src/shared/utf8.c

   1 /*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/
   2
   3 /***
   4   This file is part of systemd.
   5
   6   Copyright 2008-2011 Kay Sievers
   7   Copyright 2012 Lennart Poettering
   8
   9   systemd is free software; you can redistribute it and/or modify it
  10   under the terms of the GNU Lesser General Public License as published by
  11   the Free Software Foundation; either version 2.1 of the License, or
  12   (at your option) any later version.
  13
  14   systemd is distributed in the hope that it will be useful, but
  15   WITHOUT ANY WARRANTY; without even the implied warranty of
  16   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
  17   Lesser General Public License for more details.
  18
  19   You should have received a copy of the GNU Lesser General Public License
  20   along with systemd; If not, see <http://www.gnu.org/licenses/>.
  21 ***/
  22
  23 /* Parts of this file are based on the GLIB utf8 validation functions. The
  24  * original license text follows. */
  25
  26 /* gutf8.c - Operations on UTF-8 strings.
  27  *
  28  * Copyright (C) 1999 Tom Tromey
  29  * Copyright (C) 2000 Red Hat, Inc.
  30  *
  31  * This library is free software; you can redistribute it and/or
  32  * modify it under the terms of the GNU Library General Public
  33  * License as published by the Free Software Foundation; either
  34  * version 2 of the License, or (at your option) any later version.
  35  *
  36  * This library is distributed in the hope that it will be useful,
  37  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  38  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  39  * Library General Public License for more details.
  40  *
  41  * You should have received a copy of the GNU Library General Public
  42  * License along with this library; if not, write to the Free Software
  43  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301  USA
  44  */
  45
  46 #include <errno.h>
  47 #include <stdlib.h>
  48 #include <inttypes.h>
  49 #include <string.h>
  50 #include <stdbool.h>
  51
  52 #include "utf8.h"
  53 #include "util.h"
  54
  55 static inline bool is_unicode_valid(uint32_t ch) {
  56
  57         if (ch >= 0x110000) /* End of unicode space */
  58                 return false;
  59         if ((ch & 0xFFFFF800) == 0xD800) /* Reserved area for UTF-16 */
  60                 return false;
  61         if ((ch >= 0xFDD0) && (ch <= 0xFDEF)) /* Reserved */
  62                 return false;
  63         if ((ch & 0xFFFE) == 0xFFFE) /* BOM (Byte Order Mark) */
  64                 return false;
  65
  66         return true;
  67 }
  68
  69 static bool is_unicode_control(uint32_t ch) {
  70
  71         /*
  72           0 to ' '-1 is the C0 range.
  73           DEL=0x7F, and DEL+1 to 0x9F is C1 range.
  74           '\t' is in C0 range, but more or less harmless and commonly used.
  75         */
  76
  77         return (ch < ' ' && ch != '\t' && ch != '\n') ||
  78                 (0x7F <= ch && ch <= 0x9F);
  79 }
  80
  81 /* count of characters used to encode one unicode char */
  82 static int utf8_encoded_expected_len(const char *str) {
  83         unsigned char c = (unsigned char)str[0];
  84
  85         if (c < 0x80)
  86                 return 1;
  87         if ((c & 0xe0) == 0xc0)
  88                 return 2;
  89         if ((c & 0xf0) == 0xe0)
  90                 return 3;
  91         if ((c & 0xf8) == 0xf0)
  92                 return 4;
  93         if ((c & 0xfc) == 0xf8)
  94                 return 5;
  95         if ((c & 0xfe) == 0xfc)
  96                 return 6;
  97         return 0;
  98 }
  99
 100 /* decode one unicode char */
 101 int utf8_encoded_to_unichar(const char *str) {
 102         int unichar;
 103         int len;
 104         int i;
 105
 106         len = utf8_encoded_expected_len(str);
 107         switch (len) {
 108         case 1:
 109                 return (int)str[0];
 110         case 2:
 111                 unichar = str[0] & 0x1f;
 112                 break;
 113         case 3:
 114                 unichar = (int)str[0] & 0x0f;
 115                 break;
 116         case 4:
 117                 unichar = (int)str[0] & 0x07;
 118                 break;
 119         case 5:
 120                 unichar = (int)str[0] & 0x03;
 121                 break;
 122         case 6:
 123                 unichar = (int)str[0] & 0x01;
 124                 break;
 125         default:
 126                 return -1;
 127         }
 128
 129         for (i = 1; i < len; i++) {
 130                 if (((int)str[i] & 0xc0) != 0x80)
 131                         return -1;
 132                 unichar <<= 6;
 133                 unichar |= (int)str[i] & 0x3f;
 134         }
 135
 136         return unichar;
 137 }
 138
 139 bool utf8_is_printable(const char* str, size_t length) {
 140         const uint8_t *p;
 141
 142         assert(str);
 143
 144         for (p = (const uint8_t*) str; length;) {
 145                 int encoded_len = utf8_encoded_valid_unichar((const char *)p);
 146                 int val = utf8_encoded_to_unichar((const char*)p);
 147
 148                 if (encoded_len < 0 || val < 0 || is_unicode_control(val))
 149                         return false;
 150
 151                 length -= encoded_len;
 152                 p += encoded_len;
 153         }
 154
 155         return true;
 156 }
 157
 158 const char *utf8_is_valid(const char *str) {
 159         const uint8_t *p;
 160
 161         assert(str);
 162
 163         for (p = (const uint8_t*) str; *p; ) {
 164                 int len;
 165
 166                 len = utf8_encoded_valid_unichar((const char *)p);
 167
 168                 if (len < 0)
 169                         return NULL;
 170
 171                 p += len;
 172         }
 173
 174         return str;
 175 }
 176
 177 char *ascii_is_valid(const char *str) {
 178         const char *p;
 179
 180         assert(str);
 181
 182         for (p = str; *p; p++)
 183                 if ((unsigned char) *p >= 128)
 184                         return NULL;
 185
 186         return (char*) str;
 187 }
 188
 189 char *utf16_to_utf8(const void *s, size_t length) {
 190         char *r;
 191         const uint8_t *f;
 192         uint8_t *t;
 193
 194         r = new(char, (length*3+1)/2 + 1);
 195         if (!r)
 196                 return NULL;
 197
 198         t = (uint8_t*) r;
 199
 200         for (f = s; f < (const uint8_t*) s + length; f += 2) {
 201                 uint16_t c;
 202
 203                 c = (f[1] << 8) | f[0];
 204
 205                 if (c == 0) {
 206                         *t = 0;
 207                         return r;
 208                 } else if (c < 0x80) {
 209                         *(t++) = (uint8_t) c;
 210                 } else if (c < 0x800) {
 211                         *(t++) = (uint8_t) (0xc0 | (c >> 6));
 212                         *(t++) = (uint8_t) (0x80 | (c & 0x3f));
 213                 } else {
 214                         *(t++) = (uint8_t) (0xe0 | (c >> 12));
 215                         *(t++) = (uint8_t) (0x80 | ((c >> 6) & 0x3f));
 216                         *(t++) = (uint8_t) (0x80 | (c & 0x3f));
 217                 }
 218         }
 219
 220         *t = 0;
 221
 222         return r;
 223 }
 224
 225 /* expected size used to encode one unicode char */
 226 static int utf8_unichar_to_encoded_len(int unichar) {
 227         if (unichar < 0x80)
 228                 return 1;
 229         if (unichar < 0x800)
 230                 return 2;
 231         if (unichar < 0x10000)
 232                 return 3;
 233         if (unichar < 0x200000)
 234                 return 4;
 235         if (unichar < 0x4000000)
 236                 return 5;
 237         return 6;
 238 }
 239
 240 /* validate one encoded unicode char and return its length */
 241 int utf8_encoded_valid_unichar(const char *str) {
 242         int len;
 243         int unichar;
 244         int i;
 245
 246         len = utf8_encoded_expected_len(str);
 247         if (len == 0)
 248                 return -1;
 249
 250         /* ascii is valid */
 251         if (len == 1)
 252                 return 1;
 253
 254         /* check if expected encoded chars are available */
 255         for (i = 0; i < len; i++)
 256                 if ((str[i] & 0x80) != 0x80)
 257                         return -1;
 258
 259         unichar = utf8_encoded_to_unichar(str);
 260
 261         /* check if encoded length matches encoded value */
 262         if (utf8_unichar_to_encoded_len(unichar) != len)
 263                 return -1;
 264
 265         /* check if value has valid range */
 266         if (!is_unicode_valid(unichar))
 267                 return -1;
 268
 269         return len;
 270 }