udev_utils_string.c

   1 /*
   2  * udev_utils_string.c - string manipulation
   3  *
   4  * Copyright (C) 2004-2005 Kay Sievers <kay.sievers@vrfy.org>
   5  *
   6  *      This program is free software; you can redistribute it and/or modify it
   7  *      under the terms of the GNU General Public License as published by the
   8  *      Free Software Foundation version 2 of the License.
   9  *
  10  *      This program is distributed in the hope that it will be useful, but
  11  *      WITHOUT ANY WARRANTY; without even the implied warranty of
  12  *      MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13  *      General Public License for more details.
  14  *
  15  *      You should have received a copy of the GNU General Public License along
  16  *      with this program; if not, write to the Free Software Foundation, Inc.,
  17  *      675 Mass Ave, Cambridge, MA 02139, USA.
  18  *
  19  */
  20
  21
  22 #include <stdlib.h>
  23 #include <stdio.h>
  24 #include <stddef.h>
  25 #include <unistd.h>
  26 #include <fcntl.h>
  27 #include <errno.h>
  28 #include <ctype.h>
  29 #include <dirent.h>
  30 #include <syslog.h>
  31 #include <sys/utsname.h>
  32
  33 #include "udev.h"
  34
  35 /* compare string with pattern (like fnmatch(), supports * ? [0-9] [!A-Z]) */
  36 int strcmp_pattern(const char *p, const char *s)
  37 {
  38         if (s[0] == '\0') {
  39                 while (p[0] == '*')
  40                         p++;
  41                 return (p[0] != '\0');
  42         }
  43         switch (p[0]) {
  44         case '[':
  45                 {
  46                         int not = 0;
  47                         p++;
  48                         if (p[0] == '!') {
  49                                 not = 1;
  50                                 p++;
  51                         }
  52                         while ((p[0] != '\0') && (p[0] != ']')) {
  53                                 int match = 0;
  54                                 if (p[1] == '-') {
  55                                         if ((s[0] >= p[0]) && (s[0] <= p[2]))
  56                                                 match = 1;
  57                                         p += 3;
  58                                 } else {
  59                                         match = (p[0] == s[0]);
  60                                         p++;
  61                                 }
  62                                 if (match ^ not) {
  63                                         while ((p[0] != '\0') && (p[0] != ']'))
  64                                                 p++;
  65                                         if (p[0] == ']')
  66                                                 return strcmp_pattern(p+1, s+1);
  67                                 }
  68                         }
  69                 }
  70                 break;
  71         case '*':
  72                 if (strcmp_pattern(p, s+1))
  73                         return strcmp_pattern(p+1, s);
  74                 return 0;
  75         case '\0':
  76                 if (s[0] == '\0') {
  77                         return 0;
  78                 }
  79                 break;
  80         default:
  81                 if ((p[0] == s[0]) || (p[0] == '?'))
  82                         return strcmp_pattern(p+1, s+1);
  83                 break;
  84         }
  85         return 1;
  86 }
  87
  88 int string_is_true(const char *str)
  89 {
  90         if (strcasecmp(str, "true") == 0)
  91                 return 1;
  92         if (strcasecmp(str, "yes") == 0)
  93                 return 1;
  94         if (strcasecmp(str, "1") == 0)
  95                 return 1;
  96         return 0;
  97 }
  98
  99 void remove_trailing_chars(char *path, char c)
 100 {
 101         size_t len;
 102
 103         len = strlen(path);
 104         while (len > 0 && path[len-1] == c)
 105                 path[--len] = '\0';
 106 }
 107
 108 /* count of characters used to encode one unicode char */
 109 static int utf8_encoded_expected_len(const char *str)
 110 {
 111         unsigned char c = (unsigned char)str[0];
 112
 113         if (c < 0x80)
 114                 return 1;
 115         if ((c & 0xe0) == 0xc0)
 116                 return 2;
 117         if ((c & 0xf0) == 0xe0)
 118                 return 3;
 119         if ((c & 0xf8) == 0xf0)
 120                 return 4;
 121         if ((c & 0xfc) == 0xf8)
 122                 return 5;
 123         if ((c & 0xfe) == 0xfc)
 124                 return 6;
 125         return 0;
 126 }
 127
 128 /* decode one unicode char */
 129 static int utf8_encoded_to_unichar(const char *str)
 130 {
 131         int unichar;
 132         int len;
 133         int i;
 134
 135         len = utf8_encoded_expected_len(str);
 136         switch (len) {
 137         case 1:
 138                 return (int)str[0];
 139         case 2:
 140                 unichar = str[0] & 0x1f;
 141                 break;
 142         case 3:
 143                 unichar = (int)str[0] & 0x0f;
 144                 break;
 145         case 4:
 146                 unichar = (int)str[0] & 0x07;
 147                 break;
 148         case 5:
 149                 unichar = (int)str[0] & 0x03;
 150                 break;
 151         case 6:
 152                 unichar = (int)str[0] & 0x01;
 153                 break;
 154         default:
 155                 return -1;
 156         }
 157
 158         for (i = 1; i < len; i++) {
 159                 if (((int)str[i] & 0xc0) != 0x80)
 160                         return -1;
 161                 unichar <<= 6;
 162                 unichar |= (int)str[i] & 0x3f;
 163         }
 164
 165         return unichar;
 166 }
 167
 168 /* expected size used to encode one unicode char */
 169 static int utf8_unichar_to_encoded_len(int unichar)
 170 {
 171         if (unichar < 0x80)
 172                 return 1;
 173         if (unichar < 0x800)
 174                 return 2;
 175         if (unichar < 0x10000)
 176                 return 3;
 177         if (unichar < 0x200000)
 178                 return 4;
 179         if (unichar < 0x4000000)
 180                 return 5;
 181         return 6;
 182 }
 183
 184 /* check if unicode char has a valid numeric range */
 185 static int utf8_unichar_valid_range(int unichar)
 186 {
 187         if (unichar > 0x10ffff)
 188                 return 0;
 189         if ((unichar & 0xfffff800) == 0xd800)
 190                 return 0;
 191         if ((unichar > 0xfdcf) && (unichar < 0xfdf0))
 192                 return 0;
 193         if ((unichar & 0xffff) == 0xffff)
 194                 return 0;
 195         return 1;
 196 }
 197
 198 /* validate one encoded unicode char and return its length */
 199 int utf8_encoded_valid_unichar(const char *str)
 200 {
 201         int len;
 202         int unichar;
 203         int i;
 204
 205         len = utf8_encoded_expected_len(str);
 206         if (len == 0)
 207                 return -1;
 208
 209         /* ascii is valid */
 210         if (len == 1)
 211                 return 1;
 212
 213         /* check if expected encoded chars are available */
 214         for (i = 0; i < len; i++)
 215                 if ((str[i] & 0x80) != 0x80)
 216                         return -1;
 217
 218         unichar = utf8_encoded_to_unichar(str);
 219
 220         /* check if encoded length matches encoded value */
 221         if (utf8_unichar_to_encoded_len(unichar) != len)
 222                 return -1;
 223
 224         /* check if value has valid range */
 225         if (!utf8_unichar_valid_range(unichar))
 226                 return -1;
 227
 228         return len;
 229 }
 230
 231 /* replace everything but whitelisted plain ascii and valid utf8 */
 232 int replace_untrusted_chars(char *str)
 233 {
 234         size_t i = 0;
 235         int replaced = 0;
 236
 237         while (str[i] != '\0') {
 238                 int len;
 239
 240                 /* valid printable ascii char */
 241                 if ((str[i] >= '0' && str[i] <= '9') ||
 242                     (str[i] >= 'A' && str[i] <= 'Z') ||
 243                     (str[i] >= 'a' && str[i] <= 'z') ||
 244                     strchr(" #$%+-./:=?@_", str[i])) {
 245                         i++;
 246                         continue;
 247                 }
 248                 /* valid utf8 is accepted */
 249                 len = utf8_encoded_valid_unichar(&str[i]);
 250                 if (len > 1) {
 251                         i += len;
 252                         continue;
 253                 }
 254
 255                 /* everything else is garbage */
 256                 str[i] = '_';
 257                 i++;
 258                 replaced++;
 259         }
 260
 261         return replaced;
 262 }