Subversion Repositories wimsdev

Rev

Rev 7676 | Rev 8185 | Go to most recent revision | Blame | Compare with Previous | Last modification | View Log | RSS feed

  1. /*    Copyright (C) 1998-2003 XIAO, Gang of Universite de Nice - Sophia Antipolis
  2.  *
  3.  *  This program is free software; you can redistribute it and/or modify
  4.  *  it under the terms of the GNU General Public License as published by
  5.  *  the Free Software Foundation; either version 2 of the License, or
  6.  *  (at your option) any later version.
  7.  *
  8.  *  This program is distributed in the hope that it will be useful,
  9.  *  but WITHOUT ANY WARRANTY; without even the implied warranty of
  10.  *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  11.  *  GNU General Public License for more details.
  12.  *
  13.  *  You should have received a copy of the GNU General Public License
  14.  *  along with this program; if not, write to the Free Software
  15.  *  Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  16.  */
  17.  
  18. /* Versatile translation according to a dictionary */
  19.  
  20. /*************** Customization: change values hereafter ****************/
  21.  
  22. /* limit of dictionary entries */
  23. #define entrylim 32768
  24. /* limit of dictionary length */
  25. #define diclim 1024*1024
  26.  
  27. /***************** Nothing should need change hereafter *****************/
  28.  
  29.  
  30. char inpbuf[MAX_LINELEN+1], outbuf[2*MAX_LINELEN+2];
  31. char *dicbuf;
  32. struct entry {
  33.     unsigned char *original, *replace;
  34.     int olen,earlier;
  35. } entry[entrylim];
  36. int entrycount;
  37.  
  38. enum {
  39.     unk_delete, unk_leave, unk_replace
  40. };
  41.  
  42. int has_digits=0;
  43. int unknown_type=unk_delete;
  44. int nocase=0,leaveline=0;
  45. char *unknown, unkbuf[1024];
  46.  
  47. int compare(int i1, const char *s2)
  48. {
  49.     int k;
  50.     if(nocase) k=strncasecmp((char*)entry[i1].original,s2,entry[i1].olen);
  51.     else k=strncmp((char*)entry[i1].original,s2,entry[i1].olen);
  52.     if(k==0 && (isalnum(*(s2+entry[i1].olen)) || (*(s2+entry[i1].olen)&128)!=0)) return -1;
  53.     else return k;
  54. }
  55.  
  56. /* searches a list. Returns index if found, -1 if nomatch.
  57.  * Uses binary search, list must be sorted. */
  58. int search_list2(struct entry *list, int items, size_t item_size, const char *str)
  59. {
  60.     int i1,i2,j,k,t,t1;
  61.     unsigned char c;
  62.  
  63.     if(items<=0) return -1;
  64.     j=0; c=str[0];
  65.     k=list[0].original[0]-c; if(k==0) k=compare(0,str);
  66.     if(k==0) goto more; if(k>0) return -1;
  67.     j=items-1; k=list[j].original[0]-c; if(k==0) k=compare(j,str);
  68.     if(k==0) return j;
  69.     if(k>0) for(i1=0,i2=j;i2>i1+1;) {
  70.       j=i1+(i2-i1)/2;
  71.       k=list[j].original[0]-c; if(k==0) k=compare(j,str);
  72.       if(k==0) goto more;
  73.       if(k>0) {i2=j; continue;}
  74.       if(k<0) {i1=j; continue;}
  75.     }
  76.     if(k>0) {j--;k=compare(j,str);}
  77.     more:
  78.     if((t=list[j].earlier)<0) {
  79.       if(k==0) return j; else return -1;
  80.     }
  81.     if(compare(t,str)!=0) return -1;
  82.     for(j=t1=t,k=0;j<items && list[j].earlier==t1 && (k=compare(j,str))<=0; j++) {
  83.       if(k==0) t=j;
  84.     }
  85.     return t;
  86. }
  87.  
  88. /* change all spaces into ' ', and collapse multiple occurences */
  89. void singlespace2(char *p)
  90. {
  91.     char *pp, *p2;
  92.     for(pp=p;*pp;pp++) {
  93.       if(!isspace(*pp)) continue;
  94.       if(leaveline) {
  95.           if(*pp==13) ovlstrcpy(pp,pp+1);
  96.           if(*pp=='\n') {
  97.             pp++;
  98.             gopt: for(p2=pp; isspace(*p2) && *p2!='\n'; p2++);
  99.             if(p2>pp) ovlstrcpy(pp,p2); pp--;
  100.           }
  101.           else {
  102.             pp++; if(!isspace(*pp) || *pp=='\n') continue;
  103.             goto gopt;
  104.           }
  105.       }
  106.       else {
  107.           if(*pp!=' ') *pp=' ';
  108.           if(!isspace(*(pp+1))) continue;
  109.           for(pp++,p2=pp;isspace(*p2);p2++);
  110.           ovlstrcpy(pp,p2); pp--;
  111.       }
  112.     }
  113. }
  114.  
  115. #include "suffix.c"
  116.  
  117. /* Prepare dictionary */
  118. void prepare_dic(char *fname)
  119. {
  120.     int i,l;
  121.     FILE *dicf;
  122.     char *p1, *p2, *pp;
  123.     long int flen;
  124.  
  125.     entrycount=0;
  126.     dicf=fopen(fname,"r"); if(dicf==NULL) return;
  127.     fseek(dicf,0,SEEK_END);flen=ftell(dicf); fseek(dicf,0,SEEK_SET);
  128.     if(flen>diclim) return;
  129.     dicbuf=xmalloc(flen+16);flen=fread(dicbuf,1,flen,dicf);
  130.     fclose(dicf);
  131.     if(flen>0 && flen<diclim) dicbuf[flen]=0;
  132.     else return;
  133.     for(i=0,p1=dicbuf;p1!=NULL && *p1!=0 && i<entrylim;p1=p2) {
  134.       p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0;
  135.       pp=strchr(p1,':'); if(pp==NULL) continue;
  136.       *pp++=0;
  137.       strip_trailing_spaces2(p1); strip_trailing_spaces2(pp);
  138.       singlespace2(p1);
  139.       p1=find_word_start(p1); pp=find_word_start(pp);
  140.       if(*p1==0) continue;
  141.       if(has_digits==0) {
  142.           char *p;
  143.           for(p=p1;*p!=0 && p<pp && !isdigit(*p);p++);
  144.           if(isdigit(*p)) has_digits=1;
  145.       }
  146.       entry[i].original=(unsigned char*)p1;
  147.       entry[i].replace=(unsigned char*)pp;
  148.       entry[i].olen=l=strlen(p1); entry[i].earlier=-1;
  149.       if(i>0) {
  150.           int l1,l2;
  151.           l1=entry[i-1].earlier; if(l1>=0) l2=entry[l1].olen;
  152.           else {l2=entry[i-1].olen;l1=i-1;}
  153.           if(l>l2 && isspace(p1[l2])
  154.              && strncmp((char*)entry[l1].original,p1,l2)==0)
  155.             entry[i].earlier=entry[i-1].earlier=l1;
  156.       }
  157.       i++;
  158.     }
  159.     entrycount=i;
  160. }
  161.  
  162. /* now make the translation. */
  163. void translate(char *p)
  164. {
  165.     char *p1, *p2, *pp;
  166.     int t;
  167.  
  168.     if(entrycount<=0 && suffixcnt<=0) return;
  169.     snprintf(outbuf,sizeof(outbuf),"%s",p);
  170.     for(p1=find_word_start(outbuf);
  171.       p1!=NULL && p1-outbuf<MAX_LINELEN && *p1!=0;
  172.       p1=p2) {
  173.       p2=find_word_end(p1);
  174.       for(pp=p1;pp<p2 &&
  175.           ((!has_digits && isalpha(*pp)) ||
  176.            (has_digits && isalnum(*pp)) || (*pp&128)!=0 ||
  177.            strchr("_",*pp)!=NULL);pp++);
  178.       p2=find_word_start(p2);
  179.       if(pp==p1 ||
  180.          (has_digits==0 && isdigit(*pp)) ||
  181.          (*pp!=0 && !isspace(*pp) && strchr(",.?!/;",*pp)==NULL)) continue;
  182.       t=search_list2(entry,entrycount,sizeof(entry[0]),p1);
  183.       if(t<0) {
  184.           switch(unknown_type) {
  185.             case unk_leave: break;
  186.             case unk_delete: {
  187.                 ovlstrcpy(p1,find_word_start(pp)); p2=p1;
  188.                 break;
  189.             }
  190.             case unk_replace: {
  191.                 string_modify3(outbuf,p1,pp,unkbuf);
  192.                 p2=find_word_start(p1+strlen(unkbuf));
  193.             }
  194.           }
  195.           continue;
  196.       }
  197.       string_modify3(outbuf,p1,p1+strlen((char*)entry[t].original),
  198.                   (char*)entry[t].replace);
  199.       p2=find_word_start(p1+strlen((char*)entry[t].replace));
  200.     }
  201.     snprintf(p,MAX_LINELEN,"%s",outbuf);
  202. }
  203.  
  204.