src/basic/utf8.c

   1 /***
   2   This file is part of systemd.
   3
   4   Copyright 2008-2011 Kay Sievers
   5   Copyright 2012 Lennart Poettering
   6
   7   systemd is free software; you can redistribute it and/or modify it
   8   under the terms of the GNU Lesser General Public License as published by
   9   the Free Software Foundation; either version 2.1 of the License, or
  10   (at your option) any later version.
  11
  12   systemd is distributed in the hope that it will be useful, but
  13   WITHOUT ANY WARRANTY; without even the implied warranty of
  14   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  15   Lesser General Public License for more details.
  16
  17   You should have received a copy of the GNU Lesser General Public License
  18   along with systemd; If not, see <http://www.gnu.org/licenses/>.
  19 ***/
  20
  21 /* Parts of this file are based on the GLIB utf8 validation functions. The
  22  * original license text follows. */
  23
  24 /* gutf8.c - Operations on UTF-8 strings.
  25  *
  26  * Copyright (C) 1999 Tom Tromey
  27  * Copyright (C) 2000 Red Hat, Inc.
  28  *
  29  * This library is free software; you can redistribute it and/or
  30  * modify it under the terms of the GNU Library General Public
  31  * License as published by the Free Software Foundation; either
  32  * version 2 of the License, or (at your option) any later version.
  33  *
  34  * This library is distributed in the hope that it will be useful,
  35  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  36  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  37  * Library General Public License for more details.
  38  *
  39  * You should have received a copy of the GNU Library General Public
  40  * License along with this library; if not, write to the Free Software
  41  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
  42  */
  43
  44 #include <errno.h>
  45 #include <stdbool.h>
  46 #include <stdlib.h>
  47 #include <string.h>
  48
  49 #include "alloc-util.h"
  50 #include "hexdecoct.h"
  51 #include "macro.h"
  52 #include "utf8.h"
  53
  54 bool unichar_is_valid(char32_t ch) {
  55
  56         if (ch >= 0x110000) /* End of unicode space */
  57                 return false;
  58         if ((ch & 0xFFFFF800) == 0xD800) /* Reserved area for UTF-16 */
  59                 return false;
  60         if ((ch >= 0xFDD0) && (ch <= 0xFDEF)) /* Reserved */
  61                 return false;
  62         if ((ch & 0xFFFE) == 0xFFFE) /* BOM (Byte Order Mark) */
  63                 return false;
  64
  65         return true;
  66 }
  67
  68 static bool unichar_is_control(char32_t ch) {
  69
  70         /*
  71           0 to ' '-1 is the C0 range.
  72           DEL=0x7F, and DEL+1 to 0x9F is C1 range.
  73           '\t' is in C0 range, but more or less harmless and commonly used.
  74         */
  75
  76         return (ch < ' ' && !IN_SET(ch, '\t', '\n')) ||
  77                 (0x7F <= ch && ch <= 0x9F);
  78 }
  79
  80 /* count of characters used to encode one unicode char */
  81 static int utf8_encoded_expected_len(const char *str) {
  82         unsigned char c;
  83
  84         assert(str);
  85
  86         c = (unsigned char) str[0];
  87         if (c < 0x80)
  88                 return 1;
  89         if ((c & 0xe0) == 0xc0)
  90                 return 2;
  91         if ((c & 0xf0) == 0xe0)
  92                 return 3;
  93         if ((c & 0xf8) == 0xf0)
  94                 return 4;
  95         if ((c & 0xfc) == 0xf8)
  96                 return 5;
  97         if ((c & 0xfe) == 0xfc)
  98                 return 6;
  99
 100         return 0;
 101 }
 102
 103 /* decode one unicode char */
 104 int utf8_encoded_to_unichar(const char *str, char32_t *ret_unichar) {
 105         char32_t unichar;
 106         int len, i;
 107
 108         assert(str);
 109
 110         len = utf8_encoded_expected_len(str);
 111
 112         switch (len) {
 113         case 1:
 114                 *ret_unichar = (char32_t)str[0];
 115                 return 0;
 116         case 2:
 117                 unichar = str[0] & 0x1f;
 118                 break;
 119         case 3:
 120                 unichar = (char32_t)str[0] & 0x0f;
 121                 break;
 122         case 4:
 123                 unichar = (char32_t)str[0] & 0x07;
 124                 break;
 125         case 5:
 126                 unichar = (char32_t)str[0] & 0x03;
 127                 break;
 128         case 6:
 129                 unichar = (char32_t)str[0] & 0x01;
 130                 break;
 131         default:
 132                 return -EINVAL;
 133         }
 134
 135         for (i = 1; i < len; i++) {
 136                 if (((char32_t)str[i] & 0xc0) != 0x80)
 137                         return -EINVAL;
 138                 unichar <<= 6;
 139                 unichar |= (char32_t)str[i] & 0x3f;
 140         }
 141
 142         *ret_unichar = unichar;
 143
 144         return 0;
 145 }
 146
 147 bool utf8_is_printable_newline(const char* str, size_t length, bool newline) {
 148         const char *p;
 149
 150         assert(str);
 151
 152         for (p = str; length;) {
 153                 int encoded_len, r;
 154                 char32_t val;
 155
 156                 encoded_len = utf8_encoded_valid_unichar(p);
 157                 if (encoded_len < 0 ||
 158                     (size_t) encoded_len > length)
 159                         return false;
 160
 161                 r = utf8_encoded_to_unichar(p, &val);
 162                 if (r < 0 ||
 163                     unichar_is_control(val) ||
 164                     (!newline && val == '\n'))
 165                         return false;
 166
 167                 length -= encoded_len;
 168                 p += encoded_len;
 169         }
 170
 171         return true;
 172 }
 173
 174 const char *utf8_is_valid(const char *str) {
 175         const uint8_t *p;
 176
 177         assert(str);
 178
 179         for (p = (const uint8_t*) str; *p; ) {
 180                 int len;
 181
 182                 len = utf8_encoded_valid_unichar((const char *)p);
 183                 if (len < 0)
 184                         return NULL;
 185
 186                 p += len;
 187         }
 188
 189         return str;
 190 }
 191
 192 char *utf8_escape_invalid(const char *str) {
 193         char *p, *s;
 194
 195         assert(str);
 196
 197         p = s = malloc(strlen(str) * 4 + 1);
 198         if (!p)
 199                 return NULL;
 200
 201         while (*str) {
 202                 int len;
 203
 204                 len = utf8_encoded_valid_unichar(str);
 205                 if (len > 0) {
 206                         s = mempcpy(s, str, len);
 207                         str += len;
 208                 } else {
 209                         s = stpcpy(s, UTF8_REPLACEMENT_CHARACTER);
 210                         str += 1;
 211                 }
 212         }
 213
 214         *s = '\0';
 215
 216         return p;
 217 }
 218
 219 char *utf8_escape_non_printable(const char *str) {
 220         char *p, *s;
 221
 222         assert(str);
 223
 224         p = s = malloc(strlen(str) * 4 + 1);
 225         if (!p)
 226                 return NULL;
 227
 228         while (*str) {
 229                 int len;
 230
 231                 len = utf8_encoded_valid_unichar(str);
 232                 if (len > 0) {
 233                         if (utf8_is_printable(str, len)) {
 234                                 s = mempcpy(s, str, len);
 235                                 str += len;
 236                         } else {
 237                                 while (len > 0) {
 238                                         *(s++) = '\\';
 239                                         *(s++) = 'x';
 240                                         *(s++) = hexchar((int) *str >> 4);
 241                                         *(s++) = hexchar((int) *str);
 242
 243                                         str += 1;
 244                                         len--;
 245                                 }
 246                         }
 247                 } else {
 248                         s = stpcpy(s, UTF8_REPLACEMENT_CHARACTER);
 249                         str += 1;
 250                 }
 251         }
 252
 253         *s = '\0';
 254
 255         return p;
 256 }
 257
 258 char *ascii_is_valid(const char *str) {
 259         const char *p;
 260
 261         assert(str);
 262
 263         for (p = str; *p; p++)
 264                 if ((unsigned char) *p >= 128)
 265                         return NULL;
 266
 267         return (char*) str;
 268 }
 269
 270 /**
 271  * utf8_encode_unichar() - Encode single UCS-4 character as UTF-8
 272  * @out_utf8: output buffer of at least 4 bytes or NULL
 273  * @g: UCS-4 character to encode
 274  *
 275  * This encodes a single UCS-4 character as UTF-8 and writes it into @out_utf8.
 276  * The length of the character is returned. It is not zero-terminated! If the
 277  * output buffer is NULL, only the length is returned.
 278  *
 279  * Returns: The length in bytes that the UTF-8 representation does or would
 280  *          occupy.
 281  */
 282 size_t utf8_encode_unichar(char *out_utf8, char32_t g) {
 283
 284         if (g < (1 << 7)) {
 285                 if (out_utf8)
 286                         out_utf8[0] = g & 0x7f;
 287                 return 1;
 288         } else if (g < (1 << 11)) {
 289                 if (out_utf8) {
 290                         out_utf8[0] = 0xc0 | ((g >> 6) & 0x1f);
 291                         out_utf8[1] = 0x80 | (g & 0x3f);
 292                 }
 293                 return 2;
 294         } else if (g < (1 << 16)) {
 295                 if (out_utf8) {
 296                         out_utf8[0] = 0xe0 | ((g >> 12) & 0x0f);
 297                         out_utf8[1] = 0x80 | ((g >> 6) & 0x3f);
 298                         out_utf8[2] = 0x80 | (g & 0x3f);
 299                 }
 300                 return 3;
 301         } else if (g < (1 << 21)) {
 302                 if (out_utf8) {
 303                         out_utf8[0] = 0xf0 | ((g >> 18) & 0x07);
 304                         out_utf8[1] = 0x80 | ((g >> 12) & 0x3f);
 305                         out_utf8[2] = 0x80 | ((g >> 6) & 0x3f);
 306                         out_utf8[3] = 0x80 | (g & 0x3f);
 307                 }
 308                 return 4;
 309         }
 310
 311         return 0;
 312 }
 313
 314 char *utf16_to_utf8(const void *s, size_t length) {
 315         const uint8_t *f;
 316         char *r, *t;
 317
 318         r = new(char, (length * 4 + 1) / 2 + 1);
 319         if (!r)
 320                 return NULL;
 321
 322         f = s;
 323         t = r;
 324
 325         while (f < (const uint8_t*) s + length) {
 326                 char16_t w1, w2;
 327
 328                 /* see RFC 2781 section 2.2 */
 329
 330                 w1 = f[1] << 8 | f[0];
 331                 f += 2;
 332
 333                 if (!utf16_is_surrogate(w1)) {
 334                         t += utf8_encode_unichar(t, w1);
 335
 336                         continue;
 337                 }
 338
 339                 if (utf16_is_trailing_surrogate(w1))
 340                         continue;
 341                 else if (f >= (const uint8_t*) s + length)
 342                         break;
 343
 344                 w2 = f[1] << 8 | f[0];
 345                 f += 2;
 346
 347                 if (!utf16_is_trailing_surrogate(w2)) {
 348                         f -= 2;
 349                         continue;
 350                 }
 351
 352                 t += utf8_encode_unichar(t, utf16_surrogate_pair_to_unichar(w1, w2));
 353         }
 354
 355         *t = 0;
 356         return r;
 357 }
 358
 359 /* expected size used to encode one unicode char */
 360 static int utf8_unichar_to_encoded_len(char32_t unichar) {
 361
 362         if (unichar < 0x80)
 363                 return 1;
 364         if (unichar < 0x800)
 365                 return 2;
 366         if (unichar < 0x10000)
 367                 return 3;
 368         if (unichar < 0x200000)
 369                 return 4;
 370         if (unichar < 0x4000000)
 371                 return 5;
 372
 373         return 6;
 374 }
 375
 376 /* validate one encoded unicode char and return its length */
 377 int utf8_encoded_valid_unichar(const char *str) {
 378         int len, i, r;
 379         char32_t unichar;
 380
 381         assert(str);
 382
 383         len = utf8_encoded_expected_len(str);
 384         if (len == 0)
 385                 return -EINVAL;
 386
 387         /* ascii is valid */
 388         if (len == 1)
 389                 return 1;
 390
 391         /* check if expected encoded chars are available */
 392         for (i = 0; i < len; i++)
 393                 if ((str[i] & 0x80) != 0x80)
 394                         return -EINVAL;
 395
 396         r = utf8_encoded_to_unichar(str, &unichar);
 397         if (r < 0)
 398                 return r;
 399
 400         /* check if encoded length matches encoded value */
 401         if (utf8_unichar_to_encoded_len(unichar) != len)
 402                 return -EINVAL;
 403
 404         /* check if value has valid range */
 405         if (!unichar_is_valid(unichar))
 406                 return -EINVAL;
 407
 408         return len;
 409 }