locale/weight.h

   1 /* Copyright (C) 1996, 1997, 1998, 1999 Free Software Foundation, Inc.
   2    This file is part of the GNU C Library.
   3    Written by Ulrich Drepper, <drepper@cygnus.com>.
   4
   5    The GNU C Library is free software; you can redistribute it and/or
   6    modify it under the terms of the GNU Library General Public License as
   7    published by the Free Software Foundation; either version 2 of the
   8    License, or (at your option) any later version.
   9
  10    The GNU C Library is distributed in the hope that it will be useful,
  11    but WITHOUT ANY WARRANTY; without even the implied warranty of
  12    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  13    Library General Public License for more details.
  14
  15    You should have received a copy of the GNU Library General Public
  16    License along with the GNU C Library; see the file COPYING.LIB.  If not,
  17    write to the Free Software Foundation, Inc., 59 Temple Place - Suite 330,
  18    Boston, MA 02111-1307, USA.  */
  19
  20 #include <alloca.h>
  21 #include <errno.h>
  22 #include <langinfo.h>
  23 #include "localeinfo.h"
  24
  25 #ifndef STRING_TYPE
  26 # error STRING_TYPE not defined
  27 #endif
  28
  29 #ifndef USTRING_TYPE
  30 # error USTRING_TYPE not defined
  31 #endif
  32
  33 typedef struct weight_t
  34 {
  35   struct weight_t *prev;
  36   struct weight_t *next;
  37   struct data_pair
  38     {
  39       int number;
  40       const uint32_t *value;
  41     } data[0];
  42 } weight_t;
  43
  44
  45 /* The following five macros grant access to the values in the
  46    collate locale file that do not depend on byte order.  */
  47 #ifndef USE_IN_EXTENDED_LOCALE_MODEL
  48 # define collate_nrules \
  49   (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_NRULES))
  50 # define collate_hash_size \
  51   (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_HASH_SIZE))
  52 # define collate_hash_layers \
  53   (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_HASH_LAYERS))
  54 # define collate_undefined \
  55   (_NL_CURRENT_WORD (LC_COLLATE, _NL_COLLATE_UNDEFINED_WC))
  56 # define collate_rules \
  57   ((uint32_t *) _NL_CURRENT (LC_COLLATE, _NL_COLLATE_RULES))
  58
  59 static __inline void get_weight (const STRING_TYPE **str, weight_t *result);
  60 static __inline void
  61 get_weight (const STRING_TYPE **str, weight_t *result)
  62 #else
  63 # define collate_nrules \
  64   current->values[_NL_ITEM_INDEX (_NL_COLLATE_NRULES)].word
  65 # define collate_hash_size \
  66   current->values[_NL_ITEM_INDEX (_NL_COLLATE_HASH_SIZE)].word
  67 # define collate_hash_layers \
  68   current->values[_NL_ITEM_INDEX (_NL_COLLATE_HASH_LAYERS)].word
  69 # define collate_undefined \
  70   current->values[_NL_ITEM_INDEX (_NL_COLLATE_UNDEFINED_WC)].word
  71 # define collate_rules \
  72   ((uint32_t *) current->values[_NL_ITEM_INDEX (_NL_COLLATE_RULES)].string)
  73
  74 static __inline void get_weight (const STRING_TYPE **str, weight_t *result,
  75                                  struct locale_data *current,
  76                                  const uint32_t *__collate_tablewc,
  77                                  const uint32_t *__collate_extrawc);
  78 static __inline void
  79 get_weight (const STRING_TYPE **str, weight_t *result,
  80             struct locale_data *current, const uint32_t *__collate_tablewc,
  81             const uint32_t *__collate_extrawc)
  82 #endif
  83 {
  84   unsigned int ch = *((USTRING_TYPE *) (*str))++;
  85   size_t slot;
  86
  87   if (sizeof (STRING_TYPE) == 1)
  88     slot = ch * (collate_nrules + 1);
  89   else
  90     {
  91       const size_t level_size = collate_hash_size * (collate_nrules + 1);
  92       size_t level;
  93
  94       slot = (ch % collate_hash_size) * (collate_nrules + 1);
  95
  96       level = 0;
  97       while (__collate_tablewc[slot] != (uint32_t) ch)
  98         {
  99           if (__collate_tablewc[slot + 1] == 0
 100               || ++level >= collate_hash_layers)
 101             {
 102               size_t idx = collate_undefined;
 103               size_t cnt;
 104
 105               for (cnt = 0; cnt < collate_nrules; ++cnt)
 106                 {
 107                   result->data[cnt].number = __collate_extrawc[idx++];
 108                   result->data[cnt].value = &__collate_extrawc[idx];
 109                   idx += result->data[cnt].number;
 110                 }
 111               /* The Unix standard requires that a character outside
 112                  the domain is signalled by setting `errno'.  */
 113               __set_errno (EINVAL);
 114               return;
 115             }
 116           slot += level_size;
 117         }
 118     }
 119
 120   if (__collate_tablewc[slot + 1] != (uint32_t) FORWARD_CHAR)
 121     {
 122       /* We have a simple form.  One value for each weight.  */
 123       size_t cnt;
 124
 125       for (cnt = 0; cnt < collate_nrules; ++cnt)
 126         {
 127           result->data[cnt].number = 1;
 128           result->data[cnt].value = &__collate_tablewc[slot + 1 + cnt];
 129         }
 130       return;
 131     }
 132
 133   /* We now look for any collation element which starts with CH.
 134      There might none, but the last list member is a catch-all case
 135      because it is simple the character CH.  The value of this entry
 136      might be the same as UNDEFINED.  */
 137   slot = __collate_tablewc[slot + 2];
 138
 139   while (1)
 140     {
 141       size_t idx;
 142
 143       /* This is a comparison between a uint32_t array (aka wchar_t) and
 144          an 8-bit string.  */
 145       for (idx = 0; __collate_extrawc[slot + 2 + idx] != 0; ++idx)
 146         if (__collate_extrawc[slot + 2 + idx] != (uint32_t) (*str)[idx])
 147           break;
 148
 149       /* When the loop finished with all character of the collation
 150          element used, we found the longest prefix.  */
 151       if (__collate_extrawc[slot + 2 + idx] == 0)
 152         {
 153           size_t cnt;
 154
 155           *str += idx;
 156           idx += slot + 3;
 157           for (cnt = 0; cnt < collate_nrules; ++cnt)
 158             {
 159               result->data[cnt].number = __collate_extrawc[idx++];
 160               result->data[cnt].value = &__collate_extrawc[idx];
 161               idx += result->data[cnt].number;
 162             }
 163           return;
 164         }
 165
 166       /* To next entry in list.  */
 167       slot += __collate_extrawc[slot];
 168     }
 169 }
 170
 171
 172 /* To process a string efficiently we retrieve all information about
 173    the string at once.  The following macro constructs a double linked
 174    list of this information.  It is a macro because we use `alloca'
 175    and we use a double linked list because of the backward collation
 176    order.
 177
 178    We have this strange extra macro since the functions which use the
 179    given locale (not the global one) cannot use the global tables.  */
 180 #ifndef USE_IN_EXTENDED_LOCALE_MODEL
 181 # define call_get_weight(strp, newp) get_weight ((strp), (newp))
 182 #else
 183 # define call_get_weight(strp, newp) \
 184   get_weight ((strp), (newp), current, collate_table, collate_extra)
 185 #endif
 186
 187 #define get_string(str, forw, backw) \
 188   do                                                                          \
 189     {                                                                         \
 190       weight_t *newp;                                                         \
 191       while (*str != '\0')                                                    \
 192         {                                                                     \
 193           newp = (weight_t *) alloca (sizeof (weight_t)                       \
 194                                       + (collate_nrules                       \
 195                                          * sizeof (struct data_pair)));       \
 196                                                                               \
 197           newp->prev = backw;                                                 \
 198           if (backw == NULL)                                                  \
 199             forw = newp;                                                      \
 200           else                                                                \
 201             backw->next = newp;                                               \
 202           newp->next = NULL;                                                  \
 203           backw = newp;                                                       \
 204           call_get_weight (&str, newp);                                       \
 205         }                                                                     \
 206     }                                                                         \
 207   while (0)