aboutsummaryrefslogtreecommitdiff
path: root/lib/efi_loader/efi_unicode_collation.c
blob: 36be798f64b8f8005c466594d284bdc86c9dc2d5 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
// SPDX-License-Identifier: GPL-2.0+
/*
 * EFI Unicode collation protocol
 *
 * Copyright (c) 2018 Heinrich Schuchardt <xypron.glpk@gmx.de>
 */

#include <common.h>
#include <charset.h>
#include <cp1250.h>
#include <cp437.h>
#include <efi_loader.h>

/* Characters that may not be used in FAT 8.3 file names */
static const char illegal[] = "+,<=>:;\"/\\|?*[]\x7f";

/*
 * EDK2 assumes codepage 1250 when creating FAT 8.3 file names.
 * Linux defaults to codepage 437 for FAT 8.3 file names.
 */
#if CONFIG_FAT_DEFAULT_CODEPAGE == 1250
/* Unicode code points for code page 1250 characters 0x80 - 0xff */
static const u16 codepage[] = CP1250;
#else
/* Unicode code points for code page 437 characters 0x80 - 0xff */
static const u16 *codepage = codepage_437;
#endif

/* GUID of the EFI_UNICODE_COLLATION_PROTOCOL2 */
const efi_guid_t efi_guid_unicode_collation_protocol2 =
	EFI_UNICODE_COLLATION_PROTOCOL2_GUID;

/**
 * efi_stri_coll() - compare utf-16 strings case-insenitively
 *
 * @this:	unicode collation protocol instance
 * @s1:		first string
 * @s2:		second string
 *
 * This function implements the StriColl() service of the
 * EFI_UNICODE_COLLATION_PROTOCOL2.
 *
 * See the Unified Extensible Firmware Interface (UEFI) specification for
 * details.
 *
 * Return:	0: s1 == s2, > 0: s1 > s2, < 0: s1 < s2
 */
static efi_intn_t EFIAPI efi_stri_coll(
		struct efi_unicode_collation_protocol *this, u16 *s1, u16 *s2)
{
	s32 c1, c2;
	efi_intn_t ret = 0;

	EFI_ENTRY("%p, %ls, %ls", this, s1, s2);
	for (; *s1 | *s2; ++s1, ++s2) {
		c1 = utf_to_upper(*s1);
		c2 = utf_to_upper(*s2);
		if (c1 < c2) {
			ret = -1;
			goto out;
		} else if (c1 > c2) {
			ret = 1;
			goto out;
		}
	}
out:
	EFI_EXIT(EFI_SUCCESS);
	return ret;
}

/**
 * next_lower() - get next codepoint converted to lower case
 *
 * @string:	pointer to u16 string, on return advanced by one codepoint
 * Return:	first codepoint of string converted to lower case
 */
static s32 next_lower(const u16 **string)
{
	return utf_to_lower(utf16_get(string));
}

/**
 * metai_match() - compare utf-16 string with a pattern string case-insenitively
 *
 * @string:	string to compare
 * @pattern:	pattern string
 *
 * The pattern string may use these:
 *	- * matches >= 0 characters
 *	- ? matches 1 character
 *	- [<char1><char2>...<charN>] match any character in the set
 *	- [<char1>-<char2>] matches any character in the range
 *
 * This function is called my efi_metai_match().
 *
 * For '*' pattern searches this function calls itself recursively.
 * Performance-wise this is suboptimal, especially for multiple '*' wildcards.
 * But it results in simple code.
 *
 * Return:	true if the string is matched.
 */
static bool metai_match(const u16 *string, const u16 *pattern)
{
	s32 first, s, p;

	for (; *string && *pattern;) {
		const u16 *string_old = string;

		s = next_lower(&string);
		p = next_lower(&pattern);

		switch (p) {
		case '*':
			/* Match 0 or more characters */
			for (;; s = next_lower(&string)) {
				if (metai_match(string_old, pattern))
					return true;
				if (!s)
					return false;
				string_old = string;
			}
		case '?':
			/* Match any one character */
			break;
		case '[':
			/* Match any character in the set */
			p = next_lower(&pattern);
			first = p;
			if (first == ']')
				/* Empty set */
				return false;
			p = next_lower(&pattern);
			if (p == '-') {
				/* Range */
				p = next_lower(&pattern);
				if (s < first || s > p)
					return false;
				p = next_lower(&pattern);
				if (p != ']')
					return false;
			} else {
				/* Set */
				bool hit = false;

				if (s == first)
					hit = true;
				for (; p && p != ']';
				     p = next_lower(&pattern)) {
					if (p == s)
						hit = true;
				}
				if (!hit || p != ']')
					return false;
			}
			break;
		default:
			/* Match one character */
			if (p != s)
				return false;
		}
	}
	if (!*pattern && !*string)
		return true;
	return false;
}

/**
 * efi_metai_match() - compare utf-16 string with a pattern string
 *		       case-insenitively
 *
 * @this:	unicode collation protocol instance
 * @string:	string to compare
 * @pattern:	pattern string
 *
 * The pattern string may use these:
 *	- * matches >= 0 characters
 *	- ? matches 1 character
 *	- [<char1><char2>...<charN>] match any character in the set
 *	- [<char1>-<char2>] matches any character in the range
 *
 * This function implements the MetaMatch() service of the
 * EFI_UNICODE_COLLATION_PROTOCOL2.
 *
 * Return:	true if the string is matched.
 */
static bool EFIAPI efi_metai_match(struct efi_unicode_collation_protocol *this,
				   const u16 *string, const u16 *pattern)
{
	bool ret;

	EFI_ENTRY("%p, %ls, %ls", this, string, pattern);
	ret =  metai_match(string, pattern);
	EFI_EXIT(EFI_SUCCESS);
	return ret;
}

/**
 * efi_str_lwr() - convert to lower case
 *
 * @this:	unicode collation protocol instance
 * @string:	string to convert
 *
 * The conversion is done in place. As long as upper and lower letters use the
 * same number of words this does not pose a problem.
 *
 * This function implements the StrLwr() service of the
 * EFI_UNICODE_COLLATION_PROTOCOL2.
 */
static void EFIAPI efi_str_lwr(struct efi_unicode_collation_protocol *this,
			       u16 *string)
{
	EFI_ENTRY("%p, %ls", this, string);
	for (; *string; ++string)
		*string = utf_to_lower(*string);
	EFI_EXIT(EFI_SUCCESS);
}

/**
 * efi_str_upr() - convert to upper case
 *
 * @this:	unicode collation protocol instance
 * @string:	string to convert
 *
 * The conversion is done in place. As long as upper and lower letters use the
 * same number of words this does not pose a problem.
 *
 * This function implements the StrUpr() service of the
 * EFI_UNICODE_COLLATION_PROTOCOL2.
 */
static void EFIAPI efi_str_upr(struct efi_unicode_collation_protocol *this,
			       u16 *string)
{
	EFI_ENTRY("%p, %ls", this, string);
	for (; *string; ++string)
		*string = utf_to_upper(*string);
	EFI_EXIT(EFI_SUCCESS);
}

/**
 * efi_fat_to_str() - convert an 8.3 file name from an OEM codepage to Unicode
 *
 * @this:	unicode collation protocol instance
 * @fat_size:	size of the string to convert
 * @fat:	string to convert
 * @string:	converted string
 *
 * This function implements the FatToStr() service of the
 * EFI_UNICODE_COLLATION_PROTOCOL2.
 */
static void EFIAPI efi_fat_to_str(struct efi_unicode_collation_protocol *this,
				  efi_uintn_t fat_size, char *fat, u16 *string)
{
	efi_uintn_t i;
	u16 c;

	EFI_ENTRY("%p, %zu, %s, %p", this, fat_size, fat, string);
	for (i = 0; i < fat_size; ++i) {
		c = (unsigned char)fat[i];
		if (c > 0x80)
			c = codepage[i - 0x80];
		string[i] = c;
		if (!c)
			break;
	}
	string[i] = 0;
	EFI_EXIT(EFI_SUCCESS);
}

/**
 * efi_fat_to_str() - convert a utf-16 string to legal characters for a FAT
 *                    file name in an OEM code page
 *
 * @this:	unicode collation protocol instance
 * @string:	Unicode string to convert
 * @fat_size:	size of the target buffer
 * @fat:	converted string
 *
 * This function implements the StrToFat() service of the
 * EFI_UNICODE_COLLATION_PROTOCOL2.
 *
 * Return:	true if an illegal character was substituted by '_'.
 */
static bool EFIAPI efi_str_to_fat(struct efi_unicode_collation_protocol *this,
				  const u16 *string, efi_uintn_t fat_size,
				  char *fat)
{
	efi_uintn_t i;
	s32 c;
	bool ret = false;

	EFI_ENTRY("%p, %ls, %zu, %p", this, string, fat_size, fat);
	for (i = 0; i < fat_size;) {
		c = utf16_get(&string);
		switch (c) {
		/* Ignore period and space */
		case '.':
		case ' ':
			continue;
		case 0:
			break;
		}
		c = utf_to_upper(c);
		if (utf_to_cp(&c, codepage) ||
		    (c && (c < 0x20 || strchr(illegal, c)))) {
			ret = true;
			c = '_';
		}

		fat[i] = c;
		if (!c)
			break;
		++i;
	}
	EFI_EXIT(EFI_SUCCESS);
	return ret;
}

const struct efi_unicode_collation_protocol efi_unicode_collation_protocol2 = {
	.stri_coll = efi_stri_coll,
	.metai_match = efi_metai_match,
	.str_lwr = efi_str_lwr,
	.str_upr = efi_str_upr,
	.fat_to_str = efi_fat_to_str,
	.str_to_fat = efi_str_to_fat,
	.supported_languages = "en",
};