1 /* SPDX-License-Identifier: LGPL-2.1+ */
3 /* Parts of this file are based on the GLIB utf8 validation functions. The
4 * original license text follows. */
6 /* gutf8.c - Operations on UTF-8 strings.
8 * Copyright (C) 1999 Tom Tromey
9 * Copyright (C) 2000 Red Hat, Inc.
11 * This library is free software; you can redistribute it and/or
12 * modify it under the terms of the GNU Library General Public
13 * License as published by the Free Software Foundation; either
14 * version 2 of the License, or (at your option) any later version.
16 * This library is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 * Library General Public License for more details.
21 * You should have received a copy of the GNU Library General Public
22 * License along with this library; if not, write to the Free Software
23 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
31 #include "alloc-util.h"
33 #include "hexdecoct.h"
37 bool unichar_is_valid(char32_t ch
) {
39 if (ch
>= 0x110000) /* End of unicode space */
41 if ((ch
& 0xFFFFF800) == 0xD800) /* Reserved area for UTF-16 */
43 if ((ch
>= 0xFDD0) && (ch
<= 0xFDEF)) /* Reserved */
45 if ((ch
& 0xFFFE) == 0xFFFE) /* BOM (Byte Order Mark) */
51 static bool unichar_is_control(char32_t ch
) {
54 0 to ' '-1 is the C0 range.
55 DEL=0x7F, and DEL+1 to 0x9F is C1 range.
56 '\t' is in C0 range, but more or less harmless and commonly used.
59 return (ch
< ' ' && !IN_SET(ch
, '\t', '\n')) ||
60 (0x7F <= ch
&& ch
<= 0x9F);
63 /* count of characters used to encode one unicode char */
64 static int utf8_encoded_expected_len(const char *str
) {
69 c
= (unsigned char) str
[0];
72 if ((c
& 0xe0) == 0xc0)
74 if ((c
& 0xf0) == 0xe0)
76 if ((c
& 0xf8) == 0xf0)
78 if ((c
& 0xfc) == 0xf8)
80 if ((c
& 0xfe) == 0xfc)
86 /* decode one unicode char */
87 int utf8_encoded_to_unichar(const char *str
, char32_t
*ret_unichar
) {
93 len
= utf8_encoded_expected_len(str
);
97 *ret_unichar
= (char32_t
)str
[0];
100 unichar
= str
[0] & 0x1f;
103 unichar
= (char32_t
)str
[0] & 0x0f;
106 unichar
= (char32_t
)str
[0] & 0x07;
109 unichar
= (char32_t
)str
[0] & 0x03;
112 unichar
= (char32_t
)str
[0] & 0x01;
118 for (i
= 1; i
< len
; i
++) {
119 if (((char32_t
)str
[i
] & 0xc0) != 0x80)
122 unichar
|= (char32_t
)str
[i
] & 0x3f;
125 *ret_unichar
= unichar
;
130 bool utf8_is_printable_newline(const char* str
, size_t length
, bool newline
) {
135 for (p
= str
; length
;) {
139 encoded_len
= utf8_encoded_valid_unichar(p
);
140 if (encoded_len
< 0 ||
141 (size_t) encoded_len
> length
)
144 r
= utf8_encoded_to_unichar(p
, &val
);
146 unichar_is_control(val
) ||
147 (!newline
&& val
== '\n'))
150 length
-= encoded_len
;
157 const char *utf8_is_valid(const char *str
) {
162 for (p
= (const uint8_t*) str
; *p
; ) {
165 len
= utf8_encoded_valid_unichar((const char *)p
);
175 char *utf8_escape_invalid(const char *str
) {
180 p
= s
= malloc(strlen(str
) * 4 + 1);
187 len
= utf8_encoded_valid_unichar(str
);
189 s
= mempcpy(s
, str
, len
);
192 s
= stpcpy(s
, UTF8_REPLACEMENT_CHARACTER
);
202 char *utf8_escape_non_printable(const char *str
) {
207 p
= s
= malloc(strlen(str
) * 4 + 1);
214 len
= utf8_encoded_valid_unichar(str
);
216 if (utf8_is_printable(str
, len
)) {
217 s
= mempcpy(s
, str
, len
);
223 *(s
++) = hexchar((int) *str
>> 4);
224 *(s
++) = hexchar((int) *str
);
231 s
= stpcpy(s
, UTF8_REPLACEMENT_CHARACTER
);
241 char *ascii_is_valid(const char *str
) {
244 /* Check whether the string consists of valid ASCII bytes,
245 * i.e values between 0 and 127, inclusive. */
249 for (p
= str
; *p
; p
++)
250 if ((unsigned char) *p
>= 128)
256 char *ascii_is_valid_n(const char *str
, size_t len
) {
259 /* Very similar to ascii_is_valid(), but checks exactly len
260 * bytes and rejects any NULs in that range. */
264 for (i
= 0; i
< len
; i
++)
265 if ((unsigned char) str
[i
] >= 128 || str
[i
] == 0)
272 * utf8_encode_unichar() - Encode single UCS-4 character as UTF-8
273 * @out_utf8: output buffer of at least 4 bytes or NULL
274 * @g: UCS-4 character to encode
276 * This encodes a single UCS-4 character as UTF-8 and writes it into @out_utf8.
277 * The length of the character is returned. It is not zero-terminated! If the
278 * output buffer is NULL, only the length is returned.
280 * Returns: The length in bytes that the UTF-8 representation does or would
283 size_t utf8_encode_unichar(char *out_utf8
, char32_t g
) {
287 out_utf8
[0] = g
& 0x7f;
289 } else if (g
< (1 << 11)) {
291 out_utf8
[0] = 0xc0 | ((g
>> 6) & 0x1f);
292 out_utf8
[1] = 0x80 | (g
& 0x3f);
295 } else if (g
< (1 << 16)) {
297 out_utf8
[0] = 0xe0 | ((g
>> 12) & 0x0f);
298 out_utf8
[1] = 0x80 | ((g
>> 6) & 0x3f);
299 out_utf8
[2] = 0x80 | (g
& 0x3f);
302 } else if (g
< (1 << 21)) {
304 out_utf8
[0] = 0xf0 | ((g
>> 18) & 0x07);
305 out_utf8
[1] = 0x80 | ((g
>> 12) & 0x3f);
306 out_utf8
[2] = 0x80 | ((g
>> 6) & 0x3f);
307 out_utf8
[3] = 0x80 | (g
& 0x3f);
315 char *utf16_to_utf8(const void *s
, size_t length
) {
319 r
= new(char, (length
* 4 + 1) / 2 + 1);
326 while (f
< (const uint8_t*) s
+ length
) {
329 /* see RFC 2781 section 2.2 */
331 w1
= f
[1] << 8 | f
[0];
334 if (!utf16_is_surrogate(w1
)) {
335 t
+= utf8_encode_unichar(t
, w1
);
340 if (utf16_is_trailing_surrogate(w1
))
342 else if (f
>= (const uint8_t*) s
+ length
)
345 w2
= f
[1] << 8 | f
[0];
348 if (!utf16_is_trailing_surrogate(w2
)) {
353 t
+= utf8_encode_unichar(t
, utf16_surrogate_pair_to_unichar(w1
, w2
));
360 /* expected size used to encode one unicode char */
361 static int utf8_unichar_to_encoded_len(char32_t unichar
) {
367 if (unichar
< 0x10000)
369 if (unichar
< 0x200000)
371 if (unichar
< 0x4000000)
377 /* validate one encoded unicode char and return its length */
378 int utf8_encoded_valid_unichar(const char *str
) {
384 len
= utf8_encoded_expected_len(str
);
392 /* check if expected encoded chars are available */
393 for (i
= 0; i
< len
; i
++)
394 if ((str
[i
] & 0x80) != 0x80)
397 r
= utf8_encoded_to_unichar(str
, &unichar
);
401 /* check if encoded length matches encoded value */
402 if (utf8_unichar_to_encoded_len(unichar
) != len
)
405 /* check if value has valid range */
406 if (!unichar_is_valid(unichar
))
412 size_t utf8_n_codepoints(const char *str
) {
415 /* Returns the number of UTF-8 codepoints in this string, or (size_t) -1 if the string is not valid UTF-8. */
420 k
= utf8_encoded_valid_unichar(str
);
431 size_t utf8_console_width(const char *str
) {
434 /* Returns the approximate width a string will take on screen when printed on a character cell
435 * terminal/console. */
440 if (utf8_encoded_to_unichar(str
, &c
) < 0)
443 str
= utf8_next_char(str
);
445 n
+= unichar_iswide(c
) ? 2 : 1;