src/intl/gettext/explodename.c

   1 /* Copyright (C) 1995-1998, 2000, 2001 Free Software Foundation, Inc.
   2    Contributed by Ulrich Drepper <drepper@gnu.ai.mit.edu>, 1995.
   3
   4    This program is free software; you can redistribute it and/or modify
   5    it under the terms of the GNU General Public License as published by
   6    the Free Software Foundation; either version 2, or (at your option)
   7    any later version.
   8
   9    This program is distributed in the hope that it will be useful,
  10    but WITHOUT ANY WARRANTY; without even the implied warranty of
  11    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12    GNU General Public License for more details.
  13
  14    You should have received a copy of the GNU General Public License
  15    along with this program; if not, write to the Free Software Foundation,
  16    Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.  */
  17
  18 #ifdef HAVE_CONFIG_H
  19 #include "config.h"
  20 #endif
  21
  22 #include <stdlib.h>
  23 #include <string.h>
  24 #include <sys/types.h>
  25
  26 #include "elinks.h"
  27
  28 #include "intl/gettext/loadinfo.h"
  29
  30 /* On some strange systems still no definition of NULL is found.  Sigh!  */
  31 #ifndef NULL
  32 #if defined __STDC__ && __STDC__
  33 #define NULL ((void *) 0)
  34 #else
  35 #define NULL 0
  36 #endif
  37 #endif
  38
  39 /* @@ end of prolog @@ */
  40
  41 unsigned char *
  42 _nl_find_language(const unsigned char *name)
  43 {
  44         while (name[0] != '\0' && name[0] != '_' && name[0] != '@'
  45                && name[0] != '+' && name[0] != ',')
  46                 ++name;
  47
  48         return (unsigned char *) name;
  49 }
  50
  51 int
  52 _nl_explode_name(unsigned char *name, const unsigned char **language, const unsigned char **modifier,
  53                  const unsigned char **territory, const unsigned char **codeset,
  54                  const unsigned char **normalized_codeset, const unsigned char **special,
  55                  const unsigned char **sponsor, const unsigned char **revision)
  56 {
  57         enum { undecided, xpg, cen } syntax;
  58         unsigned char *cp;
  59         int mask;
  60
  61         *modifier = NULL;
  62         *territory = NULL;
  63         *codeset = NULL;
  64         *normalized_codeset = NULL;
  65         *special = NULL;
  66         *sponsor = NULL;
  67         *revision = NULL;
  68
  69         /* Now we determine the single parts of the locale name.  First
  70            look for the language.  Termination symbols are `_' and `@' if
  71            we use XPG4 style, and `_', `+', and `,' if we use CEN syntax.  */
  72         mask = 0;
  73         syntax = undecided;
  74         *language = cp = name;
  75         cp = _nl_find_language(*language);
  76
  77         if (*language == cp)
  78                 /* This does not make sense: language has to be specified.  Use
  79                    this entry as it is without exploding.  Perhaps it is an alias.  */
  80                 cp = strchr(*language, '\0');
  81         else if (cp[0] == '_') {
  82                 /* Next is the territory.  */
  83                 cp[0] = '\0';
  84                 *territory = ++cp;
  85
  86                 while (cp[0] != '\0' && cp[0] != '.' && cp[0] != '@'
  87                        && cp[0] != '+' && cp[0] != ',' && cp[0] != '_')
  88                         ++cp;
  89
  90                 mask |= TERRITORY;
  91
  92                 if (cp[0] == '.') {
  93                         /* Next is the codeset.  */
  94                         syntax = xpg;
  95                         cp[0] = '\0';
  96                         *codeset = ++cp;
  97
  98                         while (cp[0] != '\0' && cp[0] != '@')
  99                                 ++cp;
 100
 101                         mask |= XPG_CODESET;
 102
 103                         if (*codeset != cp && (*codeset)[0] != '\0') {
 104                                 *normalized_codeset =
 105                                         _nl_normalize_codeset(*codeset,
 106                                                               cp - *codeset);
 107                                 if (strcmp(*codeset, *normalized_codeset) == 0)
 108                                         free((unsigned char *) *normalized_codeset);
 109                                 else
 110                                         mask |= XPG_NORM_CODESET;
 111                         }
 112                 }
 113         }
 114
 115         if (cp[0] == '@' || (syntax != xpg && cp[0] == '+')) {
 116                 /* Next is the modifier.  */
 117                 syntax = cp[0] == '@' ? xpg : cen;
 118                 cp[0] = '\0';
 119                 *modifier = ++cp;
 120
 121                 while (syntax == cen && cp[0] != '\0' && cp[0] != '+'
 122                        && cp[0] != ',' && cp[0] != '_')
 123                         ++cp;
 124
 125                 mask |= XPG_MODIFIER | CEN_AUDIENCE;
 126         }
 127
 128         if (syntax != xpg && (cp[0] == '+' || cp[0] == ',' || cp[0] == '_')) {
 129                 syntax = cen;
 130
 131                 if (cp[0] == '+') {
 132                         /* Next is special application (CEN syntax).  */
 133                         cp[0] = '\0';
 134                         *special = ++cp;
 135
 136                         while (cp[0] != '\0' && cp[0] != ',' && cp[0] != '_')
 137                                 ++cp;
 138
 139                         mask |= CEN_SPECIAL;
 140                 }
 141
 142                 if (cp[0] == ',') {
 143                         /* Next is sponsor (CEN syntax).  */
 144                         cp[0] = '\0';
 145                         *sponsor = ++cp;
 146
 147                         while (cp[0] != '\0' && cp[0] != '_')
 148                                 ++cp;
 149
 150                         mask |= CEN_SPONSOR;
 151                 }
 152
 153                 if (cp[0] == '_') {
 154                         /* Next is revision (CEN syntax).  */
 155                         cp[0] = '\0';
 156                         *revision = ++cp;
 157
 158                         mask |= CEN_REVISION;
 159                 }
 160         }
 161
 162         /* For CEN syntax values it might be important to have the
 163            separator character in the file name, not for XPG syntax.  */
 164         if (syntax == xpg) {
 165                 if (*territory != NULL && (*territory)[0] == '\0')
 166                         mask &= ~TERRITORY;
 167
 168                 if (*codeset != NULL && (*codeset)[0] == '\0')
 169                         mask &= ~XPG_CODESET;
 170
 171                 if (*modifier != NULL && (*modifier)[0] == '\0')
 172                         mask &= ~XPG_MODIFIER;
 173         }
 174
 175         return mask;
 176 }