extensions/spellcheck/hunspell/src/dictmgr.cpp

   1 /******* BEGIN LICENSE BLOCK *******
   2  * Version: MPL 1.1/GPL 2.0/LGPL 2.1
   3  *
   4  * The contents of this file are subject to the Mozilla Public License Version
   5  * 1.1 (the "License"); you may not use this file except in compliance with
   6  * the License. You may obtain a copy of the License at
   7  * http://www.mozilla.org/MPL/
   8  *
   9  * Software distributed under the License is distributed on an "AS IS" basis,
  10  * WITHOUT WARRANTY OF ANY KIND, either express or implied. See the License
  11  * for the specific language governing rights and limitations under the
  12  * License.
  13  *
  14  * The Initial Developers of the Original Code are Kevin Hendricks (MySpell)
  15  * and László Németh (Hunspell). Portions created by the Initial Developers
  16  * are Copyright (C) 2002-2005 the Initial Developers. All Rights Reserved.
  17  *
  18  * Contributor(s): László Németh (nemethl@gyorsposta.hu)
  19  *
  20  * Alternatively, the contents of this file may be used under the terms of
  21  * either the GNU General Public License Version 2 or later (the "GPL"), or
  22  * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
  23  * in which case the provisions of the GPL or the LGPL are applicable instead
  24  * of those above. If you wish to allow use of your version of this file only
  25  * under the terms of either the GPL or the LGPL, and not to allow others to
  26  * use your version of this file under the terms of the MPL, indicate your
  27  * decision by deleting the provisions above and replace them with the notice
  28  * and other provisions required by the GPL or the LGPL. If you do not delete
  29  * the provisions above, a recipient may use your version of this file under
  30  * the terms of any one of the MPL, the GPL or the LGPL.
  31  *
  32  ******* END LICENSE BLOCK *******/
  33
  34 #ifndef MOZILLA_CLIENT
  35 #include <cstdlib>
  36 #include <cstring>
  37 #include <cstdio>
  38 #include <cctype>
  39 #else
  40 #include <stdlib.h>
  41 #include <string.h>
  42 #include <stdio.h>
  43 #include <ctype.h>
  44 #endif
  45
  46 #include "dictmgr.hxx"
  47
  48 #ifndef MOZILLA_CLIENT
  49 #ifndef W32
  50 using namespace std;
  51 #endif
  52 #endif
  53
  54 DictMgr::DictMgr(const char * dictpath, const char * etype)
  55 {
  56   // load list of etype entries
  57   numdict = 0;
  58   pdentry = (dictentry *)malloc(MAXDICTIONARIES*sizeof(struct dictentry));
  59   if (pdentry) {
  60      if (parse_file(dictpath, etype)) {
  61         numdict = 0;
  62         // no dictionary.lst found is okay
  63      }
  64   } else {
  65      numdict = 0;
  66   }
  67 }
  68
  69
  70 DictMgr::~DictMgr()
  71 {
  72   dictentry * pdict = NULL;
  73   if (pdentry) {
  74      pdict = pdentry;
  75      for (int i=0;i<numdict;i++) {
  76         if (pdict->lang) {
  77             free(pdict->lang);
  78             pdict->lang = NULL;
  79         }
  80         if (pdict->region) {
  81             free(pdict->region);
  82             pdict->region=NULL;
  83         }
  84         if (pdict->filename) {
  85             free(pdict->filename);
  86             pdict->filename = NULL;
  87         }
  88         pdict++;
  89      }
  90      free(pdentry);
  91      pdentry = NULL;
  92      pdict = NULL;
  93   }
  94   numdict = 0;
  95 }
  96
  97
  98 // read in list of etype entries and build up structure to describe them
  99 int  DictMgr::parse_file(const char * dictpath, const char * etype)
 100 {
 101
 102     int i;
 103     char line[MAXDICTENTRYLEN+1];
 104     dictentry * pdict = pdentry;
 105
 106     // open the dictionary list file
 107     FILE * dictlst;
 108     dictlst = fopen(dictpath,"r");
 109     if (!dictlst) {
 110       return 1;
 111     }
 112
 113     // step one is to parse the dictionary list building up the
 114     // descriptive structures
 115
 116     // read in each line ignoring any that dont start with etype
 117     while (fgets(line,MAXDICTENTRYLEN,dictlst)) {
 118        mychomp(line);
 119
 120        /* parse in a dictionary entry */
 121        if (strncmp(line,etype,4) == 0) {
 122           if (numdict < MAXDICTIONARIES) {
 123              char * tp = line;
 124              char * piece;
 125              i = 0;
 126              while ((piece=mystrsep(&tp,' '))) {
 127                 if (*piece != '\0') {
 128                     switch(i) {
 129                        case 0: break;
 130                        case 1: pdict->lang = mystrdup(piece); break;
 131                        case 2: if (strcmp (piece, "ANY") == 0)
 132                                  pdict->region = mystrdup("");
 133                                else
 134                                  pdict->region = mystrdup(piece);
 135                                break;
 136                        case 3: pdict->filename = mystrdup(piece); break;
 137                        default: break;
 138                     }
 139                     i++;
 140                 }
 141                 free(piece);
 142              }
 143              if (i == 4) {
 144                  numdict++;
 145                  pdict++;
 146              } else {
 147                  fprintf(stderr,"dictionary list corruption in line \"%s\"\n",line);
 148                  fflush(stderr);
 149              }
 150           }
 151        }
 152     }
 153     fclose(dictlst);
 154     return 0;
 155 }
 156
 157 // return text encoding of dictionary
 158 int DictMgr::get_list(dictentry ** ppentry)
 159 {
 160   *ppentry = pdentry;
 161   return numdict;
 162 }
 163
 164
 165
 166 // strip strings into token based on single char delimiter
 167 // acts like strsep() but only uses a delim char and not
 168 // a delim string
 169
 170 char * DictMgr::mystrsep(char ** stringp, const char delim)
 171 {
 172   char * rv = NULL;
 173   char * mp = *stringp;
 174   int n = strlen(mp);
 175   if (n > 0) {
 176      char * dp = (char *)memchr(mp,(int)((unsigned char)delim),n);
 177      if (dp) {
 178         *stringp = dp+1;
 179         int nc = (int)((unsigned long)dp - (unsigned long)mp);
 180         rv = (char *) malloc(nc+1);
 181         if (rv) {
 182            memcpy(rv,mp,nc);
 183            *(rv+nc) = '\0';
 184            return rv;
 185         }
 186      } else {
 187        rv = (char *) malloc(n+1);
 188        if (rv) {
 189           memcpy(rv, mp, n);
 190           *(rv+n) = '\0';
 191           *stringp = mp + n;
 192           return rv;
 193        }
 194      }
 195   }
 196   return NULL;
 197 }
 198
 199
 200 // replaces strdup with ansi version
 201 char * DictMgr::mystrdup(const char * s)
 202 {
 203   char * d = NULL;
 204   if (s) {
 205      int sl = strlen(s);
 206      d = (char *) malloc(((sl+1) * sizeof(char)));
 207      if (d) memcpy(d,s,((sl+1)*sizeof(char)));
 208   }
 209   return d;
 210 }
 211
 212
 213 // remove cross-platform text line end characters
 214 void DictMgr:: mychomp(char * s)
 215 {
 216   int k = strlen(s);
 217   if ((k > 0) && ((*(s+k-1)=='\r') || (*(s+k-1)=='\n'))) *(s+k-1) = '\0';
 218   if ((k > 1) && (*(s+k-2) == '\r')) *(s+k-2) = '\0';
 219 }