]>
Commit | Line | Data |
---|---|---|
53e1b683 | 1 | /* SPDX-License-Identifier: LGPL-2.1+ */ |
7f110ff9 | 2 | |
036ae95a | 3 | /* Parts of this file are based on the GLIB utf8 validation functions. The |
7f110ff9 LP |
4 | * original license text follows. */ |
5 | ||
6 | /* gutf8.c - Operations on UTF-8 strings. | |
7 | * | |
8 | * Copyright (C) 1999 Tom Tromey | |
9 | * Copyright (C) 2000 Red Hat, Inc. | |
10 | * | |
11 | * This library is free software; you can redistribute it and/or | |
23757887 | 12 | * modify it under the terms of the GNU Library General Public |
7f110ff9 LP |
13 | * License as published by the Free Software Foundation; either |
14 | * version 2 of the License, or (at your option) any later version. | |
15 | * | |
16 | * This library is distributed in the hope that it will be useful, | |
17 | * but WITHOUT ANY WARRANTY; without even the implied warranty of | |
23757887 SK |
18 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
19 | * Library General Public License for more details. | |
7f110ff9 | 20 | * |
23757887 SK |
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 | |
7f110ff9 LP |
24 | */ |
25 | ||
26 | #include <errno.h> | |
7f110ff9 | 27 | #include <stdbool.h> |
cf0fbc49 TA |
28 | #include <stdlib.h> |
29 | #include <string.h> | |
7f110ff9 | 30 | |
b5efdb8a | 31 | #include "alloc-util.h" |
3f536d5b | 32 | #include "gunicode.h" |
e4e73a63 | 33 | #include "hexdecoct.h" |
11c3a366 | 34 | #include "macro.h" |
4c6d5139 | 35 | #include "string-util.h" |
7f110ff9 LP |
36 | #include "utf8.h" |
37 | ||
c932fb71 | 38 | bool unichar_is_valid(char32_t ch) { |
7f110ff9 LP |
39 | |
40 | if (ch >= 0x110000) /* End of unicode space */ | |
41 | return false; | |
42 | if ((ch & 0xFFFFF800) == 0xD800) /* Reserved area for UTF-16 */ | |
43 | return false; | |
44 | if ((ch >= 0xFDD0) && (ch <= 0xFDEF)) /* Reserved */ | |
45 | return false; | |
46 | if ((ch & 0xFFFE) == 0xFFFE) /* BOM (Byte Order Mark) */ | |
47 | return false; | |
48 | ||
49 | return true; | |
50 | } | |
51 | ||
c932fb71 | 52 | static bool unichar_is_control(char32_t ch) { |
ba961854 ZJS |
53 | |
54 | /* | |
55 | 0 to ' '-1 is the C0 range. | |
56 | DEL=0x7F, and DEL+1 to 0x9F is C1 range. | |
57 | '\t' is in C0 range, but more or less harmless and commonly used. | |
58 | */ | |
59 | ||
4c701096 | 60 | return (ch < ' ' && !IN_SET(ch, '\t', '\n')) || |
ba961854 ZJS |
61 | (0x7F <= ch && ch <= 0x9F); |
62 | } | |
63 | ||
7991ac34 | 64 | /* count of characters used to encode one unicode char */ |
84319aa7 | 65 | static size_t utf8_encoded_expected_len(uint8_t c) { |
7991ac34 DR |
66 | if (c < 0x80) |
67 | return 1; | |
68 | if ((c & 0xe0) == 0xc0) | |
69 | return 2; | |
70 | if ((c & 0xf0) == 0xe0) | |
71 | return 3; | |
72 | if ((c & 0xf8) == 0xf0) | |
73 | return 4; | |
74 | if ((c & 0xfc) == 0xf8) | |
75 | return 5; | |
76 | if ((c & 0xfe) == 0xfc) | |
77 | return 6; | |
7e8185ef | 78 | |
7991ac34 DR |
79 | return 0; |
80 | } | |
ba961854 | 81 | |
7991ac34 | 82 | /* decode one unicode char */ |
c932fb71 SL |
83 | int utf8_encoded_to_unichar(const char *str, char32_t *ret_unichar) { |
84 | char32_t unichar; | |
7c421857 | 85 | size_t len, i; |
7e8185ef LP |
86 | |
87 | assert(str); | |
ba961854 | 88 | |
84319aa7 | 89 | len = utf8_encoded_expected_len(str[0]); |
7e8185ef | 90 | |
7991ac34 DR |
91 | switch (len) { |
92 | case 1: | |
c932fb71 SL |
93 | *ret_unichar = (char32_t)str[0]; |
94 | return 0; | |
7991ac34 DR |
95 | case 2: |
96 | unichar = str[0] & 0x1f; | |
97 | break; | |
98 | case 3: | |
c932fb71 | 99 | unichar = (char32_t)str[0] & 0x0f; |
7991ac34 DR |
100 | break; |
101 | case 4: | |
c932fb71 | 102 | unichar = (char32_t)str[0] & 0x07; |
7991ac34 DR |
103 | break; |
104 | case 5: | |
c932fb71 | 105 | unichar = (char32_t)str[0] & 0x03; |
7991ac34 DR |
106 | break; |
107 | case 6: | |
c932fb71 | 108 | unichar = (char32_t)str[0] & 0x01; |
7991ac34 DR |
109 | break; |
110 | default: | |
7e8185ef | 111 | return -EINVAL; |
ba961854 ZJS |
112 | } |
113 | ||
7991ac34 | 114 | for (i = 1; i < len; i++) { |
c932fb71 | 115 | if (((char32_t)str[i] & 0xc0) != 0x80) |
7e8185ef | 116 | return -EINVAL; |
7c421857 | 117 | |
7991ac34 | 118 | unichar <<= 6; |
c932fb71 | 119 | unichar |= (char32_t)str[i] & 0x3f; |
7991ac34 DR |
120 | } |
121 | ||
c932fb71 SL |
122 | *ret_unichar = unichar; |
123 | ||
124 | return 0; | |
ba961854 ZJS |
125 | } |
126 | ||
0ade5ffe | 127 | bool utf8_is_printable_newline(const char* str, size_t length, bool newline) { |
6ed62be0 | 128 | const char *p; |
7f110ff9 LP |
129 | |
130 | assert(str); | |
131 | ||
92e068b4 | 132 | for (p = str; length > 0;) { |
c932fb71 SL |
133 | int encoded_len, r; |
134 | char32_t val; | |
7e8185ef | 135 | |
92e068b4 ZJS |
136 | encoded_len = utf8_encoded_valid_unichar(p, length); |
137 | if (encoded_len < 0) | |
144b3d9e | 138 | return false; |
92e068b4 | 139 | assert(encoded_len > 0 && (size_t) encoded_len <= length); |
144b3d9e | 140 | |
c932fb71 SL |
141 | r = utf8_encoded_to_unichar(p, &val); |
142 | if (r < 0 || | |
f3ee6297 | 143 | unichar_is_control(val) || |
0ade5ffe | 144 | (!newline && val == '\n')) |
7991ac34 | 145 | return false; |
7f110ff9 | 146 | |
7991ac34 | 147 | length -= encoded_len; |
a7176505 | 148 | p += encoded_len; |
7f110ff9 LP |
149 | } |
150 | ||
7991ac34 | 151 | return true; |
7f110ff9 LP |
152 | } |
153 | ||
e71fb4b3 LP |
154 | char *utf8_is_valid(const char *str) { |
155 | const char *p; | |
7f110ff9 LP |
156 | |
157 | assert(str); | |
158 | ||
e71fb4b3 LP |
159 | p = str; |
160 | while (*p) { | |
faaa5728 LP |
161 | int len; |
162 | ||
92e068b4 | 163 | len = utf8_encoded_valid_unichar(p, (size_t) -1); |
7991ac34 DR |
164 | if (len < 0) |
165 | return NULL; | |
166 | ||
167 | p += len; | |
168 | } | |
7f110ff9 | 169 | |
e71fb4b3 | 170 | return (char*) str; |
7f110ff9 LP |
171 | } |
172 | ||
550a40ec ZJS |
173 | char *utf8_escape_invalid(const char *str) { |
174 | char *p, *s; | |
175 | ||
176 | assert(str); | |
177 | ||
178 | p = s = malloc(strlen(str) * 4 + 1); | |
179 | if (!p) | |
180 | return NULL; | |
181 | ||
182 | while (*str) { | |
183 | int len; | |
184 | ||
92e068b4 | 185 | len = utf8_encoded_valid_unichar(str, (size_t) -1); |
550a40ec ZJS |
186 | if (len > 0) { |
187 | s = mempcpy(s, str, len); | |
188 | str += len; | |
189 | } else { | |
3c6d3052 | 190 | s = stpcpy(s, UTF8_REPLACEMENT_CHARACTER); |
550a40ec ZJS |
191 | str += 1; |
192 | } | |
193 | } | |
7e8185ef | 194 | |
550a40ec | 195 | *s = '\0'; |
4c6d5139 | 196 | (void) str_realloc(&p); |
550a40ec ZJS |
197 | return p; |
198 | } | |
199 | ||
da88f542 ZJS |
200 | static int utf8_char_console_width(const char *str) { |
201 | char32_t c; | |
202 | int r; | |
203 | ||
204 | r = utf8_encoded_to_unichar(str, &c); | |
205 | if (r < 0) | |
206 | return r; | |
207 | ||
208 | /* TODO: we should detect combining characters */ | |
209 | ||
210 | return unichar_iswide(c) ? 2 : 1; | |
211 | } | |
212 | ||
213 | char *utf8_escape_non_printable_full(const char *str, size_t console_width) { | |
214 | char *p, *s, *prev_s; | |
215 | size_t n = 0; /* estimated print width */ | |
fec84576 WC |
216 | |
217 | assert(str); | |
218 | ||
da88f542 ZJS |
219 | if (console_width == 0) |
220 | return strdup(""); | |
221 | ||
222 | p = s = prev_s = malloc(strlen(str) * 4 + 1); | |
fec84576 WC |
223 | if (!p) |
224 | return NULL; | |
225 | ||
da88f542 | 226 | for (;;) { |
fec84576 | 227 | int len; |
da88f542 ZJS |
228 | char *saved_s = s; |
229 | ||
230 | if (!*str) /* done! */ | |
231 | goto finish; | |
fec84576 | 232 | |
92e068b4 | 233 | len = utf8_encoded_valid_unichar(str, (size_t) -1); |
fec84576 WC |
234 | if (len > 0) { |
235 | if (utf8_is_printable(str, len)) { | |
da88f542 ZJS |
236 | int w; |
237 | ||
238 | w = utf8_char_console_width(str); | |
239 | assert(w >= 0); | |
240 | if (n + w > console_width) | |
241 | goto truncation; | |
242 | ||
fec84576 WC |
243 | s = mempcpy(s, str, len); |
244 | str += len; | |
da88f542 ZJS |
245 | n += w; |
246 | ||
fec84576 | 247 | } else { |
da88f542 ZJS |
248 | for (; len > 0; len--) { |
249 | if (n + 4 > console_width) | |
250 | goto truncation; | |
251 | ||
fec84576 WC |
252 | *(s++) = '\\'; |
253 | *(s++) = 'x'; | |
254 | *(s++) = hexchar((int) *str >> 4); | |
255 | *(s++) = hexchar((int) *str); | |
fec84576 | 256 | |
3c6d3052 | 257 | str += 1; |
da88f542 | 258 | n += 4; |
3c6d3052 | 259 | } |
fec84576 WC |
260 | } |
261 | } else { | |
da88f542 ZJS |
262 | if (n + 1 > console_width) |
263 | goto truncation; | |
264 | ||
265 | s = mempcpy(s, UTF8_REPLACEMENT_CHARACTER, strlen(UTF8_REPLACEMENT_CHARACTER)); | |
fec84576 | 266 | str += 1; |
da88f542 | 267 | n += 1; |
fec84576 | 268 | } |
da88f542 ZJS |
269 | |
270 | prev_s = saved_s; | |
fec84576 WC |
271 | } |
272 | ||
da88f542 ZJS |
273 | truncation: |
274 | /* Try to go back one if we don't have enough space for the ellipsis */ | |
275 | if (n + 1 >= console_width) | |
276 | s = prev_s; | |
277 | ||
278 | s = mempcpy(s, "…", strlen("…")); | |
fec84576 | 279 | |
da88f542 ZJS |
280 | finish: |
281 | *s = '\0'; | |
4c6d5139 | 282 | (void) str_realloc(&p); |
fec84576 WC |
283 | return p; |
284 | } | |
285 | ||
7f110ff9 LP |
286 | char *ascii_is_valid(const char *str) { |
287 | const char *p; | |
288 | ||
294a3121 ZJS |
289 | /* Check whether the string consists of valid ASCII bytes, |
290 | * i.e values between 0 and 127, inclusive. */ | |
291 | ||
7f110ff9 LP |
292 | assert(str); |
293 | ||
294 | for (p = str; *p; p++) | |
295 | if ((unsigned char) *p >= 128) | |
296 | return NULL; | |
297 | ||
298 | return (char*) str; | |
299 | } | |
300 | ||
294a3121 ZJS |
301 | char *ascii_is_valid_n(const char *str, size_t len) { |
302 | size_t i; | |
303 | ||
304 | /* Very similar to ascii_is_valid(), but checks exactly len | |
305 | * bytes and rejects any NULs in that range. */ | |
306 | ||
307 | assert(str); | |
308 | ||
309 | for (i = 0; i < len; i++) | |
310 | if ((unsigned char) str[i] >= 128 || str[i] == 0) | |
311 | return NULL; | |
312 | ||
313 | return (char*) str; | |
314 | } | |
315 | ||
2bb4c7e3 TG |
316 | /** |
317 | * utf8_encode_unichar() - Encode single UCS-4 character as UTF-8 | |
318 | * @out_utf8: output buffer of at least 4 bytes or NULL | |
319 | * @g: UCS-4 character to encode | |
320 | * | |
321 | * This encodes a single UCS-4 character as UTF-8 and writes it into @out_utf8. | |
322 | * The length of the character is returned. It is not zero-terminated! If the | |
323 | * output buffer is NULL, only the length is returned. | |
324 | * | |
325 | * Returns: The length in bytes that the UTF-8 representation does or would | |
326 | * occupy. | |
327 | */ | |
c932fb71 | 328 | size_t utf8_encode_unichar(char *out_utf8, char32_t g) { |
f3ee6297 | 329 | |
2bb4c7e3 TG |
330 | if (g < (1 << 7)) { |
331 | if (out_utf8) | |
332 | out_utf8[0] = g & 0x7f; | |
e7eebcfc | 333 | return 1; |
2bb4c7e3 TG |
334 | } else if (g < (1 << 11)) { |
335 | if (out_utf8) { | |
336 | out_utf8[0] = 0xc0 | ((g >> 6) & 0x1f); | |
337 | out_utf8[1] = 0x80 | (g & 0x3f); | |
338 | } | |
e7eebcfc | 339 | return 2; |
2bb4c7e3 TG |
340 | } else if (g < (1 << 16)) { |
341 | if (out_utf8) { | |
342 | out_utf8[0] = 0xe0 | ((g >> 12) & 0x0f); | |
343 | out_utf8[1] = 0x80 | ((g >> 6) & 0x3f); | |
344 | out_utf8[2] = 0x80 | (g & 0x3f); | |
345 | } | |
e7eebcfc | 346 | return 3; |
2bb4c7e3 TG |
347 | } else if (g < (1 << 21)) { |
348 | if (out_utf8) { | |
349 | out_utf8[0] = 0xf0 | ((g >> 18) & 0x07); | |
350 | out_utf8[1] = 0x80 | ((g >> 12) & 0x3f); | |
351 | out_utf8[2] = 0x80 | ((g >> 6) & 0x3f); | |
352 | out_utf8[3] = 0x80 | (g & 0x3f); | |
353 | } | |
354 | return 4; | |
e7eebcfc | 355 | } |
f3ee6297 LP |
356 | |
357 | return 0; | |
e7eebcfc LP |
358 | } |
359 | ||
2ac2ff3f | 360 | char *utf16_to_utf8(const char16_t *s, size_t length /* bytes! */) { |
2e3d0692 | 361 | const uint8_t *f; |
e7eebcfc | 362 | char *r, *t; |
2e3d0692 | 363 | |
2ac2ff3f LP |
364 | assert(s); |
365 | ||
366 | /* Input length is in bytes, i.e. the shortest possible character takes 2 bytes. Each unicode character may | |
367 | * take up to 4 bytes in UTF-8. Let's also account for a trailing NUL byte. */ | |
368 | if (length * 2 < length) | |
369 | return NULL; /* overflow */ | |
370 | ||
371 | r = new(char, length * 2 + 1); | |
2e3d0692 LP |
372 | if (!r) |
373 | return NULL; | |
374 | ||
2ac2ff3f | 375 | f = (const uint8_t*) s; |
04166cb7 TG |
376 | t = r; |
377 | ||
2ac2ff3f | 378 | while (f + 1 < (const uint8_t*) s + length) { |
c932fb71 | 379 | char16_t w1, w2; |
04166cb7 TG |
380 | |
381 | /* see RFC 2781 section 2.2 */ | |
382 | ||
383 | w1 = f[1] << 8 | f[0]; | |
384 | f += 2; | |
385 | ||
386 | if (!utf16_is_surrogate(w1)) { | |
dcd12626 | 387 | t += utf8_encode_unichar(t, w1); |
04166cb7 TG |
388 | continue; |
389 | } | |
390 | ||
391 | if (utf16_is_trailing_surrogate(w1)) | |
2ac2ff3f LP |
392 | continue; /* spurious trailing surrogate, ignore */ |
393 | ||
394 | if (f + 1 >= (const uint8_t*) s + length) | |
04166cb7 TG |
395 | break; |
396 | ||
397 | w2 = f[1] << 8 | f[0]; | |
398 | f += 2; | |
399 | ||
400 | if (!utf16_is_trailing_surrogate(w2)) { | |
401 | f -= 2; | |
2ac2ff3f | 402 | continue; /* surrogate missing its trailing surrogate, ignore */ |
04166cb7 TG |
403 | } |
404 | ||
405 | t += utf8_encode_unichar(t, utf16_surrogate_pair_to_unichar(w1, w2)); | |
406 | } | |
2e3d0692 LP |
407 | |
408 | *t = 0; | |
7b4d7cc0 | 409 | return r; |
2e3d0692 | 410 | } |
02a36bc9 | 411 | |
80b0a597 LP |
412 | size_t utf16_encode_unichar(char16_t *out, char32_t c) { |
413 | ||
414 | /* Note that this encodes as little-endian. */ | |
415 | ||
416 | switch (c) { | |
417 | ||
418 | case 0 ... 0xd7ffU: | |
419 | case 0xe000U ... 0xffffU: | |
420 | out[0] = htole16(c); | |
421 | return 1; | |
422 | ||
423 | case 0x10000U ... 0x10ffffU: | |
424 | c -= 0x10000U; | |
425 | out[0] = htole16((c >> 10) + 0xd800U); | |
426 | out[1] = htole16((c & 0x3ffU) + 0xdc00U); | |
427 | return 2; | |
428 | ||
429 | default: /* A surrogate (invalid) */ | |
430 | return 0; | |
431 | } | |
432 | } | |
433 | ||
434 | char16_t *utf8_to_utf16(const char *s, size_t length) { | |
435 | char16_t *n, *p; | |
436 | size_t i; | |
437 | int r; | |
438 | ||
439 | assert(s); | |
440 | ||
441 | n = new(char16_t, length + 1); | |
442 | if (!n) | |
443 | return NULL; | |
444 | ||
445 | p = n; | |
446 | ||
447 | for (i = 0; i < length;) { | |
448 | char32_t unichar; | |
449 | size_t e; | |
450 | ||
84319aa7 | 451 | e = utf8_encoded_expected_len(s[i]); |
80b0a597 LP |
452 | if (e <= 1) /* Invalid and single byte characters are copied as they are */ |
453 | goto copy; | |
454 | ||
455 | if (i + e > length) /* sequence longer than input buffer, then copy as-is */ | |
456 | goto copy; | |
457 | ||
458 | r = utf8_encoded_to_unichar(s + i, &unichar); | |
459 | if (r < 0) /* sequence invalid, then copy as-is */ | |
460 | goto copy; | |
461 | ||
462 | p += utf16_encode_unichar(p, unichar); | |
463 | i += e; | |
464 | continue; | |
465 | ||
466 | copy: | |
467 | *(p++) = htole16(s[i++]); | |
468 | } | |
469 | ||
470 | *p = 0; | |
471 | return n; | |
472 | } | |
473 | ||
474 | size_t char16_strlen(const char16_t *s) { | |
475 | size_t n = 0; | |
476 | ||
477 | assert(s); | |
478 | ||
479 | while (*s != 0) | |
480 | n++, s++; | |
481 | ||
482 | return n; | |
483 | } | |
484 | ||
02a36bc9 | 485 | /* expected size used to encode one unicode char */ |
c932fb71 | 486 | static int utf8_unichar_to_encoded_len(char32_t unichar) { |
7e8185ef | 487 | |
02a36bc9 DR |
488 | if (unichar < 0x80) |
489 | return 1; | |
490 | if (unichar < 0x800) | |
491 | return 2; | |
492 | if (unichar < 0x10000) | |
493 | return 3; | |
494 | if (unichar < 0x200000) | |
495 | return 4; | |
496 | if (unichar < 0x4000000) | |
497 | return 5; | |
7e8185ef | 498 | |
02a36bc9 DR |
499 | return 6; |
500 | } | |
501 | ||
502 | /* validate one encoded unicode char and return its length */ | |
92e068b4 | 503 | int utf8_encoded_valid_unichar(const char *str, size_t length /* bytes */) { |
c932fb71 | 504 | char32_t unichar; |
7c421857 LP |
505 | size_t len, i; |
506 | int r; | |
7e8185ef LP |
507 | |
508 | assert(str); | |
92e068b4 ZJS |
509 | assert(length > 0); |
510 | ||
511 | /* We read until NUL, at most length bytes. (size_t) -1 may be used to disable the length check. */ | |
02a36bc9 | 512 | |
84319aa7 | 513 | len = utf8_encoded_expected_len(str[0]); |
02a36bc9 | 514 | if (len == 0) |
7e8185ef | 515 | return -EINVAL; |
02a36bc9 | 516 | |
92e068b4 ZJS |
517 | /* Do we have a truncated multi-byte character? */ |
518 | if (len > length) | |
519 | return -EINVAL; | |
520 | ||
02a36bc9 DR |
521 | /* ascii is valid */ |
522 | if (len == 1) | |
523 | return 1; | |
524 | ||
525 | /* check if expected encoded chars are available */ | |
526 | for (i = 0; i < len; i++) | |
527 | if ((str[i] & 0x80) != 0x80) | |
7e8185ef | 528 | return -EINVAL; |
02a36bc9 | 529 | |
c932fb71 SL |
530 | r = utf8_encoded_to_unichar(str, &unichar); |
531 | if (r < 0) | |
532 | return r; | |
02a36bc9 DR |
533 | |
534 | /* check if encoded length matches encoded value */ | |
7c421857 | 535 | if (utf8_unichar_to_encoded_len(unichar) != (int) len) |
7e8185ef | 536 | return -EINVAL; |
02a36bc9 DR |
537 | |
538 | /* check if value has valid range */ | |
f3ee6297 | 539 | if (!unichar_is_valid(unichar)) |
7e8185ef | 540 | return -EINVAL; |
02a36bc9 | 541 | |
7c421857 | 542 | return (int) len; |
02a36bc9 | 543 | } |
65ee8660 LP |
544 | |
545 | size_t utf8_n_codepoints(const char *str) { | |
546 | size_t n = 0; | |
547 | ||
548 | /* Returns the number of UTF-8 codepoints in this string, or (size_t) -1 if the string is not valid UTF-8. */ | |
549 | ||
550 | while (*str != 0) { | |
551 | int k; | |
552 | ||
92e068b4 | 553 | k = utf8_encoded_valid_unichar(str, (size_t) -1); |
65ee8660 LP |
554 | if (k < 0) |
555 | return (size_t) -1; | |
556 | ||
557 | str += k; | |
558 | n++; | |
559 | } | |
560 | ||
561 | return n; | |
562 | } | |
3f536d5b LP |
563 | |
564 | size_t utf8_console_width(const char *str) { | |
565 | size_t n = 0; | |
566 | ||
567 | /* Returns the approximate width a string will take on screen when printed on a character cell | |
568 | * terminal/console. */ | |
569 | ||
da88f542 ZJS |
570 | while (*str) { |
571 | int w; | |
3f536d5b | 572 | |
da88f542 ZJS |
573 | w = utf8_char_console_width(str); |
574 | if (w < 0) | |
3f536d5b LP |
575 | return (size_t) -1; |
576 | ||
da88f542 | 577 | n += w; |
3f536d5b | 578 | str = utf8_next_char(str); |
3f536d5b LP |
579 | } |
580 | ||
581 | return n; | |
582 | } |