]>
Commit | Line | Data |
---|---|---|
6952ebae | 1 | /* SPDX-License-Identifier: LGPL-2.0-or-later */ |
7f110ff9 | 2 | |
6952ebae | 3 | /* Parts of this file are based on the GLIB utf8 validation functions. The original copyright follows. |
7f110ff9 | 4 | * |
6952ebae | 5 | * gutf8.c - Operations on UTF-8 strings. |
7f110ff9 LP |
6 | * Copyright (C) 1999 Tom Tromey |
7 | * Copyright (C) 2000 Red Hat, Inc. | |
7f110ff9 LP |
8 | */ |
9 | ||
10 | #include <errno.h> | |
7f110ff9 | 11 | #include <stdbool.h> |
cf0fbc49 | 12 | #include <stdlib.h> |
7f110ff9 | 13 | |
b5efdb8a | 14 | #include "alloc-util.h" |
3f536d5b | 15 | #include "gunicode.h" |
e4e73a63 | 16 | #include "hexdecoct.h" |
11c3a366 | 17 | #include "macro.h" |
4c6d5139 | 18 | #include "string-util.h" |
7f110ff9 LP |
19 | #include "utf8.h" |
20 | ||
c932fb71 | 21 | bool unichar_is_valid(char32_t ch) { |
7f110ff9 LP |
22 | |
23 | if (ch >= 0x110000) /* End of unicode space */ | |
24 | return false; | |
25 | if ((ch & 0xFFFFF800) == 0xD800) /* Reserved area for UTF-16 */ | |
26 | return false; | |
27 | if ((ch >= 0xFDD0) && (ch <= 0xFDEF)) /* Reserved */ | |
28 | return false; | |
29 | if ((ch & 0xFFFE) == 0xFFFE) /* BOM (Byte Order Mark) */ | |
30 | return false; | |
31 | ||
32 | return true; | |
33 | } | |
34 | ||
c932fb71 | 35 | static bool unichar_is_control(char32_t ch) { |
ba961854 ZJS |
36 | |
37 | /* | |
38 | 0 to ' '-1 is the C0 range. | |
39 | DEL=0x7F, and DEL+1 to 0x9F is C1 range. | |
40 | '\t' is in C0 range, but more or less harmless and commonly used. | |
41 | */ | |
42 | ||
4c701096 | 43 | return (ch < ' ' && !IN_SET(ch, '\t', '\n')) || |
ba961854 ZJS |
44 | (0x7F <= ch && ch <= 0x9F); |
45 | } | |
46 | ||
7991ac34 | 47 | /* count of characters used to encode one unicode char */ |
84319aa7 | 48 | static size_t utf8_encoded_expected_len(uint8_t c) { |
7991ac34 DR |
49 | if (c < 0x80) |
50 | return 1; | |
51 | if ((c & 0xe0) == 0xc0) | |
52 | return 2; | |
53 | if ((c & 0xf0) == 0xe0) | |
54 | return 3; | |
55 | if ((c & 0xf8) == 0xf0) | |
56 | return 4; | |
57 | if ((c & 0xfc) == 0xf8) | |
58 | return 5; | |
59 | if ((c & 0xfe) == 0xfc) | |
60 | return 6; | |
7e8185ef | 61 | |
7991ac34 DR |
62 | return 0; |
63 | } | |
ba961854 | 64 | |
7991ac34 | 65 | /* decode one unicode char */ |
c932fb71 SL |
66 | int utf8_encoded_to_unichar(const char *str, char32_t *ret_unichar) { |
67 | char32_t unichar; | |
fe96c0f8 | 68 | size_t len; |
7e8185ef LP |
69 | |
70 | assert(str); | |
ba961854 | 71 | |
84319aa7 | 72 | len = utf8_encoded_expected_len(str[0]); |
7e8185ef | 73 | |
7991ac34 DR |
74 | switch (len) { |
75 | case 1: | |
c932fb71 | 76 | *ret_unichar = (char32_t)str[0]; |
9579e9a5 | 77 | return 1; |
7991ac34 DR |
78 | case 2: |
79 | unichar = str[0] & 0x1f; | |
80 | break; | |
81 | case 3: | |
c932fb71 | 82 | unichar = (char32_t)str[0] & 0x0f; |
7991ac34 DR |
83 | break; |
84 | case 4: | |
c932fb71 | 85 | unichar = (char32_t)str[0] & 0x07; |
7991ac34 DR |
86 | break; |
87 | case 5: | |
c932fb71 | 88 | unichar = (char32_t)str[0] & 0x03; |
7991ac34 DR |
89 | break; |
90 | case 6: | |
c932fb71 | 91 | unichar = (char32_t)str[0] & 0x01; |
7991ac34 DR |
92 | break; |
93 | default: | |
7e8185ef | 94 | return -EINVAL; |
ba961854 ZJS |
95 | } |
96 | ||
fe96c0f8 | 97 | for (size_t i = 1; i < len; i++) { |
c932fb71 | 98 | if (((char32_t)str[i] & 0xc0) != 0x80) |
7e8185ef | 99 | return -EINVAL; |
7c421857 | 100 | |
7991ac34 | 101 | unichar <<= 6; |
c932fb71 | 102 | unichar |= (char32_t)str[i] & 0x3f; |
7991ac34 DR |
103 | } |
104 | ||
c932fb71 | 105 | *ret_unichar = unichar; |
9579e9a5 | 106 | return len; |
ba961854 ZJS |
107 | } |
108 | ||
618727da | 109 | bool utf8_is_printable_newline(const char* str, size_t length, bool allow_newline) { |
7f110ff9 LP |
110 | assert(str); |
111 | ||
a456086b | 112 | for (const char *p = str; length > 0;) { |
9579e9a5 | 113 | int encoded_len; |
c932fb71 | 114 | char32_t val; |
7e8185ef | 115 | |
92e068b4 ZJS |
116 | encoded_len = utf8_encoded_valid_unichar(p, length); |
117 | if (encoded_len < 0) | |
144b3d9e | 118 | return false; |
92e068b4 | 119 | assert(encoded_len > 0 && (size_t) encoded_len <= length); |
144b3d9e | 120 | |
9579e9a5 | 121 | if (utf8_encoded_to_unichar(p, &val) < 0 || |
f3ee6297 | 122 | unichar_is_control(val) || |
618727da | 123 | (!allow_newline && val == '\n')) |
7991ac34 | 124 | return false; |
7f110ff9 | 125 | |
7991ac34 | 126 | length -= encoded_len; |
a7176505 | 127 | p += encoded_len; |
7f110ff9 LP |
128 | } |
129 | ||
7991ac34 | 130 | return true; |
7f110ff9 LP |
131 | } |
132 | ||
80ab31a4 ZJS |
133 | char *utf8_is_valid_n(const char *str, size_t len_bytes) { |
134 | /* Check if the string is composed of valid utf8 characters. If length len_bytes is given, stop after | |
135 | * len_bytes. Otherwise, stop at NUL. */ | |
7f110ff9 LP |
136 | |
137 | assert(str); | |
138 | ||
f5fbe71d | 139 | for (const char *p = str; len_bytes != SIZE_MAX ? (size_t) (p - str) < len_bytes : *p != '\0'; ) { |
faaa5728 LP |
140 | int len; |
141 | ||
f5fbe71d | 142 | if (_unlikely_(*p == '\0') && len_bytes != SIZE_MAX) |
80ab31a4 ZJS |
143 | return NULL; /* embedded NUL */ |
144 | ||
145 | len = utf8_encoded_valid_unichar(p, | |
f5fbe71d | 146 | len_bytes != SIZE_MAX ? len_bytes - (p - str) : SIZE_MAX); |
80ab31a4 ZJS |
147 | if (_unlikely_(len < 0)) |
148 | return NULL; /* invalid character */ | |
7991ac34 DR |
149 | |
150 | p += len; | |
151 | } | |
7f110ff9 | 152 | |
e71fb4b3 | 153 | return (char*) str; |
7f110ff9 LP |
154 | } |
155 | ||
550a40ec ZJS |
156 | char *utf8_escape_invalid(const char *str) { |
157 | char *p, *s; | |
158 | ||
159 | assert(str); | |
160 | ||
161 | p = s = malloc(strlen(str) * 4 + 1); | |
162 | if (!p) | |
163 | return NULL; | |
164 | ||
165 | while (*str) { | |
166 | int len; | |
167 | ||
f5fbe71d | 168 | len = utf8_encoded_valid_unichar(str, SIZE_MAX); |
550a40ec ZJS |
169 | if (len > 0) { |
170 | s = mempcpy(s, str, len); | |
171 | str += len; | |
172 | } else { | |
3c6d3052 | 173 | s = stpcpy(s, UTF8_REPLACEMENT_CHARACTER); |
550a40ec ZJS |
174 | str += 1; |
175 | } | |
176 | } | |
7e8185ef | 177 | |
550a40ec | 178 | *s = '\0'; |
523e1b14 | 179 | return str_realloc(p); |
550a40ec ZJS |
180 | } |
181 | ||
da88f542 ZJS |
182 | static int utf8_char_console_width(const char *str) { |
183 | char32_t c; | |
184 | int r; | |
185 | ||
186 | r = utf8_encoded_to_unichar(str, &c); | |
187 | if (r < 0) | |
188 | return r; | |
189 | ||
190 | /* TODO: we should detect combining characters */ | |
191 | ||
192 | return unichar_iswide(c) ? 2 : 1; | |
193 | } | |
194 | ||
fc96e5c0 | 195 | char *utf8_escape_non_printable_full(const char *str, size_t console_width, bool force_ellipsis) { |
da88f542 ZJS |
196 | char *p, *s, *prev_s; |
197 | size_t n = 0; /* estimated print width */ | |
fec84576 WC |
198 | |
199 | assert(str); | |
200 | ||
da88f542 ZJS |
201 | if (console_width == 0) |
202 | return strdup(""); | |
203 | ||
204 | p = s = prev_s = malloc(strlen(str) * 4 + 1); | |
fec84576 WC |
205 | if (!p) |
206 | return NULL; | |
207 | ||
da88f542 | 208 | for (;;) { |
fec84576 | 209 | int len; |
da88f542 ZJS |
210 | char *saved_s = s; |
211 | ||
fc96e5c0 ZJS |
212 | if (!*str) { /* done! */ |
213 | if (force_ellipsis) | |
214 | goto truncation; | |
215 | else | |
216 | goto finish; | |
217 | } | |
fec84576 | 218 | |
f5fbe71d | 219 | len = utf8_encoded_valid_unichar(str, SIZE_MAX); |
fec84576 WC |
220 | if (len > 0) { |
221 | if (utf8_is_printable(str, len)) { | |
da88f542 ZJS |
222 | int w; |
223 | ||
224 | w = utf8_char_console_width(str); | |
225 | assert(w >= 0); | |
226 | if (n + w > console_width) | |
227 | goto truncation; | |
228 | ||
fec84576 WC |
229 | s = mempcpy(s, str, len); |
230 | str += len; | |
da88f542 ZJS |
231 | n += w; |
232 | ||
fec84576 | 233 | } else { |
da88f542 ZJS |
234 | for (; len > 0; len--) { |
235 | if (n + 4 > console_width) | |
236 | goto truncation; | |
237 | ||
fec84576 WC |
238 | *(s++) = '\\'; |
239 | *(s++) = 'x'; | |
240 | *(s++) = hexchar((int) *str >> 4); | |
241 | *(s++) = hexchar((int) *str); | |
fec84576 | 242 | |
3c6d3052 | 243 | str += 1; |
da88f542 | 244 | n += 4; |
3c6d3052 | 245 | } |
fec84576 WC |
246 | } |
247 | } else { | |
da88f542 ZJS |
248 | if (n + 1 > console_width) |
249 | goto truncation; | |
250 | ||
251 | s = mempcpy(s, UTF8_REPLACEMENT_CHARACTER, strlen(UTF8_REPLACEMENT_CHARACTER)); | |
fec84576 | 252 | str += 1; |
da88f542 | 253 | n += 1; |
fec84576 | 254 | } |
da88f542 ZJS |
255 | |
256 | prev_s = saved_s; | |
fec84576 WC |
257 | } |
258 | ||
da88f542 ZJS |
259 | truncation: |
260 | /* Try to go back one if we don't have enough space for the ellipsis */ | |
fc96e5c0 | 261 | if (n + 1 > console_width) |
da88f542 ZJS |
262 | s = prev_s; |
263 | ||
264 | s = mempcpy(s, "…", strlen("…")); | |
fec84576 | 265 | |
da88f542 ZJS |
266 | finish: |
267 | *s = '\0'; | |
523e1b14 | 268 | return str_realloc(p); |
fec84576 WC |
269 | } |
270 | ||
7f110ff9 | 271 | char *ascii_is_valid(const char *str) { |
294a3121 ZJS |
272 | /* Check whether the string consists of valid ASCII bytes, |
273 | * i.e values between 0 and 127, inclusive. */ | |
274 | ||
7f110ff9 LP |
275 | assert(str); |
276 | ||
a456086b | 277 | for (const char *p = str; *p; p++) |
7f110ff9 LP |
278 | if ((unsigned char) *p >= 128) |
279 | return NULL; | |
280 | ||
281 | return (char*) str; | |
282 | } | |
283 | ||
294a3121 | 284 | char *ascii_is_valid_n(const char *str, size_t len) { |
294a3121 ZJS |
285 | /* Very similar to ascii_is_valid(), but checks exactly len |
286 | * bytes and rejects any NULs in that range. */ | |
287 | ||
288 | assert(str); | |
289 | ||
fe96c0f8 | 290 | for (size_t i = 0; i < len; i++) |
294a3121 ZJS |
291 | if ((unsigned char) str[i] >= 128 || str[i] == 0) |
292 | return NULL; | |
293 | ||
294 | return (char*) str; | |
295 | } | |
296 | ||
9b49a3b4 ZJS |
297 | int utf8_to_ascii(const char *str, char replacement_char, char **ret) { |
298 | /* Convert to a string that has only ASCII chars, replacing anything that is not ASCII | |
299 | * by replacement_char. */ | |
300 | ||
301 | _cleanup_free_ char *ans = new(char, strlen(str) + 1); | |
302 | if (!ans) | |
303 | return -ENOMEM; | |
304 | ||
305 | char *q = ans; | |
306 | ||
307 | for (const char *p = str; *p; q++) { | |
308 | int l; | |
309 | ||
310 | l = utf8_encoded_valid_unichar(p, SIZE_MAX); | |
311 | if (l < 0) /* Non-UTF-8, let's not even try to propagate the garbage */ | |
312 | return l; | |
313 | ||
314 | if (l == 1) | |
315 | *q = *p; | |
316 | else | |
317 | /* non-ASCII, we need to replace it */ | |
318 | *q = replacement_char; | |
319 | ||
320 | p += l; | |
321 | } | |
322 | *q = '\0'; | |
323 | ||
324 | *ret = TAKE_PTR(ans); | |
325 | return 0; | |
326 | } | |
327 | ||
2bb4c7e3 TG |
328 | /** |
329 | * utf8_encode_unichar() - Encode single UCS-4 character as UTF-8 | |
330 | * @out_utf8: output buffer of at least 4 bytes or NULL | |
331 | * @g: UCS-4 character to encode | |
332 | * | |
333 | * This encodes a single UCS-4 character as UTF-8 and writes it into @out_utf8. | |
334 | * The length of the character is returned. It is not zero-terminated! If the | |
335 | * output buffer is NULL, only the length is returned. | |
336 | * | |
337 | * Returns: The length in bytes that the UTF-8 representation does or would | |
338 | * occupy. | |
339 | */ | |
c932fb71 | 340 | size_t utf8_encode_unichar(char *out_utf8, char32_t g) { |
f3ee6297 | 341 | |
2bb4c7e3 TG |
342 | if (g < (1 << 7)) { |
343 | if (out_utf8) | |
344 | out_utf8[0] = g & 0x7f; | |
e7eebcfc | 345 | return 1; |
2bb4c7e3 TG |
346 | } else if (g < (1 << 11)) { |
347 | if (out_utf8) { | |
348 | out_utf8[0] = 0xc0 | ((g >> 6) & 0x1f); | |
349 | out_utf8[1] = 0x80 | (g & 0x3f); | |
350 | } | |
e7eebcfc | 351 | return 2; |
2bb4c7e3 TG |
352 | } else if (g < (1 << 16)) { |
353 | if (out_utf8) { | |
354 | out_utf8[0] = 0xe0 | ((g >> 12) & 0x0f); | |
355 | out_utf8[1] = 0x80 | ((g >> 6) & 0x3f); | |
356 | out_utf8[2] = 0x80 | (g & 0x3f); | |
357 | } | |
e7eebcfc | 358 | return 3; |
2bb4c7e3 TG |
359 | } else if (g < (1 << 21)) { |
360 | if (out_utf8) { | |
361 | out_utf8[0] = 0xf0 | ((g >> 18) & 0x07); | |
362 | out_utf8[1] = 0x80 | ((g >> 12) & 0x3f); | |
363 | out_utf8[2] = 0x80 | ((g >> 6) & 0x3f); | |
364 | out_utf8[3] = 0x80 | (g & 0x3f); | |
365 | } | |
366 | return 4; | |
e7eebcfc | 367 | } |
f3ee6297 LP |
368 | |
369 | return 0; | |
e7eebcfc LP |
370 | } |
371 | ||
2ac2ff3f | 372 | char *utf16_to_utf8(const char16_t *s, size_t length /* bytes! */) { |
2e3d0692 | 373 | const uint8_t *f; |
e7eebcfc | 374 | char *r, *t; |
2e3d0692 | 375 | |
ba091282 LP |
376 | if (length == 0) |
377 | return new0(char, 1); | |
378 | ||
2ac2ff3f LP |
379 | assert(s); |
380 | ||
ba091282 LP |
381 | if (length == SIZE_MAX) { |
382 | length = char16_strlen(s); | |
383 | ||
384 | if (length > SIZE_MAX/2) | |
385 | return NULL; /* overflow */ | |
386 | ||
387 | length *= 2; | |
388 | } | |
389 | ||
2ac2ff3f LP |
390 | /* Input length is in bytes, i.e. the shortest possible character takes 2 bytes. Each unicode character may |
391 | * take up to 4 bytes in UTF-8. Let's also account for a trailing NUL byte. */ | |
ba091282 | 392 | if (length > (SIZE_MAX - 1) / 2) |
2ac2ff3f LP |
393 | return NULL; /* overflow */ |
394 | ||
395 | r = new(char, length * 2 + 1); | |
2e3d0692 LP |
396 | if (!r) |
397 | return NULL; | |
398 | ||
2ac2ff3f | 399 | f = (const uint8_t*) s; |
04166cb7 TG |
400 | t = r; |
401 | ||
2ac2ff3f | 402 | while (f + 1 < (const uint8_t*) s + length) { |
c932fb71 | 403 | char16_t w1, w2; |
04166cb7 TG |
404 | |
405 | /* see RFC 2781 section 2.2 */ | |
406 | ||
407 | w1 = f[1] << 8 | f[0]; | |
408 | f += 2; | |
409 | ||
410 | if (!utf16_is_surrogate(w1)) { | |
dcd12626 | 411 | t += utf8_encode_unichar(t, w1); |
04166cb7 TG |
412 | continue; |
413 | } | |
414 | ||
415 | if (utf16_is_trailing_surrogate(w1)) | |
2ac2ff3f LP |
416 | continue; /* spurious trailing surrogate, ignore */ |
417 | ||
418 | if (f + 1 >= (const uint8_t*) s + length) | |
04166cb7 TG |
419 | break; |
420 | ||
421 | w2 = f[1] << 8 | f[0]; | |
422 | f += 2; | |
423 | ||
424 | if (!utf16_is_trailing_surrogate(w2)) { | |
425 | f -= 2; | |
2ac2ff3f | 426 | continue; /* surrogate missing its trailing surrogate, ignore */ |
04166cb7 TG |
427 | } |
428 | ||
429 | t += utf8_encode_unichar(t, utf16_surrogate_pair_to_unichar(w1, w2)); | |
430 | } | |
2e3d0692 LP |
431 | |
432 | *t = 0; | |
7b4d7cc0 | 433 | return r; |
2e3d0692 | 434 | } |
02a36bc9 | 435 | |
80b0a597 LP |
436 | size_t utf16_encode_unichar(char16_t *out, char32_t c) { |
437 | ||
438 | /* Note that this encodes as little-endian. */ | |
439 | ||
440 | switch (c) { | |
441 | ||
442 | case 0 ... 0xd7ffU: | |
443 | case 0xe000U ... 0xffffU: | |
444 | out[0] = htole16(c); | |
445 | return 1; | |
446 | ||
447 | case 0x10000U ... 0x10ffffU: | |
448 | c -= 0x10000U; | |
449 | out[0] = htole16((c >> 10) + 0xd800U); | |
450 | out[1] = htole16((c & 0x3ffU) + 0xdc00U); | |
451 | return 2; | |
452 | ||
453 | default: /* A surrogate (invalid) */ | |
454 | return 0; | |
455 | } | |
456 | } | |
457 | ||
458 | char16_t *utf8_to_utf16(const char *s, size_t length) { | |
459 | char16_t *n, *p; | |
80b0a597 LP |
460 | int r; |
461 | ||
ba091282 LP |
462 | if (length == 0) |
463 | return new0(char16_t, 1); | |
464 | ||
80b0a597 LP |
465 | assert(s); |
466 | ||
ba091282 LP |
467 | if (length == SIZE_MAX) |
468 | length = strlen(s); | |
469 | ||
470 | if (length > SIZE_MAX - 1) | |
471 | return NULL; /* overflow */ | |
472 | ||
80b0a597 LP |
473 | n = new(char16_t, length + 1); |
474 | if (!n) | |
475 | return NULL; | |
476 | ||
477 | p = n; | |
478 | ||
fe96c0f8 | 479 | for (size_t i = 0; i < length;) { |
80b0a597 LP |
480 | char32_t unichar; |
481 | size_t e; | |
482 | ||
84319aa7 | 483 | e = utf8_encoded_expected_len(s[i]); |
80b0a597 LP |
484 | if (e <= 1) /* Invalid and single byte characters are copied as they are */ |
485 | goto copy; | |
486 | ||
487 | if (i + e > length) /* sequence longer than input buffer, then copy as-is */ | |
488 | goto copy; | |
489 | ||
490 | r = utf8_encoded_to_unichar(s + i, &unichar); | |
491 | if (r < 0) /* sequence invalid, then copy as-is */ | |
492 | goto copy; | |
493 | ||
494 | p += utf16_encode_unichar(p, unichar); | |
495 | i += e; | |
496 | continue; | |
497 | ||
498 | copy: | |
499 | *(p++) = htole16(s[i++]); | |
500 | } | |
501 | ||
502 | *p = 0; | |
503 | return n; | |
504 | } | |
505 | ||
506 | size_t char16_strlen(const char16_t *s) { | |
507 | size_t n = 0; | |
508 | ||
509 | assert(s); | |
510 | ||
511 | while (*s != 0) | |
512 | n++, s++; | |
513 | ||
514 | return n; | |
515 | } | |
516 | ||
02a36bc9 | 517 | /* expected size used to encode one unicode char */ |
c932fb71 | 518 | static int utf8_unichar_to_encoded_len(char32_t unichar) { |
7e8185ef | 519 | |
02a36bc9 DR |
520 | if (unichar < 0x80) |
521 | return 1; | |
522 | if (unichar < 0x800) | |
523 | return 2; | |
524 | if (unichar < 0x10000) | |
525 | return 3; | |
526 | if (unichar < 0x200000) | |
527 | return 4; | |
528 | if (unichar < 0x4000000) | |
529 | return 5; | |
7e8185ef | 530 | |
02a36bc9 DR |
531 | return 6; |
532 | } | |
533 | ||
534 | /* validate one encoded unicode char and return its length */ | |
92e068b4 | 535 | int utf8_encoded_valid_unichar(const char *str, size_t length /* bytes */) { |
c932fb71 | 536 | char32_t unichar; |
fe96c0f8 | 537 | size_t len; |
7c421857 | 538 | int r; |
7e8185ef LP |
539 | |
540 | assert(str); | |
92e068b4 ZJS |
541 | assert(length > 0); |
542 | ||
f5fbe71d | 543 | /* We read until NUL, at most length bytes. SIZE_MAX may be used to disable the length check. */ |
02a36bc9 | 544 | |
84319aa7 | 545 | len = utf8_encoded_expected_len(str[0]); |
02a36bc9 | 546 | if (len == 0) |
7e8185ef | 547 | return -EINVAL; |
02a36bc9 | 548 | |
92e068b4 ZJS |
549 | /* Do we have a truncated multi-byte character? */ |
550 | if (len > length) | |
551 | return -EINVAL; | |
552 | ||
02a36bc9 DR |
553 | /* ascii is valid */ |
554 | if (len == 1) | |
555 | return 1; | |
556 | ||
557 | /* check if expected encoded chars are available */ | |
fe96c0f8 | 558 | for (size_t i = 0; i < len; i++) |
02a36bc9 | 559 | if ((str[i] & 0x80) != 0x80) |
7e8185ef | 560 | return -EINVAL; |
02a36bc9 | 561 | |
c932fb71 SL |
562 | r = utf8_encoded_to_unichar(str, &unichar); |
563 | if (r < 0) | |
564 | return r; | |
02a36bc9 DR |
565 | |
566 | /* check if encoded length matches encoded value */ | |
7c421857 | 567 | if (utf8_unichar_to_encoded_len(unichar) != (int) len) |
7e8185ef | 568 | return -EINVAL; |
02a36bc9 DR |
569 | |
570 | /* check if value has valid range */ | |
f3ee6297 | 571 | if (!unichar_is_valid(unichar)) |
7e8185ef | 572 | return -EINVAL; |
02a36bc9 | 573 | |
7c421857 | 574 | return (int) len; |
02a36bc9 | 575 | } |
65ee8660 LP |
576 | |
577 | size_t utf8_n_codepoints(const char *str) { | |
578 | size_t n = 0; | |
579 | ||
f5fbe71d | 580 | /* Returns the number of UTF-8 codepoints in this string, or SIZE_MAX if the string is not valid UTF-8. */ |
65ee8660 LP |
581 | |
582 | while (*str != 0) { | |
583 | int k; | |
584 | ||
f5fbe71d | 585 | k = utf8_encoded_valid_unichar(str, SIZE_MAX); |
65ee8660 | 586 | if (k < 0) |
f5fbe71d | 587 | return SIZE_MAX; |
65ee8660 LP |
588 | |
589 | str += k; | |
590 | n++; | |
591 | } | |
592 | ||
593 | return n; | |
594 | } | |
3f536d5b LP |
595 | |
596 | size_t utf8_console_width(const char *str) { | |
597 | size_t n = 0; | |
598 | ||
599 | /* Returns the approximate width a string will take on screen when printed on a character cell | |
600 | * terminal/console. */ | |
601 | ||
da88f542 ZJS |
602 | while (*str) { |
603 | int w; | |
3f536d5b | 604 | |
da88f542 ZJS |
605 | w = utf8_char_console_width(str); |
606 | if (w < 0) | |
f5fbe71d | 607 | return SIZE_MAX; |
3f536d5b | 608 | |
da88f542 | 609 | n += w; |
3f536d5b | 610 | str = utf8_next_char(str); |
3f536d5b LP |
611 | } |
612 | ||
613 | return n; | |
614 | } |