code.delx.au - pulseaudio/blob - src/polypcore/utf8.c

   1 /* $Id */
   2
   3 /* This file is based on the GLIB utf8 validation functions. The
   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
  12  * modify it under the terms of the GNU Lesser General Public
  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
  18  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  19  * Lesser General Public License for more details.
  20  *
  21  * You should have received a copy of the GNU Lesser General Public
  22  * License along with this library; if not, write to the
  23  * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  24  * Boston, MA 02111-1307, USA.
  25  */
  26
  27 #ifdef HAVE_CONFIG_H
  28 #include <config.h>
  29 #endif
  30
  31 #include <assert.h>
  32 #include <stdlib.h>
  33 #include <inttypes.h>
  34 #include <string.h>
  35
  36 #include <polyp/xmalloc.h>
  37
  38 #include "utf8.h"
  39
  40 #define FILTER_CHAR '_'
  41
  42 static inline int is_unicode_valid(uint32_t ch) {
  43     if (ch >= 0x110000) /* End of unicode space */
  44         return 0;
  45     if ((ch & 0xFFFFF800) == 0xD800) /* Reserved area for UTF-16 */
  46         return 0;
  47     if ((ch >= 0xFDD0) && (ch <= 0xFDEF)) /* Reserved */
  48         return 0;
  49     if ((ch & 0xFFFE) == 0xFFFE) /* BOM (Byte Order Mark) */
  50         return 0;
  51     return 1;
  52 }
  53
  54 static inline int is_continuation_char(uint8_t ch) {
  55     if ((ch & 0xc0) != 0x80) /* 10xxxxxx */
  56         return 0;
  57     return 1;
  58 }
  59
  60 static inline void merge_continuation_char(uint32_t *u_ch, uint8_t ch) {
  61     *u_ch <<= 6;
  62     *u_ch |= ch & 0x3f;
  63 }
  64
  65 static char* utf8_validate(const char *str, char *output) {
  66     uint32_t val = 0;
  67     uint32_t min = 0;
  68     const uint8_t *p, *last;
  69     int size;
  70     uint8_t *o;
  71
  72     o = (uint8_t*) output;
  73     for (p = (const uint8_t*) str; *p; p++) {
  74         if (*p < 128) {
  75             if (o)
  76                 *o = *p;
  77         } else {
  78             last = p;
  79
  80             if ((*p & 0xe0) == 0xc0) { /* 110xxxxx two-char seq. */
  81                 size = 2;
  82                 min = 128;
  83                 val = *p & 0x1e;
  84                 goto ONE_REMAINING;
  85             } else if ((*p & 0xf0) == 0xe0) { /* 1110xxxx three-char seq.*/
  86                 size = 3;
  87                 min = (1 << 11);
  88                 val = *p & 0x0f;
  89                 goto TWO_REMAINING;
  90             } else if ((*p & 0xf8) == 0xf0) { /* 11110xxx four-char seq */
  91                 size = 4;
  92                 min = (1 << 16);
  93                 val = *p & 0x07;
  94             } else {
  95                 size = 1;
  96                 goto error;
  97             }
  98
  99             p++;
 100             if (!is_continuation_char(*p))
 101                 goto error;
 102             merge_continuation_char(&val, *p);
 103
 104 TWO_REMAINING:
 105             p++;
 106             if (!is_continuation_char(*p))
 107                 goto error;
 108             merge_continuation_char(&val, *p);
 109
 110 ONE_REMAINING:
 111             p++;
 112             if (!is_continuation_char(*p))
 113                 goto error;
 114             merge_continuation_char(&val, *p);
 115
 116             if (val < min)
 117                 goto error;
 118
 119             if (!is_unicode_valid(val))
 120                 goto error;
 121
 122             if (o) {
 123                 memcpy(o, last, size);
 124                 o += size - 1;
 125             }
 126
 127             if (o)
 128                 o++;
 129
 130             continue;
 131
 132 error:
 133             if (o) {
 134                 *o = FILTER_CHAR;
 135                 p = last; /* We retry at the next character */
 136             } else
 137                 goto failure;
 138         }
 139
 140         if (o)
 141             o++;
 142     }
 143
 144     if (o) {
 145         *o = '\0';
 146         return output;
 147     }
 148
 149     return (char*) str;
 150
 151 failure:
 152     return NULL;
 153 }
 154
 155 const char* pa_utf8_valid (const char *str) {
 156     return utf8_validate(str, NULL);
 157 }
 158
 159 char* pa_utf8_filter (const char *str) {
 160     char *new_str;
 161
 162     new_str = pa_xnew(char, strlen(str) + 1);
 163
 164     return utf8_validate(str, new_str);
 165 }