, d1);
255 	ut_asserteq(6, pos - buf);
256 	ut_assert(!unicode_test_u16_strcmp(buf, c1, SIZE_MAX));
257 
258 	pos = buf;
259 	utf8_utf16_strcpy(&pos, d2);
260 	ut_asserteq(8, pos - buf);
261 	ut_assert(!unicode_test_u16_strcmp(buf, c2, SIZE_MAX));
262 
263 	pos = buf;
264 	utf8_utf16_strcpy(&pos, d3);
265 	ut_asserteq(3, pos - buf);
266 	ut_assert(!unicode_test_u16_strcmp(buf, c3, SIZE_MAX));
267 
268 	pos = buf;
269 	utf8_utf16_strcpy(&pos, d4);
270 	ut_asserteq(6, pos - buf);
271 	ut_assert(!unicode_test_u16_strcmp(buf, c4, SIZE_MAX));
272 
273 	/* Illegal utf-8 strings */
274 	pos = buf;
275 	utf8_utf16_strcpy(&pos, j1);
276 	ut_asserteq(4, pos - buf);
277 	ut_assert(!unicode_test_u16_strcmp(buf, L"j1?l", SIZE_MAX));
278 
279 	pos = buf;
280 	utf8_utf16_strcpy(&pos, j2);
281 	ut_asserteq(4, pos - buf);
282 	ut_assert(!unicode_test_u16_strcmp(buf, L"j2?l", SIZE_MAX));
283 
284 	pos = buf;
285 	utf8_utf16_strcpy(&pos, j3);
286 	ut_asserteq(3, pos - buf);
287 	ut_assert(!unicode_test_u16_strcmp(buf, L"j3?", SIZE_MAX));
288 
289 	return 0;
290 }
291 UNICODE_TEST(unicode_test_utf8_utf16_strcpy);
292 
293 static int unicode_test_utf8_utf16_strncpy(struct unit_test_state *uts)
294 {
295 	u16 buf[16];
296 	u16 *pos;
297 
298 	pos = buf;
299 	memset(buf, 0, sizeof(buf));
300 	utf8_utf16_strncpy(&pos, d1, 4);
301 	ut_asserteq(4, pos - buf);
302 	ut_assert(!buf[4]);
303 	ut_assert(!unicode_test_u16_strcmp(buf, c1, 4));
304 
305 	pos = buf;
306 	memset(buf, 0, sizeof(buf));
307 	utf8_utf16_strncpy(&pos, d2, 10);
308 	ut_asserteq(8, pos - buf);
309 	ut_assert(buf[4]);
310 	ut_assert(!unicode_test_u16_strcmp(buf, c2, SIZE_MAX));
311 
312 	pos = buf;
313 	memset(buf, 0, sizeof(buf));
314 	utf8_utf16_strncpy(&pos, d3, 2);
315 	ut_asserteq(2, pos - buf);
316 	ut_assert(!buf[2]);
317 	ut_assert(!unicode_test_u16_strcmp(buf, c3, 2));
318 
319 	pos = buf;
320 	memset(buf, 0, sizeof(buf));
321 	utf8_utf16_strncpy(&pos, d4, 2);
322 	ut_asserteq(4, pos - buf);
323 	ut_assert(!buf[4]);
324 	ut_assert(!unicode_test_u16_strcmp(buf, c4, 4));
325 
326 	pos = buf;
327 	memset(buf, 0, sizeof(buf));
328 	utf8_utf16_strncpy(&pos, d4, 10);
329 	ut_asserteq(6, pos - buf);
330 	ut_assert(buf[5]);
331 	ut_assert(!unicode_test_u16_strcmp(buf, c4, SIZE_MAX));
332 
333 	return 0;
334 }
335 UNICODE_TEST(unicode_test_utf8_utf16_strncpy);
336 
337 static int unicode_test_utf16_get(struct unit_test_state *uts)
338 {
339 	const u16 *s;
340 	s32 code;
341 	int i;
342 
343 	/* Check characters less than 0x10000 */
344 	s = c2;
345 	for (i = 0; i < 9; ++i) {
346 		code = utf16_get((const u16 **)&s);
347 		ut_asserteq(c2[i], code);
348 		if (!code)
349 			break;
350 	}
351 	ut_asserteq_ptr(c2 + 8, s);
352 
353 	/* Check character greater 0xffff */
354 	s = c4;
355 	code = utf16_get((const u16 **)&s);
356 	ut_asserteq(0x0001048d, code);
357 	ut_asserteq_ptr(c4 + 2, s);
358 
359 	return 0;
360 }
361 UNICODE_TEST(unicode_test_utf16_get);
362 
363 static int unicode_test_utf16_put(struct unit_test_state *uts)
364 {
365 	u16 buffer[4] = { 0, };
366 	u16 *pos;
367 
368 	/* Commercial at, translates to one word */
369 	pos = buffer;
370 	ut_assert(!utf16_put('@', &pos));
371 	ut_asserteq(1, pos - buffer);
372 	ut_asserteq((u16)'@', buffer[0]);
373 	ut_assert(!buffer[1]);
374 
375 	/* Hamster face, translates to two words */
376 	pos = buffer;
377 	ut_assert(!utf16_put(0x1f439, &pos));
378 	ut_asserteq(2, pos - buffer);
379 	ut_asserteq((u16)0xd83d, buffer[0]);
380 	ut_asserteq((u16)0xdc39, buffer[1]);
381 	ut_assert(!buffer[2]);
382 
383 	/* Illegal code */
384 	pos = buffer;
385 	ut_asserteq(-1, utf16_put(0xd888, &pos));
386 
387 	return 0;
388 }
389 UNICODE_TEST(unicode_test_utf16_put);
390 
391 static int unicode_test_utf16_strnlen(struct unit_test_state *uts)
392 {
393 	ut_asserteq(3, utf16_strnlen(c1, 3));
394 	ut_asserteq(6, utf16_strnlen(c1, 13));
395 	ut_asserteq(6, utf16_strnlen(c2, 6));
396 	ut_asserteq(2, utf16_strnlen(c3, 2));
397 	ut_asserteq(2, utf16_strnlen(c4, 2));
398 	ut_asserteq(3, utf16_strnlen(c4, 3));
399 
400 	/* illegal utf-16 word sequences */
401 	ut_asserteq(4, utf16_strnlen(i1, 16));
402 	ut_asserteq(4, utf16_strnlen(i2, 16));
403 	ut_asserteq(3, utf16_strnlen(i3, 16));
404 
405 	return 0;
406 }
407 UNICODE_TEST(unicode_test_utf16_strnlen);
408 
409 static int unicode_test_utf16_utf8_strlen(struct unit_test_state *uts)
410 {
411 	ut_asserteq(6, utf16_utf8_strlen(c1));
412 	ut_asserteq(9, utf16_utf8_strlen(c2));
413 	ut_asserteq(9, utf16_utf8_strlen(c3));
414 	ut_asserteq(12, utf16_utf8_strlen(c4));
415 
416 	/* illegal utf-16 word sequences */
417 	ut_asserteq(4, utf16_utf8_strlen(i1));
418 	ut_asserteq(4, utf16_utf8_strlen(i2));
419 	ut_asserteq(3, utf16_utf8_strlen(i3));
420 
421 	return 0;
422 }
423 UNICODE_TEST(unicode_test_utf16_utf8_strlen);
424 
425 static int unicode_test_utf16_utf8_strnlen(struct unit_test_state *uts)
426 {
427 	ut_asserteq(3, utf16_utf8_strnlen(c1, 3));
428 	ut_asserteq(6, utf16_utf8_strnlen(c1, 13));
429 	ut_asserteq(7, utf16_utf8_strnlen(c2, 6));
430 	ut_asserteq(6, utf16_utf8_strnlen(c3, 2));
431 	ut_asserteq(8, utf16_utf8_strnlen(c4, 2));
432 	ut_asserteq(12, utf16_utf8_strnlen(c4, 3));
433 	return 0;
434 }
435 UNICODE_TEST(unicode_test_utf16_utf8_strnlen);
436 
437 static int unicode_test_utf16_utf8_strcpy(struct unit_test_state *uts)
438 {
439 	char buf[16];
440 	char *pos;
441 
442 	pos = buf;
443 	utf16_utf8_strcpy(&pos, c1);
444 	ut_asserteq(6, pos - buf);
445 	ut_asserteq_str(d1, buf);
446 
447 	pos = buf;
448 	utf16_utf8_strcpy(&pos, c2);
449 	ut_asserteq(9, pos - buf);
450 	ut_asserteq_str(d2, buf);
451 
452 	pos = buf;
453 	utf16_utf8_strcpy(&pos, c3);
454 	ut_asserteq(9, pos - buf);
455 	ut_asserteq_str(d3, buf);
456 
457 	pos = buf;
458 	utf16_utf8_strcpy(&pos, c4);
459 	ut_asserteq(12, pos - buf);
460 	ut_asserteq_str(d4, buf);
461 
462 	/* Illegal utf-16 strings */
463 	pos = buf;
464 	utf16_utf8_strcpy(&pos, i1);
465 	ut_asserteq(4, pos - buf);
466 	ut_asserteq_str("i1?l", buf);
467 
468 	pos = buf;
469 	utf16_utf8_strcpy(&pos, i2);
470 	ut_asserteq(4, pos - buf);
471 	ut_asserteq_str("i2?l", buf);
472 
473 	pos = buf;
474 	utf16_utf8_strcpy(&pos, i3);
475 	ut_asserteq(3, pos - buf);
476 	ut_asserteq_str("i3?", buf);
477 
478 	return 0;
479 }
480 UNICODE_TEST(unicode_test_utf16_utf8_strcpy);
481 
482 static int unicode_test_utf16_utf8_strncpy(struct unit_test_state *uts)
483 {
484 	char buf[16];
485 	char *pos;
486 
487 	pos = buf;
488 	memset(buf, 0, sizeof(buf));
489 	utf16_utf8_strncpy(&pos, c1, 4);
490 	ut_asserteq(4, pos - buf);
491 	ut_assert(!buf[4]);
492 	ut_assert(!strncmp(buf, d1, 4));
493 
494 	pos = buf;
495 	memset(buf, 0, sizeof(buf));
496 	utf16_utf8_strncpy(&pos, c2, 10);
497 	ut_asserteq(9, pos - buf);
498 	ut_assert(buf[4]);
499 	ut_assert(!strncmp(buf, d2, SIZE_MAX));
500 
501 	pos = buf;
502 	memset(buf, 0, sizeof(buf));
503 	utf16_utf8_strncpy(&pos, c3, 2);
504 	ut_asserteq(6, pos - buf);
505 	ut_assert(!buf[6]);
506 	ut_assert(!strncmp(buf, d3, 6));
507 
508 	pos = buf;
509 	memset(buf, 0, sizeof(buf));
510 	utf16_utf8_strncpy(&pos, c4, 2);
511 	ut_asserteq(8, pos - buf);
512 	ut_assert(!buf[8]);
513 	ut_assert(!strncmp(buf, d4, 8));
514 
515 	pos = buf;
516 	memset(buf, 0, sizeof(buf));
517 	utf16_utf8_strncpy(&pos, c4, 10);
518 	ut_asserteq(12, pos - buf);
519 	ut_assert(buf[5]);
520 	ut_assert(!strncmp(buf, d4, SIZE_MAX));
521 
522 	return 0;
523 }
524 UNICODE_TEST(unicode_test_utf16_utf8_strncpy);
525 
526 static int unicode_test_utf_to_lower(struct unit_test_state *uts)
527 {
528 	ut_asserteq('@', utf_to_lower('@'));
529 	ut_asserteq('a', utf_to_lower('A'));
530 	ut_asserteq('z', utf_to_lower('Z'));
531 	ut_asserteq('[', utf_to_lower('['));
532 	ut_asserteq('m', utf_to_lower('m'));
533 	/* Latin letter O with diaresis (umlaut) */
534 	ut_asserteq(0x00f6, utf_to_lower(0x00d6));
535 #ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
536 	/* Cyrillic letter I*/
537 	ut_asserteq(0x0438, utf_to_lower(0x0418));
538 #endif
539 	return 0;
540 }
541 UNICODE_TEST(unicode_test_utf_to_lower);
542 
543 static int unicode_test_utf_to_upper(struct unit_test_state *uts)
544 {
545 	ut_asserteq('`', utf_to_upper('`'));
546 	ut_asserteq('A', utf_to_upper('a'));
547 	ut_asserteq('Z', utf_to_upper('z'));
548 	ut_asserteq('{', utf_to_upper('{'));
549 	ut_asserteq('M', utf_to_upper('M'));
550 	/* Latin letter O with diaresis (umlaut) */
551 	ut_asserteq(0x00d6, utf_to_upper(0x00f6));
552 #ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
553 	/* Cyrillic letter I */
554 	ut_asserteq(0x0418, utf_to_upper(0x0438));
555 #endif
556 	return 0;
557 }
558 UNICODE_TEST(unicode_test_utf_to_upper);
559 
560 int do_ut_unicode(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
561 {
562 	struct unit_test *tests = ll_entry_start(struct unit_test, unicode_test);
563 	const int n_ents = ll_entry_count(struct unit_test, unicode_test);
564 
565 	return cmd_ut_category("Unicode", tests, n_ents, argc, argv);
566 }
567