Subversion Repositories wimsdev

Rev

Rev 8123 | 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. /* Sort dictionary */
  19.  
  20. /*************** Customization: change values hereafter ****************/
  21.  
  22. /* limit of dictionary entries */
  23. #define entrylim 512*1024
  24. /* limit of dictionary length */
  25. #define diclim 32*1024*1024
  26. /* separation character */
  27. char sepchar=':', grpchar=0;
  28.  
  29. /***************** Nothing should need change hereafter *****************/
  30.  
  31. #include "suffix.h"
  32. #include "../wims.h"
  33. #include "../Lib/libwims.h"
  34.  
  35. char inpbuf[MAX_LINELEN+1], outbuf[2*MAX_LINELEN+2];
  36. char *dicbuf;
  37. char dicname[1024], suffixname[1024];
  38.  
  39. struct entry {
  40.     char *original;
  41.     char *replace;
  42. } entry[entrylim];
  43. int entrycount;
  44.  
  45. int nocase=0, hassuffix=0, leaveline=0;
  46. int entrycount, ocount;
  47.  
  48. /* strip trailing spaces; return string end. */
  49. char *strip_trailing_spaces2(char *p)
  50. {
  51.     char *pp;
  52.     if(*p==0) return p;
  53.     for(pp=p+strlen(p)-1; pp>=p && isspace(*pp); *(pp--)=0);
  54.     return pp;
  55. }
  56.  
  57. int compare(const void *s1, const void *s2)
  58. {
  59.     const struct entry *p1, *p2;
  60.     p1=s1; p2=s2;
  61.     if(nocase) return strcasecmp(p1->original,p2->original);
  62.     else return strcmp(p1->original,p2->original);
  63. }
  64.  
  65. void sortdic(void)
  66. {
  67.     qsort(entry,entrycount,sizeof(entry[0]),compare);
  68. }
  69.  
  70. /* modify a string. Bufferlen must be at least MAX_LINELEN */
  71. void string_modify3(char *start, char *bad_beg, char *bad_end, char *good,...)
  72. {
  73.     char buf[MAX_LINELEN+1];
  74.     va_list vp;
  75.  
  76.     va_start(vp,good);
  77.     vsnprintf(buf,sizeof(buf),good,vp); va_end(vp);
  78.     if(strlen(start)-(bad_end-bad_beg)+strlen(buf)>=MAX_LINELEN)
  79.       return; /* this is an error situation. */
  80.     strcat(buf,bad_end);
  81.     ovlstrcpy(bad_beg,buf);
  82. }
  83.  
  84. /* change all spaces into ' ', and collapse multiple occurences */
  85. void singlespace2(char *p)
  86. {
  87.     char *pp, *p2;
  88.     for(pp=p;*pp;pp++) {
  89.       if(!isspace(*pp)) continue;
  90.       if(leaveline) {
  91.           if(*pp==13) ovlstrcpy(pp,pp+1);
  92.           if(*pp=='\n') {
  93.             pp++;
  94.             gopt: for(p2=pp; isspace(*p2) && *p2!='\n'; p2++);
  95.             if(p2>pp) ovlstrcpy(pp,p2); pp--;
  96.           }
  97.           else {
  98.             pp++; if(!isspace(*pp) || *pp=='\n') continue;
  99.             goto gopt;
  100.           }
  101.       }
  102.       else {
  103.           if(*pp!=' ') *pp=' ';
  104.           pp++; if(!isspace(*pp)) continue;
  105.           for(p2=pp;isspace(*p2);p2++);
  106.           ovlstrcpy(pp,p2); pp--;
  107.       }
  108.     }
  109. }
  110.  
  111. /* Prepare dictionary */
  112. void prepare_dic(void)
  113. {
  114.     int i;
  115.     FILE *dicf;
  116.     char *p1, *p2, *pp;
  117.     long int flen;
  118.  
  119.     entrycount=0;
  120.     dicf=fopen(dicname,"r"); if(dicf==NULL) return;
  121.     fseek(dicf,0,SEEK_END);flen=ftell(dicf); fseek(dicf,0,SEEK_SET);
  122.     if(flen>diclim) return;
  123.     dicbuf=xmalloc(2*flen+1024);flen=fread(dicbuf,1,flen,dicf);
  124.     fclose(dicf);
  125.     if(flen>0 && flen<diclim) dicbuf[flen]=0;
  126.     else return;
  127.     for(i=0,p1=dicbuf;p1!=NULL && *p1!=0 && i<entrylim;p1=p2) {
  128.       p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0;
  129.       pp=strchr(p1,sepchar); if(pp==NULL) continue;
  130.       *pp++=0;
  131.       strip_trailing_spaces2(p1); strip_trailing_spaces2(pp);
  132.       singlespace2(p1);
  133.       p1=find_word_start(p1); pp=find_word_start(pp);
  134.       if(*p1==0) continue;
  135.       entry[i].original=p1; entry[i].replace=pp; i++;
  136.     }
  137.     entrycount=i;
  138. }
  139.  
  140. void output(void)
  141. {
  142.     int i;
  143.     FILE *f;
  144.  
  145.     ocount=0;
  146.     strcat(dicname,".sorted");
  147.     f=fopen(dicname,"w"); if(f==NULL) return;
  148.     for(i=0;i<entrycount;i++) {
  149.       if(i>0 && strcmp(entry[i].original,entry[i-1].original)==0
  150.          && strcmp(entry[i].replace,entry[i-1].replace)==0)
  151.         continue;
  152.       if(grpchar!=0) {
  153.           if(i>0 && strcmp(entry[i].original,entry[i-1].original)==0)
  154.             fprintf(f,"%c%s",grpchar, entry[i].replace);
  155.           else {
  156.             if(i>0) fprintf(f,"\n");
  157.             fprintf(f,"%s%c%s",entry[i].original,sepchar,entry[i].replace);
  158.             ocount++;
  159.           }
  160.  
  161.       }
  162.       else {
  163.           fprintf(f,"%s%c%s\n",entry[i].original,sepchar,entry[i].replace);
  164.           ocount++;
  165.       }
  166.     }
  167.     if(grpchar!=0) fprintf(f,"\n");
  168.     fclose(f);
  169. }
  170.  
  171. int main(int argc, char *argv[])
  172. {
  173.     char *ss, *gr;
  174.     if(argc<2) return -1;
  175.  
  176.     ss=getenv("dicsort_separator");
  177.     if(ss!=NULL && *ss!=0) sepchar=*ss;
  178.     gr=getenv("dicsort_grouping");
  179.     if(gr!=NULL && *gr!=0) grpchar=*gr;
  180.     snprintf(dicname,sizeof(dicname)-128,"%s",argv[1]); prepare_dic();
  181.     if(argc>2) {
  182.       snprintf(suffixname,sizeof(suffixname),"%s",argv[2]);
  183.       suffix_dic(suffixname); hassuffix=1;
  184.     }
  185.     else suffixname[0]=hassuffix=0;
  186.     sortdic(); output();
  187.     printf("%s: sorted %d entries.\n",dicname, ocount);
  188.     return 0;
  189. }
  190.  
  191.