[thirdparty/u-boot.git] / test / unicode_ut.c

// SPDX-License-Identifier: GPL-2.0+
/*
 * Unit tests for Unicode functions
 *
 * Copyright (c) 2018 Heinrich Schuchardt <xypron.glpk@gmx.de>
 */

#include <common.h>
#include <charset.h>
#include <command.h>
#include <errno.h>
#include <test/test.h>
#include <test/suites.h>
#include <test/ut.h>

/* Linker list entry for a Unicode test */
#define UNICODE_TEST(_name) UNIT_TEST(_name, 0, unicode_test)

/* Constants c1-c4 and d1-d4 encode the same letters */

/* Six characters translating to one utf-8 byte each. */
static const u16 c1[] = {0x55, 0x2d, 0x42, 0x6f, 0x6f, 0x74, 0x00};
/* One character translating to two utf-8 bytes */
static const u16 c2[] = {0x6b, 0x61, 0x66, 0x62, 0xe1, 0x74, 0x75, 0x72, 0x00};
/* Three characters translating to three utf-8 bytes each */
static const u16 c3[] = {0x6f5c, 0x6c34, 0x8266, 0x00};
/* Three letters translating to four utf-8 bytes each */
static const u16 c4[] = {0xd801, 0xdc8d, 0xd801, 0xdc96, 0xd801, 0xdc87,
			 0x0000};

/* Illegal utf-16 strings */
static const u16 i1[] = {0x69, 0x31, 0xdc87, 0x6c, 0x00};
static const u16 i2[] = {0x69, 0x32, 0xd801, 0xd801, 0x6c, 0x00};
static const u16 i3[] = {0x69, 0x33, 0xd801, 0x00};

/* Six characters translating to one utf-16 word each. */
static const char d1[] = {0x55, 0x2d, 0x42, 0x6f, 0x6f, 0x74, 0x00};
/* Eight characters translating to one utf-16 word each */
static const char d2[] = {0x6b, 0x61, 0x66, 0x62, 0xc3, 0xa1, 0x74, 0x75,
			  0x72, 0x00};
/* Three characters translating to one utf-16 word each */
static const char d3[] = {0xe6, 0xbd, 0x9c, 0xe6, 0xb0, 0xb4, 0xe8, 0x89,
			  0xa6, 0x00};
/* Three letters translating to two utf-16 word each */
static const char d4[] = {0xf0, 0x90, 0x92, 0x8d, 0xf0, 0x90, 0x92, 0x96,
			  0xf0, 0x90, 0x92, 0x87, 0x00};

/* Illegal utf-8 strings */
static const char j1[] = {0x6a, 0x31, 0xa1, 0x6c, 0x00};
static const char j2[] = {0x6a, 0x32, 0xc3, 0xc3, 0x6c, 0x00};
static const char j3[] = {0x6a, 0x33, 0xf0, 0x90, 0xf0, 0x00};

static int unicode_test_u16_strlen(struct unit_test_state *uts)
{
	ut_asserteq(6, u16_strlen(c1));
	ut_asserteq(8, u16_strlen(c2));
	ut_asserteq(3, u16_strlen(c3));
	ut_asserteq(6, u16_strlen(c4));
	return 0;
}
UNICODE_TEST(unicode_test_u16_strlen);

static int unicode_test_u16_strdup(struct unit_test_state *uts)
{
	u16 *copy = u16_strdup(c4);

	ut_assert(copy != c4);
	ut_assert(!memcmp(copy, c4, sizeof(c4)));
	free(copy);
	return 0;
}
UNICODE_TEST(unicode_test_u16_strdup);

static int unicode_test_u16_strcpy(struct unit_test_state *uts)
{
	u16 *r;
	u16 copy[10];

	r = u16_strcpy(copy, c1);
	ut_assert(r == copy);
	ut_assert(!memcmp(copy, c1, sizeof(c1)));
	return 0;
}
UNICODE_TEST(unicode_test_u16_strcpy);

/* U-Boot uses UTF-16 strings in the EFI context only. */
#if CONFIG_IS_ENABLED(EFI_LOADER) && !defined(API_BUILD)
static int unicode_test_string16(struct unit_test_state *uts)
{
	char buf[20];

	/* Test length and precision */
	memset(buf, 0xff, sizeof(buf));
	sprintf(buf, "%8.6ls", c2);
	ut_asserteq(' ', buf[1]);
	ut_assert(!strncmp(&buf[2], d2, 7));
	ut_assert(!buf[9]);

	memset(buf, 0xff, sizeof(buf));
	sprintf(buf, "%8.6ls", c4);
	ut_asserteq(' ', buf[4]);
	ut_assert(!strncmp(&buf[5], d4, 12));
	ut_assert(!buf[17]);

	memset(buf, 0xff, sizeof(buf));
	sprintf(buf, "%-8.2ls", c4);
	ut_asserteq(' ', buf[8]);
	ut_assert(!strncmp(buf, d4, 8));
	ut_assert(!buf[14]);

	/* Test handling of illegal utf-16 sequences */
	memset(buf, 0xff, sizeof(buf));
	sprintf(buf, "%ls", i1);
	ut_asserteq_str("i1?l", buf);

	memset(buf, 0xff, sizeof(buf));
	sprintf(buf, "%ls", i2);
	ut_asserteq_str("i2?l", buf);

	memset(buf, 0xff, sizeof(buf));
	sprintf(buf, "%ls", i3);
	ut_asserteq_str("i3?", buf);

	return 0;
}
UNICODE_TEST(unicode_test_string16);
#endif

static int unicode_test_utf8_get(struct unit_test_state *uts)
{
	const char *s;
	s32 code;
	int i;

	/* Check characters less than 0x800 */
	s = d2;
	for (i = 0; i < 8; ++i) {
		code = utf8_get((const char **)&s);
		/* c2 is the utf-8 encoding of d2 */
		ut_asserteq(c2[i], code);
		if (!code)
			break;
	}
	ut_asserteq_ptr(s, d2 + 9)

	/* Check characters less than 0x10000 */
	s = d3;
	for (i = 0; i < 4; ++i) {
		code = utf8_get((const char **)&s);
		/* c3 is the utf-8 encoding of d3 */
		ut_asserteq(c3[i], code);
		if (!code)
			break;
	}
	ut_asserteq_ptr(s, d3 + 9)

	/* Check character greater 0xffff */
	s = d4;
	code = utf8_get((const char **)&s);
	ut_asserteq(0x0001048d, code);
	ut_asserteq_ptr(s, d4 + 4);

	return 0;
}
UNICODE_TEST(unicode_test_utf8_get);

static int unicode_test_utf8_put(struct unit_test_state *uts)
{
	char buffer[8] = { 0, };
	char *pos;

	/* Commercial at, translates to one character */
	pos = buffer;
	ut_assert(!utf8_put('@', &pos))
	ut_asserteq(1, pos - buffer);
	ut_asserteq('@', buffer[0]);
	ut_assert(!buffer[1]);

	/* Latin letter G with acute, translates to two charactes */
	pos = buffer;
	ut_assert(!utf8_put(0x1f4, &pos));
	ut_asserteq(2, pos - buffer);
	ut_asserteq_str("\xc7\xb4", buffer);

	/* Tagalog letter i, translates to three characters */
	pos = buffer;
	ut_assert(!utf8_put(0x1701, &pos));
	ut_asserteq(3, pos - buffer);
	ut_asserteq_str("\xe1\x9c\x81", buffer);

	/* Hamster face, translates to four characters */
	pos = buffer;
	ut_assert(!utf8_put(0x1f439, &pos));
	ut_asserteq(4, pos - buffer);
	ut_asserteq_str("\xf0\x9f\x90\xb9", buffer);

	/* Illegal code */
	pos = buffer;
	ut_asserteq(-1, utf8_put(0xd888, &pos));

	return 0;
}
UNICODE_TEST(unicode_test_utf8_put);

static int unicode_test_utf8_utf16_strlen(struct unit_test_state *uts)
{
	ut_asserteq(6, utf8_utf16_strlen(d1));
	ut_asserteq(8, utf8_utf16_strlen(d2));
	ut_asserteq(3, utf8_utf16_strlen(d3));
	ut_asserteq(6, utf8_utf16_strlen(d4));

	/* illegal utf-8 sequences */
	ut_asserteq(4, utf8_utf16_strlen(j1));
	ut_asserteq(4, utf8_utf16_strlen(j2));
	ut_asserteq(3, utf8_utf16_strlen(j3));

	return 0;
}
UNICODE_TEST(unicode_test_utf8_utf16_strlen);

static int unicode_test_utf8_utf16_strnlen(struct unit_test_state *uts)
{
	ut_asserteq(3, utf8_utf16_strnlen(d1, 3));
	ut_asserteq(6, utf8_utf16_strnlen(d1, 13));
	ut_asserteq(6, utf8_utf16_strnlen(d2, 6));
	ut_asserteq(2, utf8_utf16_strnlen(d3, 2));
	ut_asserteq(4, utf8_utf16_strnlen(d4, 2));
	ut_asserteq(6, utf8_utf16_strnlen(d4, 3));

	/* illegal utf-8 sequences */
	ut_asserteq(4, utf8_utf16_strnlen(j1, 16));
	ut_asserteq(4, utf8_utf16_strnlen(j2, 16));
	ut_asserteq(3, utf8_utf16_strnlen(j3, 16));

	return 0;
}
UNICODE_TEST(unicode_test_utf8_utf16_strnlen);

/**
 * ut_u16_strcmp() - Compare to u16 strings.
 *
 * @a1:		first string
 * @a2:		second string
 * @count:	number of u16 to compare
 * Return:	-1 if a1 < a2, 0 if a1 == a2, 1 if a1 > a2
 */
static int unicode_test_u16_strcmp(const u16 *a1, const u16 *a2, size_t count)
{
	for (; (*a1 || *a2) && count; ++a1, ++a2, --count) {
		if (*a1 < *a2)
			return -1;
		if (*a1 > *a2)
			return 1;
	}
	return 0;
}

static int unicode_test_utf8_utf16_strcpy(struct unit_test_state *uts)
{
	u16 buf[16];
	u16 *pos;

	pos = buf;
	utf8_utf16_strcpy(&pos, d1);
	ut_asserteq(6, pos - buf);
	ut_assert(!unicode_test_u16_strcmp(buf, c1, SIZE_MAX));

	pos = buf;
	utf8_utf16_strcpy(&pos, d2);
	ut_asserteq(8, pos - buf);
	ut_assert(!unicode_test_u16_strcmp(buf, c2, SIZE_MAX));

	pos = buf;
	utf8_utf16_strcpy(&pos, d3);
	ut_asserteq(3, pos - buf);
	ut_assert(!unicode_test_u16_strcmp(buf, c3, SIZE_MAX));

	pos = buf;
	utf8_utf16_strcpy(&pos, d4);
	ut_asserteq(6, pos - buf);
	ut_assert(!unicode_test_u16_strcmp(buf, c4, SIZE_MAX));

	/* Illegal utf-8 strings */
	pos = buf;
	utf8_utf16_strcpy(&pos, j1);
	ut_asserteq(4, pos - buf);
	ut_assert(!unicode_test_u16_strcmp(buf, L"j1?l", SIZE_MAX));

	pos = buf;
	utf8_utf16_strcpy(&pos, j2);
	ut_asserteq(4, pos - buf);
	ut_assert(!unicode_test_u16_strcmp(buf, L"j2?l", SIZE_MAX));

	pos = buf;
	utf8_utf16_strcpy(&pos, j3);
	ut_asserteq(3, pos - buf);
	ut_assert(!unicode_test_u16_strcmp(buf, L"j3?", SIZE_MAX));

	return 0;
}
UNICODE_TEST(unicode_test_utf8_utf16_strcpy);

static int unicode_test_utf8_utf16_strncpy(struct unit_test_state *uts)
{
	u16 buf[16];
	u16 *pos;

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf8_utf16_strncpy(&pos, d1, 4);
	ut_asserteq(4, pos - buf);
	ut_assert(!buf[4]);
	ut_assert(!unicode_test_u16_strcmp(buf, c1, 4));

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf8_utf16_strncpy(&pos, d2, 10);
	ut_asserteq(8, pos - buf);
	ut_assert(buf[4]);
	ut_assert(!unicode_test_u16_strcmp(buf, c2, SIZE_MAX));

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf8_utf16_strncpy(&pos, d3, 2);
	ut_asserteq(2, pos - buf);
	ut_assert(!buf[2]);
	ut_assert(!unicode_test_u16_strcmp(buf, c3, 2));

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf8_utf16_strncpy(&pos, d4, 2);
	ut_asserteq(4, pos - buf);
	ut_assert(!buf[4]);
	ut_assert(!unicode_test_u16_strcmp(buf, c4, 4));

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf8_utf16_strncpy(&pos, d4, 10);
	ut_asserteq(6, pos - buf);
	ut_assert(buf[5]);
	ut_assert(!unicode_test_u16_strcmp(buf, c4, SIZE_MAX));

	return 0;
}
UNICODE_TEST(unicode_test_utf8_utf16_strncpy);

static int unicode_test_utf16_get(struct unit_test_state *uts)
{
	const u16 *s;
	s32 code;
	int i;

	/* Check characters less than 0x10000 */
	s = c2;
	for (i = 0; i < 9; ++i) {
		code = utf16_get((const u16 **)&s);
		ut_asserteq(c2[i], code);
		if (!code)
			break;
	}
	ut_asserteq_ptr(c2 + 8, s);

	/* Check character greater 0xffff */
	s = c4;
	code = utf16_get((const u16 **)&s);
	ut_asserteq(0x0001048d, code);
	ut_asserteq_ptr(c4 + 2, s);

	return 0;
}
UNICODE_TEST(unicode_test_utf16_get);

static int unicode_test_utf16_put(struct unit_test_state *uts)
{
	u16 buffer[4] = { 0, };
	u16 *pos;

	/* Commercial at, translates to one word */
	pos = buffer;
	ut_assert(!utf16_put('@', &pos));
	ut_asserteq(1, pos - buffer);
	ut_asserteq((u16)'@', buffer[0]);
	ut_assert(!buffer[1]);

	/* Hamster face, translates to two words */
	pos = buffer;
	ut_assert(!utf16_put(0x1f439, &pos));
	ut_asserteq(2, pos - buffer);
	ut_asserteq((u16)0xd83d, buffer[0]);
	ut_asserteq((u16)0xdc39, buffer[1]);
	ut_assert(!buffer[2]);

	/* Illegal code */
	pos = buffer;
	ut_asserteq(-1, utf16_put(0xd888, &pos));

	return 0;
}
UNICODE_TEST(unicode_test_utf16_put);

static int unicode_test_utf16_strnlen(struct unit_test_state *uts)
{
	ut_asserteq(3, utf16_strnlen(c1, 3));
	ut_asserteq(6, utf16_strnlen(c1, 13));
	ut_asserteq(6, utf16_strnlen(c2, 6));
	ut_asserteq(2, utf16_strnlen(c3, 2));
	ut_asserteq(2, utf16_strnlen(c4, 2));
	ut_asserteq(3, utf16_strnlen(c4, 3));

	/* illegal utf-16 word sequences */
	ut_asserteq(4, utf16_strnlen(i1, 16));
	ut_asserteq(4, utf16_strnlen(i2, 16));
	ut_asserteq(3, utf16_strnlen(i3, 16));

	return 0;
}
UNICODE_TEST(unicode_test_utf16_strnlen);

static int unicode_test_utf16_utf8_strlen(struct unit_test_state *uts)
{
	ut_asserteq(6, utf16_utf8_strlen(c1));
	ut_asserteq(9, utf16_utf8_strlen(c2));
	ut_asserteq(9, utf16_utf8_strlen(c3));
	ut_asserteq(12, utf16_utf8_strlen(c4));

	/* illegal utf-16 word sequences */
	ut_asserteq(4, utf16_utf8_strlen(i1));
	ut_asserteq(4, utf16_utf8_strlen(i2));
	ut_asserteq(3, utf16_utf8_strlen(i3));

	return 0;
}
UNICODE_TEST(unicode_test_utf16_utf8_strlen);

static int unicode_test_utf16_utf8_strnlen(struct unit_test_state *uts)
{
	ut_asserteq(3, utf16_utf8_strnlen(c1, 3));
	ut_asserteq(6, utf16_utf8_strnlen(c1, 13));
	ut_asserteq(7, utf16_utf8_strnlen(c2, 6));
	ut_asserteq(6, utf16_utf8_strnlen(c3, 2));
	ut_asserteq(8, utf16_utf8_strnlen(c4, 2));
	ut_asserteq(12, utf16_utf8_strnlen(c4, 3));
	return 0;
}
UNICODE_TEST(unicode_test_utf16_utf8_strnlen);

static int unicode_test_utf16_utf8_strcpy(struct unit_test_state *uts)
{
	char buf[16];
	char *pos;

	pos = buf;
	utf16_utf8_strcpy(&pos, c1);
	ut_asserteq(6, pos - buf);
	ut_asserteq_str(d1, buf);

	pos = buf;
	utf16_utf8_strcpy(&pos, c2);
	ut_asserteq(9, pos - buf);
	ut_asserteq_str(d2, buf);

	pos = buf;
	utf16_utf8_strcpy(&pos, c3);
	ut_asserteq(9, pos - buf);
	ut_asserteq_str(d3, buf);

	pos = buf;
	utf16_utf8_strcpy(&pos, c4);
	ut_asserteq(12, pos - buf);
	ut_asserteq_str(d4, buf);

	/* Illegal utf-16 strings */
	pos = buf;
	utf16_utf8_strcpy(&pos, i1);
	ut_asserteq(4, pos - buf);
	ut_asserteq_str("i1?l", buf);

	pos = buf;
	utf16_utf8_strcpy(&pos, i2);
	ut_asserteq(4, pos - buf);
	ut_asserteq_str("i2?l", buf);

	pos = buf;
	utf16_utf8_strcpy(&pos, i3);
	ut_asserteq(3, pos - buf);
	ut_asserteq_str("i3?", buf);

	return 0;
}
UNICODE_TEST(unicode_test_utf16_utf8_strcpy);

static int unicode_test_utf16_utf8_strncpy(struct unit_test_state *uts)
{
	char buf[16];
	char *pos;

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf16_utf8_strncpy(&pos, c1, 4);
	ut_asserteq(4, pos - buf);
	ut_assert(!buf[4]);
	ut_assert(!strncmp(buf, d1, 4));

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf16_utf8_strncpy(&pos, c2, 10);
	ut_asserteq(9, pos - buf);
	ut_assert(buf[4]);
	ut_assert(!strncmp(buf, d2, SIZE_MAX));

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf16_utf8_strncpy(&pos, c3, 2);
	ut_asserteq(6, pos - buf);
	ut_assert(!buf[6]);
	ut_assert(!strncmp(buf, d3, 6));

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf16_utf8_strncpy(&pos, c4, 2);
	ut_asserteq(8, pos - buf);
	ut_assert(!buf[8]);
	ut_assert(!strncmp(buf, d4, 8));

	pos = buf;
	memset(buf, 0, sizeof(buf));
	utf16_utf8_strncpy(&pos, c4, 10);
	ut_asserteq(12, pos - buf);
	ut_assert(buf[5]);
	ut_assert(!strncmp(buf, d4, SIZE_MAX));

	return 0;
}
UNICODE_TEST(unicode_test_utf16_utf8_strncpy);

static int unicode_test_utf_to_lower(struct unit_test_state *uts)
{
	ut_asserteq('@', utf_to_lower('@'));
	ut_asserteq('a', utf_to_lower('A'));
	ut_asserteq('z', utf_to_lower('Z'));
	ut_asserteq('[', utf_to_lower('['));
	ut_asserteq('m', utf_to_lower('m'));
	/* Latin letter O with diaresis (umlaut) */
	ut_asserteq(0x00f6, utf_to_lower(0x00d6));
#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
	/* Cyrillic letter I*/
	ut_asserteq(0x0438, utf_to_lower(0x0418));
#endif
	return 0;
}
UNICODE_TEST(unicode_test_utf_to_lower);

static int unicode_test_utf_to_upper(struct unit_test_state *uts)
{
	ut_asserteq('`', utf_to_upper('`'));
	ut_asserteq('A', utf_to_upper('a'));
	ut_asserteq('Z', utf_to_upper('z'));
	ut_asserteq('{', utf_to_upper('{'));
	ut_asserteq('M', utf_to_upper('M'));
	/* Latin letter O with diaresis (umlaut) */
	ut_asserteq(0x00d6, utf_to_upper(0x00f6));
#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
	/* Cyrillic letter I */
	ut_asserteq(0x0418, utf_to_upper(0x0438));
#endif
	return 0;
}
UNICODE_TEST(unicode_test_utf_to_upper);

int do_ut_unicode(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[])
{
	struct unit_test *tests = ll_entry_start(struct unit_test, unicode_test);
	const int n_ents = ll_entry_count(struct unit_test, unicode_test);

	return cmd_ut_category("Unicode", tests, n_ents, argc, argv);
}
Commit	Line	Data
f11a164b HS	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
02b31dce HS	53	static int unicode_test_u16_strlen(struct unit_test_state *uts)
	54	{
	55	ut_asserteq(6, u16_strlen(c1));
	56	ut_asserteq(8, u16_strlen(c2));
	57	ut_asserteq(3, u16_strlen(c3));
	58	ut_asserteq(6, u16_strlen(c4));
	59	return 0;
	60	}
	61	UNICODE_TEST(unicode_test_u16_strlen);
	62
bc19681a	63	static int unicode_test_u16_strdup(struct unit_test_state *uts)
abb93cb0 HS	64	{
	65	u16 *copy = u16_strdup(c4);
	66
	67	ut_assert(copy != c4);
	68	ut_assert(!memcmp(copy, c4, sizeof(c4)));
	69	free(copy);
	70	return 0;
	71	}
bc19681a	72	UNICODE_TEST(unicode_test_u16_strdup);
abb93cb0	73
bc19681a	74	static int unicode_test_u16_strcpy(struct unit_test_state *uts)
abb93cb0 HS	75	{
	76	u16 *r;
	77	u16 copy[10];
	78
	79	r = u16_strcpy(copy, c1);
	80	ut_assert(r == copy);
	81	ut_assert(!memcmp(copy, c1, sizeof(c1)));
	82	return 0;
	83	}
bc19681a	84	UNICODE_TEST(unicode_test_u16_strcpy);
abb93cb0	85
fbba2f67 HS	86	/* U-Boot uses UTF-16 strings in the EFI context only. */
fbba2f67 HS	87	#if CONFIG_IS_ENABLED(EFI_LOADER) && !defined(API_BUILD)
bc19681a	88	static int unicode_test_string16(struct unit_test_state *uts)
fbba2f67 HS	89	{
	90	char buf[20];
	91
	92	/* Test length and precision */
	93	memset(buf, 0xff, sizeof(buf));
	94	sprintf(buf, "%8.6ls", c2);
	95	ut_asserteq(' ', buf[1]);
	96	ut_assert(!strncmp(&buf[2], d2, 7));
	97	ut_assert(!buf[9]);
	98
	99	memset(buf, 0xff, sizeof(buf));
	100	sprintf(buf, "%8.6ls", c4);
	101	ut_asserteq(' ', buf[4]);
	102	ut_assert(!strncmp(&buf[5], d4, 12));
	103	ut_assert(!buf[17]);
	104
	105	memset(buf, 0xff, sizeof(buf));
	106	sprintf(buf, "%-8.2ls", c4);
	107	ut_asserteq(' ', buf[8]);
	108	ut_assert(!strncmp(buf, d4, 8));
	109	ut_assert(!buf[14]);
	110
	111	/* Test handling of illegal utf-16 sequences */
	112	memset(buf, 0xff, sizeof(buf));
	113	sprintf(buf, "%ls", i1);
	114	ut_asserteq_str("i1?l", buf);
	115
	116	memset(buf, 0xff, sizeof(buf));
	117	sprintf(buf, "%ls", i2);
	118	ut_asserteq_str("i2?l", buf);
	119
	120	memset(buf, 0xff, sizeof(buf));
	121	sprintf(buf, "%ls", i3);
	122	ut_asserteq_str("i3?", buf);
	123
	124	return 0;
	125	}
bc19681a	126	UNICODE_TEST(unicode_test_string16);
fbba2f67 HS	127	#endif
fbba2f67 HS	128
bc19681a	129	static int unicode_test_utf8_get(struct unit_test_state *uts)
f11a164b HS	130	{
	131	const char *s;
	132	s32 code;
	133	int i;
	134
	135	/* Check characters less than 0x800 */
	136	s = d2;
	137	for (i = 0; i < 8; ++i) {
	138	code = utf8_get((const char **)&s);
	139	/* c2 is the utf-8 encoding of d2 */
	140	ut_asserteq(c2[i], code);
	141	if (!code)
	142	break;
	143	}
	144	ut_asserteq_ptr(s, d2 + 9)
	145
	146	/* Check characters less than 0x10000 */
	147	s = d3;
	148	for (i = 0; i < 4; ++i) {
	149	code = utf8_get((const char **)&s);
	150	/* c3 is the utf-8 encoding of d3 */
	151	ut_asserteq(c3[i], code);
	152	if (!code)
	153	break;
	154	}
	155	ut_asserteq_ptr(s, d3 + 9)
	156
	157	/* Check character greater 0xffff */
	158	s = d4;
	159	code = utf8_get((const char **)&s);
	160	ut_asserteq(0x0001048d, code);
	161	ut_asserteq_ptr(s, d4 + 4);
	162
	163	return 0;
	164	}
bc19681a	165	UNICODE_TEST(unicode_test_utf8_get);
f11a164b	166
bc19681a	167	static int unicode_test_utf8_put(struct unit_test_state *uts)
f11a164b HS	168	{
	169	char buffer[8] = { 0, };
	170	char *pos;
	171
	172	/* Commercial at, translates to one character */
	173	pos = buffer;
	174	ut_assert(!utf8_put('@', &pos))
	175	ut_asserteq(1, pos - buffer);
	176	ut_asserteq('@', buffer[0]);
	177	ut_assert(!buffer[1]);
	178
	179	/* Latin letter G with acute, translates to two charactes */
	180	pos = buffer;
	181	ut_assert(!utf8_put(0x1f4, &pos));
	182	ut_asserteq(2, pos - buffer);
	183	ut_asserteq_str("\xc7\xb4", buffer);
	184
	185	/* Tagalog letter i, translates to three characters */
	186	pos = buffer;
	187	ut_assert(!utf8_put(0x1701, &pos));
	188	ut_asserteq(3, pos - buffer);
	189	ut_asserteq_str("\xe1\x9c\x81", buffer);
	190
	191	/* Hamster face, translates to four characters */
	192	pos = buffer;
	193	ut_assert(!utf8_put(0x1f439, &pos));
	194	ut_asserteq(4, pos - buffer);
	195	ut_asserteq_str("\xf0\x9f\x90\xb9", buffer);
	196
	197	/* Illegal code */
	198	pos = buffer;
	199	ut_asserteq(-1, utf8_put(0xd888, &pos));
	200
	201	return 0;
	202	}
bc19681a	203	UNICODE_TEST(unicode_test_utf8_put);
f11a164b	204
bc19681a	205	static int unicode_test_utf8_utf16_strlen(struct unit_test_state *uts)
f11a164b HS	206	{
	207	ut_asserteq(6, utf8_utf16_strlen(d1));
	208	ut_asserteq(8, utf8_utf16_strlen(d2));
	209	ut_asserteq(3, utf8_utf16_strlen(d3));
	210	ut_asserteq(6, utf8_utf16_strlen(d4));
	211
	212	/* illegal utf-8 sequences */
	213	ut_asserteq(4, utf8_utf16_strlen(j1));
35cbb796	214	ut_asserteq(4, utf8_utf16_strlen(j2));
f11a164b HS	215	ut_asserteq(3, utf8_utf16_strlen(j3));
	216
	217	return 0;
	218	}
bc19681a	219	UNICODE_TEST(unicode_test_utf8_utf16_strlen);
f11a164b	220
bc19681a	221	static int unicode_test_utf8_utf16_strnlen(struct unit_test_state *uts)
f11a164b HS	222	{
	223	ut_asserteq(3, utf8_utf16_strnlen(d1, 3));
	224	ut_asserteq(6, utf8_utf16_strnlen(d1, 13));
	225	ut_asserteq(6, utf8_utf16_strnlen(d2, 6));
	226	ut_asserteq(2, utf8_utf16_strnlen(d3, 2));
	227	ut_asserteq(4, utf8_utf16_strnlen(d4, 2));
	228	ut_asserteq(6, utf8_utf16_strnlen(d4, 3));
	229
	230	/* illegal utf-8 sequences */
	231	ut_asserteq(4, utf8_utf16_strnlen(j1, 16));
35cbb796	232	ut_asserteq(4, utf8_utf16_strnlen(j2, 16));
f11a164b HS	233	ut_asserteq(3, utf8_utf16_strnlen(j3, 16));
	234
	235	return 0;
	236	}
bc19681a	237	UNICODE_TEST(unicode_test_utf8_utf16_strnlen);
f11a164b HS	238
	239	/**
	240	* ut_u16_strcmp() - Compare to u16 strings.
	241	*
	242	* @a1: first string
	243	* @a2: second string
	244	* @count: number of u16 to compare
	245	* Return: -1 if a1 < a2, 0 if a1 == a2, 1 if a1 > a2
	246	*/
bc19681a	247	static int unicode_test_u16_strcmp(const u16 a1, const u16 a2, size_t count)
f11a164b HS	248	{
	249	for (; (a1 \|\| a2) && count; ++a1, ++a2, --count) {
	250	if (a1 < a2)
	251	return -1;
	252	if (a1 > a2)
	253	return 1;
	254	}
	255	return 0;
	256	}
	257
bc19681a	258	static int unicode_test_utf8_utf16_strcpy(struct unit_test_state *uts)
f11a164b HS	259	{
	260	u16 buf[16];
	261	u16 *pos;
	262
	263	pos = buf;
	264	utf8_utf16_strcpy(&pos, d1);
	265	ut_asserteq(6, pos - buf);
bc19681a	266	ut_assert(!unicode_test_u16_strcmp(buf, c1, SIZE_MAX));
f11a164b HS	267
	268	pos = buf;
	269	utf8_utf16_strcpy(&pos, d2);
	270	ut_asserteq(8, pos - buf);
bc19681a	271	ut_assert(!unicode_test_u16_strcmp(buf, c2, SIZE_MAX));
f11a164b HS	272
	273	pos = buf;
	274	utf8_utf16_strcpy(&pos, d3);
	275	ut_asserteq(3, pos - buf);
bc19681a	276	ut_assert(!unicode_test_u16_strcmp(buf, c3, SIZE_MAX));
f11a164b HS	277
	278	pos = buf;
	279	utf8_utf16_strcpy(&pos, d4);
	280	ut_asserteq(6, pos - buf);
bc19681a	281	ut_assert(!unicode_test_u16_strcmp(buf, c4, SIZE_MAX));
f11a164b HS	282
	283	/* Illegal utf-8 strings */
	284	pos = buf;
	285	utf8_utf16_strcpy(&pos, j1);
	286	ut_asserteq(4, pos - buf);
bc19681a	287	ut_assert(!unicode_test_u16_strcmp(buf, L"j1?l", SIZE_MAX));
f11a164b HS	288
	289	pos = buf;
	290	utf8_utf16_strcpy(&pos, j2);
35cbb796	291	ut_asserteq(4, pos - buf);
bc19681a	292	ut_assert(!unicode_test_u16_strcmp(buf, L"j2?l", SIZE_MAX));
f11a164b HS	293
	294	pos = buf;
	295	utf8_utf16_strcpy(&pos, j3);
	296	ut_asserteq(3, pos - buf);
bc19681a	297	ut_assert(!unicode_test_u16_strcmp(buf, L"j3?", SIZE_MAX));
f11a164b HS	298
	299	return 0;
	300	}
bc19681a	301	UNICODE_TEST(unicode_test_utf8_utf16_strcpy);
f11a164b	302
bc19681a	303	static int unicode_test_utf8_utf16_strncpy(struct unit_test_state *uts)
f11a164b HS	304	{
	305	u16 buf[16];
	306	u16 *pos;
	307
	308	pos = buf;
	309	memset(buf, 0, sizeof(buf));
	310	utf8_utf16_strncpy(&pos, d1, 4);
	311	ut_asserteq(4, pos - buf);
	312	ut_assert(!buf[4]);
bc19681a	313	ut_assert(!unicode_test_u16_strcmp(buf, c1, 4));
f11a164b HS	314
	315	pos = buf;
	316	memset(buf, 0, sizeof(buf));
	317	utf8_utf16_strncpy(&pos, d2, 10);
	318	ut_asserteq(8, pos - buf);
	319	ut_assert(buf[4]);
bc19681a	320	ut_assert(!unicode_test_u16_strcmp(buf, c2, SIZE_MAX));
f11a164b HS	321
	322	pos = buf;
	323	memset(buf, 0, sizeof(buf));
	324	utf8_utf16_strncpy(&pos, d3, 2);
	325	ut_asserteq(2, pos - buf);
	326	ut_assert(!buf[2]);
bc19681a	327	ut_assert(!unicode_test_u16_strcmp(buf, c3, 2));
f11a164b HS	328
	329	pos = buf;
	330	memset(buf, 0, sizeof(buf));
	331	utf8_utf16_strncpy(&pos, d4, 2);
	332	ut_asserteq(4, pos - buf);
	333	ut_assert(!buf[4]);
bc19681a	334	ut_assert(!unicode_test_u16_strcmp(buf, c4, 4));
f11a164b HS	335
	336	pos = buf;
	337	memset(buf, 0, sizeof(buf));
	338	utf8_utf16_strncpy(&pos, d4, 10);
	339	ut_asserteq(6, pos - buf);
	340	ut_assert(buf[5]);
bc19681a	341	ut_assert(!unicode_test_u16_strcmp(buf, c4, SIZE_MAX));
f11a164b HS	342
	343	return 0;
	344	}
bc19681a	345	UNICODE_TEST(unicode_test_utf8_utf16_strncpy);
f11a164b	346
bc19681a	347	static int unicode_test_utf16_get(struct unit_test_state *uts)
f11a164b HS	348	{
	349	const u16 *s;
	350	s32 code;
	351	int i;
	352
	353	/* Check characters less than 0x10000 */
	354	s = c2;
	355	for (i = 0; i < 9; ++i) {
	356	code = utf16_get((const u16 **)&s);
	357	ut_asserteq(c2[i], code);
	358	if (!code)
	359	break;
	360	}
	361	ut_asserteq_ptr(c2 + 8, s);
	362
	363	/* Check character greater 0xffff */
	364	s = c4;
	365	code = utf16_get((const u16 **)&s);
	366	ut_asserteq(0x0001048d, code);
	367	ut_asserteq_ptr(c4 + 2, s);
	368
	369	return 0;
	370	}
bc19681a	371	UNICODE_TEST(unicode_test_utf16_get);
f11a164b	372
bc19681a	373	static int unicode_test_utf16_put(struct unit_test_state *uts)
f11a164b HS	374	{
	375	u16 buffer[4] = { 0, };
	376	u16 *pos;
	377
	378	/* Commercial at, translates to one word */
	379	pos = buffer;
	380	ut_assert(!utf16_put('@', &pos));
	381	ut_asserteq(1, pos - buffer);
	382	ut_asserteq((u16)'@', buffer[0]);
	383	ut_assert(!buffer[1]);
	384
	385	/* Hamster face, translates to two words */
	386	pos = buffer;
	387	ut_assert(!utf16_put(0x1f439, &pos));
	388	ut_asserteq(2, pos - buffer);
	389	ut_asserteq((u16)0xd83d, buffer[0]);
	390	ut_asserteq((u16)0xdc39, buffer[1]);
	391	ut_assert(!buffer[2]);
	392
	393	/* Illegal code */
	394	pos = buffer;
	395	ut_asserteq(-1, utf16_put(0xd888, &pos));
	396
	397	return 0;
	398	}
bc19681a	399	UNICODE_TEST(unicode_test_utf16_put);
f11a164b	400
bc19681a	401	static int unicode_test_utf16_strnlen(struct unit_test_state *uts)
f11a164b HS	402	{
	403	ut_asserteq(3, utf16_strnlen(c1, 3));
	404	ut_asserteq(6, utf16_strnlen(c1, 13));
	405	ut_asserteq(6, utf16_strnlen(c2, 6));
	406	ut_asserteq(2, utf16_strnlen(c3, 2));
	407	ut_asserteq(2, utf16_strnlen(c4, 2));
	408	ut_asserteq(3, utf16_strnlen(c4, 3));
	409
	410	/* illegal utf-16 word sequences */
	411	ut_asserteq(4, utf16_strnlen(i1, 16));
	412	ut_asserteq(4, utf16_strnlen(i2, 16));
	413	ut_asserteq(3, utf16_strnlen(i3, 16));
	414
	415	return 0;
	416	}
bc19681a	417	UNICODE_TEST(unicode_test_utf16_strnlen);
f11a164b	418
bc19681a	419	static int unicode_test_utf16_utf8_strlen(struct unit_test_state *uts)
f11a164b HS	420	{
	421	ut_asserteq(6, utf16_utf8_strlen(c1));
	422	ut_asserteq(9, utf16_utf8_strlen(c2));
	423	ut_asserteq(9, utf16_utf8_strlen(c3));
	424	ut_asserteq(12, utf16_utf8_strlen(c4));
	425
	426	/* illegal utf-16 word sequences */
	427	ut_asserteq(4, utf16_utf8_strlen(i1));
	428	ut_asserteq(4, utf16_utf8_strlen(i2));
	429	ut_asserteq(3, utf16_utf8_strlen(i3));
	430
	431	return 0;
	432	}
bc19681a	433	UNICODE_TEST(unicode_test_utf16_utf8_strlen);
f11a164b	434
bc19681a	435	static int unicode_test_utf16_utf8_strnlen(struct unit_test_state *uts)
f11a164b HS	436	{
	437	ut_asserteq(3, utf16_utf8_strnlen(c1, 3));
	438	ut_asserteq(6, utf16_utf8_strnlen(c1, 13));
	439	ut_asserteq(7, utf16_utf8_strnlen(c2, 6));
	440	ut_asserteq(6, utf16_utf8_strnlen(c3, 2));
	441	ut_asserteq(8, utf16_utf8_strnlen(c4, 2));
	442	ut_asserteq(12, utf16_utf8_strnlen(c4, 3));
	443	return 0;
	444	}
bc19681a	445	UNICODE_TEST(unicode_test_utf16_utf8_strnlen);
f11a164b	446
bc19681a	447	static int unicode_test_utf16_utf8_strcpy(struct unit_test_state *uts)
f11a164b HS	448	{
	449	char buf[16];
	450	char *pos;
	451
	452	pos = buf;
	453	utf16_utf8_strcpy(&pos, c1);
	454	ut_asserteq(6, pos - buf);
	455	ut_asserteq_str(d1, buf);
	456
	457	pos = buf;
	458	utf16_utf8_strcpy(&pos, c2);
	459	ut_asserteq(9, pos - buf);
	460	ut_asserteq_str(d2, buf);
	461
	462	pos = buf;
	463	utf16_utf8_strcpy(&pos, c3);
	464	ut_asserteq(9, pos - buf);
	465	ut_asserteq_str(d3, buf);
	466
	467	pos = buf;
	468	utf16_utf8_strcpy(&pos, c4);
	469	ut_asserteq(12, pos - buf);
	470	ut_asserteq_str(d4, buf);
	471
	472	/* Illegal utf-16 strings */
	473	pos = buf;
	474	utf16_utf8_strcpy(&pos, i1);
	475	ut_asserteq(4, pos - buf);
	476	ut_asserteq_str("i1?l", buf);
	477
	478	pos = buf;
	479	utf16_utf8_strcpy(&pos, i2);
	480	ut_asserteq(4, pos - buf);
	481	ut_asserteq_str("i2?l", buf);
	482
	483	pos = buf;
	484	utf16_utf8_strcpy(&pos, i3);
	485	ut_asserteq(3, pos - buf);
	486	ut_asserteq_str("i3?", buf);
	487
	488	return 0;
	489	}
bc19681a	490	UNICODE_TEST(unicode_test_utf16_utf8_strcpy);
f11a164b	491
bc19681a	492	static int unicode_test_utf16_utf8_strncpy(struct unit_test_state *uts)
f11a164b HS	493	{
	494	char buf[16];
	495	char *pos;
	496
	497	pos = buf;
	498	memset(buf, 0, sizeof(buf));
	499	utf16_utf8_strncpy(&pos, c1, 4);
	500	ut_asserteq(4, pos - buf);
	501	ut_assert(!buf[4]);
	502	ut_assert(!strncmp(buf, d1, 4));
	503
	504	pos = buf;
	505	memset(buf, 0, sizeof(buf));
	506	utf16_utf8_strncpy(&pos, c2, 10);
	507	ut_asserteq(9, pos - buf);
	508	ut_assert(buf[4]);
	509	ut_assert(!strncmp(buf, d2, SIZE_MAX));
	510
	511	pos = buf;
	512	memset(buf, 0, sizeof(buf));
	513	utf16_utf8_strncpy(&pos, c3, 2);
	514	ut_asserteq(6, pos - buf);
	515	ut_assert(!buf[6]);
	516	ut_assert(!strncmp(buf, d3, 6));
	517
	518	pos = buf;
	519	memset(buf, 0, sizeof(buf));
	520	utf16_utf8_strncpy(&pos, c4, 2);
	521	ut_asserteq(8, pos - buf);
	522	ut_assert(!buf[8]);
	523	ut_assert(!strncmp(buf, d4, 8));
	524
	525	pos = buf;
	526	memset(buf, 0, sizeof(buf));
	527	utf16_utf8_strncpy(&pos, c4, 10);
	528	ut_asserteq(12, pos - buf);
	529	ut_assert(buf[5]);
	530	ut_assert(!strncmp(buf, d4, SIZE_MAX));
	531
	532	return 0;
	533	}
bc19681a	534	UNICODE_TEST(unicode_test_utf16_utf8_strncpy);
f11a164b	535
bc19681a	536	static int unicode_test_utf_to_lower(struct unit_test_state *uts)
1a1012a1 HS	537	{
	538	ut_asserteq('@', utf_to_lower('@'));
	539	ut_asserteq('a', utf_to_lower('A'));
	540	ut_asserteq('z', utf_to_lower('Z'));
	541	ut_asserteq('[', utf_to_lower('['));
	542	ut_asserteq('m', utf_to_lower('m'));
	543	/* Latin letter O with diaresis (umlaut) */
	544	ut_asserteq(0x00f6, utf_to_lower(0x00d6));
	545	#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
	546	/* Cyrillic letter I*/
	547	ut_asserteq(0x0438, utf_to_lower(0x0418));
	548	#endif
	549	return 0;
	550	}
bc19681a	551	UNICODE_TEST(unicode_test_utf_to_lower);
1a1012a1	552
bc19681a	553	static int unicode_test_utf_to_upper(struct unit_test_state *uts)
1a1012a1 HS	554	{
	555	ut_asserteq('`', utf_to_upper('`'));
	556	ut_asserteq('A', utf_to_upper('a'));
	557	ut_asserteq('Z', utf_to_upper('z'));
	558	ut_asserteq('{', utf_to_upper('{'));
	559	ut_asserteq('M', utf_to_upper('M'));
	560	/* Latin letter O with diaresis (umlaut) */
	561	ut_asserteq(0x00d6, utf_to_upper(0x00f6));
	562	#ifdef CONFIG_EFI_UNICODE_CAPITALIZATION
	563	/* Cyrillic letter I */
	564	ut_asserteq(0x0418, utf_to_upper(0x0438));
	565	#endif
	566	return 0;
	567	}
bc19681a	568	UNICODE_TEST(unicode_test_utf_to_upper);
1a1012a1	569
f11a164b HS	570	int do_ut_unicode(cmd_tbl_t cmdtp, int flag, int argc, char const argv[])
	571	{
	572	struct unit_test *tests = ll_entry_start(struct unit_test, unicode_test);
	573	const int n_ents = ll_entry_count(struct unit_test, unicode_test);
	574
	575	return cmd_ut_category("Unicode", tests, n_ents, argc, argv);
	576	}