/extensions/spellcheck/hunspell/src/phonet.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 Developer of the Original Code is Bj�rn Jacke. Portions created
 15 * by the Initial Developers are Copyright (C) 2000-2007 the Initial
 16 * Developers. All Rights Reserved.
 17 * 
 18 * Contributor(s): Bj�rn Jacke (bjoern.jacke@gmx.de)
 19 *                 L�szl� N�meth (nemethl@gyorsposta.hu)
 20 *                 Caolan McNamara (caolanm@redhat.com)
 21 * 
 22 * Alternatively, the contents of this file may be used under the terms of
 23 * either the GNU General Public License Version 2 or later (the "GPL"), or
 24 * the GNU Lesser General Public License Version 2.1 or later (the "LGPL"),
 25 * in which case the provisions of the GPL or the LGPL are applicable instead
 26 * of those above. If you wish to allow use of your version of this file only
 27 * under the terms of either the GPL or the LGPL, and not to allow others to
 28 * use your version of this file under the terms of the MPL, indicate your
 29 * decision by deleting the provisions above and replace them with the notice
 30 * and other provisions required by the GPL or the LGPL. If you do not delete
 31 * the provisions above, a recipient may use your version of this file under
 32 * the terms of any one of the MPL, the GPL or the LGPL.
 33 *
 34 * Changelog:
 35 *  2000-01-05  Bj�rn Jacke <bjoern.jacke AT gmx.de>
 36 *              Initial Release insprired by the article about phonetic
 37 *              transformations out of c't 25/1999
 38 *
 39 *  2007-07-26  Bj�rn Jacke <bjoern.jacke AT gmx.de>
 40 *              Released under MPL/GPL/LGPL tri-license for Hunspell
 41 *
 42 *  2007-08-23  L�szl� N�meth <nemeth at OOo>
 43 *              Porting from Aspell to Hunspell using C-like structs
 44 *
 45 ******* END LICENSE BLOCK *******/
 46
 47#include <stdlib.h> 
 48#include <string.h>
 49#include <stdio.h> 
 50#include <ctype.h>
 51
 52#include "csutil.hxx"
 53#include "phonet.hxx"
 54
 55void init_phonet_hash(phonetable & parms) 
 56  {
 57    int i, k;
 58
 59    for (i = 0; i < HASHSIZE; i++) {
 60      parms.hash[i] = -1;
 61    }
 62
 63    for (i = 0; parms.rules[i][0] != '\0'; i += 2) {
 64      /**  set hash value  **/
 65      k = (unsigned char) parms.rules[i][0];
 66
 67      if (parms.hash[k] < 0) {
 68	parms.hash[k] = i;
 69      }
 70    }
 71  }
 72
 73// like strcpy but safe if the strings overlap
 74//   but only if dest < src
 75static inline void strmove(char * dest, char * src) {
 76  while (*src) 
 77    *dest++ = *src++;
 78  *dest = '\0';
 79}
 80
 81static int myisalpha(char ch) {
 82  if ((unsigned char) ch < 128) return isalpha(ch);
 83  return 1;
 84}
 85
 86/*  phonetic transcription algorithm                   */
 87/*  see: http://aspell.net/man-html/Phonetic-Code.html */
 88/*  convert string to uppercase before this call       */
 89int phonet (const char * inword, char * target,
 90              int len,
 91	      phonetable & parms)
 92  {
 93    /**       Do phonetic transformation.       **/
 94    /**  "len" = length of "inword" incl. '\0'. **/
 95
 96    /**  result:  >= 0:  length of "target"    **/
 97    /**            otherwise:  error            **/
 98
 99    int  i,j,k=0,n,p,z;
100    int  k0,n0,p0=-333,z0;
101    char c, c0;
102    const char * s;
103    typedef unsigned char uchar;    
104    char word[MAXPHONETUTF8LEN + 1];
105    if (len == -1) len = strlen(inword);
106    if (len > MAXPHONETUTF8LEN) return 0;
107    strcpy(word, inword);
108  
109    /**  check word  **/
110    i = j = z = 0;
111    while ((c = word[i]) != '\0') {
112      n = parms.hash[(uchar) c];
113      z0 = 0;
114
115      if (n >= 0) {
116        /**  check all rules for the same letter  **/
117        while (parms.rules[n][0] == c) {
118
119          /**  check whole string  **/
120          k = 1;   /** number of found letters  **/
121          p = 5;   /** default priority  **/
122          s = parms.rules[n];
123          s++;     /**  important for (see below)  "*(s-1)"  **/
124          
125          while (*s != '\0'  &&  word[i+k] == *s
126                 &&  !isdigit ((unsigned char) *s)  &&  strchr ("(-<^$", *s) == NULL) {
127            k++;
128            s++;
129          }
130          if (*s == '(') {
131            /**  check letters in "(..)"  **/
132            if (myisalpha(word[i+k])  // ...could be implied?
133                && strchr(s+1, word[i+k]) != NULL) {
134              k++;
135              while (*s != ')')
136                s++;
137              s++;
138            }
139          }
140          p0 = (int) *s;
141          k0 = k;
142          while (*s == '-'  &&  k > 1) {
143            k--;
144            s++;
145          }
146          if (*s == '<')
147            s++;
148          if (isdigit ((unsigned char) *s)) {
149            /**  determine priority  **/
150            p = *s - '0';
151            s++;
152          }
153          if (*s == '^'  &&  *(s+1) == '^')
154            s++;
155
156          if (*s == '\0'
157              || (*s == '^'  
158                  && (i == 0  ||  ! myisalpha(word[i-1]))
159                  && (*(s+1) != '$'
160                      || (! myisalpha(word[i+k0]) )))
161              || (*s == '$'  &&  i > 0  
162                  &&  myisalpha(word[i-1])
163                  && (! myisalpha(word[i+k0]) ))) 
164          {
165            /**  search for followup rules, if:     **/
166            /**  parms.followup and k > 1  and  NO '-' in searchstring **/
167            c0 = word[i+k-1];
168            n0 = parms.hash[(uchar) c0];
169
170//            if (parms.followup  &&  k > 1  &&  n0 >= 0
171            if (k > 1  &&  n0 >= 0
172                &&  p0 != (int) '-'  &&  word[i+k] != '\0') {
173              /**  test follow-up rule for "word[i+k]"  **/
174              while (parms.rules[n0][0] == c0) {
175
176                /**  check whole string  **/
177                k0 = k;
178                p0 = 5;
179                s = parms.rules[n0];
180                s++;
181                while (*s != '\0'  &&  word[i+k0] == *s
182                       && ! isdigit((unsigned char) *s)  &&  strchr("(-<^$",*s) == NULL) {
183                  k0++;
184                  s++;
185                }
186                if (*s == '(') {
187                  /**  check letters  **/
188                  if (myisalpha(word[i+k0])
189                      &&  strchr (s+1, word[i+k0]) != NULL) {
190                    k0++;
191                    while (*s != ')'  &&  *s != '\0')
192                      s++;
193                    if (*s == ')')
194                      s++;
195                  }
196                }
197                while (*s == '-') {
198                  /**  "k0" gets NOT reduced   **/
199                  /**  because "if (k0 == k)"  **/
200                  s++;
201                }
202                if (*s == '<')
203                  s++;
204                if (isdigit ((unsigned char) *s)) {
205                  p0 = *s - '0';
206                  s++;
207                }
208
209                if (*s == '\0'
210                    /**  *s == '^' cuts  **/
211                    || (*s == '$'  &&  ! myisalpha(word[i+k0]))) 
212                {
213                  if (k0 == k) {
214                    /**  this is just a piece of the string  **/
215                    n0 += 2;
216                    continue;
217                  }
218
219                  if (p0 < p) {
220                    /**  priority too low  **/
221                    n0 += 2;
222                    continue;
223                  }
224                  /**  rule fits; stop search  **/
225                  break;
226                }
227                n0 += 2;
228              } /**  End of "while (parms.rules[n0][0] == c0)"  **/
229
230              if (p0 >= p  && parms.rules[n0][0] == c0) {
231                n += 2;
232                continue;
233              }
234            } /** end of follow-up stuff **/
235
236            /**  replace string  **/
237            s = parms.rules[n+1];
238            p0 = (parms.rules[n][0] != '\0'
239                 &&  strchr (parms.rules[n]+1,'<') != NULL) ? 1:0;
240            if (p0 == 1 &&  z == 0) {
241              /**  rule with '<' is used  **/
242              if (j > 0  &&  *s != '\0'
243                 && (target[j-1] == c  ||  target[j-1] == *s)) {
244                j--;
245              }
246              z0 = 1;
247              z = 1;
248              k0 = 0;
249              while (*s != '\0'  &&  word[i+k0] != '\0') {
250                word[i+k0] = *s;
251                k0++;
252                s++;
253              }
254              if (k > k0)
255                strmove (&word[0]+i+k0, &word[0]+i+k);
256
257              /**  new "actual letter"  **/
258              c = word[i];
259            }
260            else { /** no '<' rule used **/
261              i += k - 1;
262              z = 0;
263              while (*s != '\0'
264                     &&  *(s+1) != '\0'  &&  j < len) {
265                if (j == 0  ||  target[j-1] != *s) {
266                  target[j] = *s;
267                  j++;
268                }
269                s++;
270              }
271              /**  new "actual letter"  **/
272              c = *s;
273              if (parms.rules[n][0] != '\0'
274                 &&  strstr (parms.rules[n]+1, "^^") != NULL) {
275                if (c != '\0') {
276                  target[j] = c;
277                  j++;
278                }
279                strmove (&word[0], &word[0]+i+1);
280                i = 0;
281                z0 = 1;
282              }
283            }
284            break;
285          }  /** end of follow-up stuff **/
286          n += 2;
287        } /**  end of while (parms.rules[n][0] == c)  **/
288      } /**  end of if (n >= 0)  **/
289      if (z0 == 0) {
290//        if (k && (assert(p0!=-333),!p0) &&  j < len &&  c != '\0'
291//           && (!parms.collapse_result  ||  j == 0  ||  target[j-1] != c)){
292        if (k && !p0 && j < len &&  c != '\0'
293           && (1 || j == 0  ||  target[j-1] != c)){
294           /**  condense only double letters  **/
295          target[j] = c;
296	  ///printf("\n setting \n");
297          j++;
298        }
299
300        i++;
301        z = 0;
302	k=0;
303      }
304    }  /**  end of   while ((c = word[i]) != '\0')  **/
305
306    target[j] = '\0';
307    return (j);
308
309  }  /**  end of function "phonet"  **/