Subversion Repositories wimsdev

Rev

Rev 8100 | Rev 8149 | 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. /*
  49. void *xmalloc(size_t n)
  50. {
  51.     void *p;
  52.     p=malloc(n);
  53.     if(p==NULL) exit(1);
  54.     return p;
  55. }
  56. */
  57. /* Points to the end of the word */
  58. /*char *find_word_end(char *p)
  59. {
  60.     int i;
  61.     for(i=0;!isspace(*p) && *p!=0 && i<MAX_LINELEN; p++,i++);
  62.     return p;
  63. }
  64. */
  65. /* Strips leading spaces */
  66. /*char *find_word_start(char *p)
  67. {
  68.     int i;
  69.     for(i=0; isspace(*p) && i<MAX_LINELEN; p++,i++);
  70.     return p;
  71. }
  72. */
  73. /* strip trailing spaces; return string end. */
  74. char *strip_trailing_spaces2(char *p)
  75. {
  76.     char *pp;
  77.     if(*p==0) return p;
  78.     for(pp=p+strlen(p)-1; pp>=p && isspace(*pp); *(pp--)=0);
  79.     return pp;
  80. }
  81.  
  82. int compare(const void *s1, const void *s2)
  83. {
  84.     const struct entry *p1, *p2;
  85.     p1=s1; p2=s2;
  86.     if(nocase) return strcasecmp(p1->original,p2->original);
  87.     else return strcmp(p1->original,p2->original);
  88. }
  89.  
  90. void sortdic(void)
  91. {
  92.     qsort(entry,entrycount,sizeof(entry[0]),compare);
  93. }
  94.  
  95. /* modify a string. Bufferlen must be at least MAX_LINELEN */
  96. void string_modify3(char *start, char *bad_beg, char *bad_end, char *good,...)
  97. {
  98.     char buf[MAX_LINELEN+1];
  99.     va_list vp;
  100.  
  101.     va_start(vp,good);
  102.     vsnprintf(buf,sizeof(buf),good,vp); va_end(vp);
  103.     if(strlen(start)-(bad_end-bad_beg)+strlen(buf)>=MAX_LINELEN)
  104.       return; /* this is an error situation. */
  105.     strcat(buf,bad_end);
  106.     ovlstrcpy(bad_beg,buf);
  107. }
  108.  
  109. /* change all spaces into ' ', and collapse multiple occurences */
  110. void singlespace2(char *p)
  111. {
  112.     char *pp, *p2;
  113.     for(pp=p;*pp;pp++) {
  114.       if(!isspace(*pp)) continue;
  115.       if(leaveline) {
  116.           if(*pp==13) ovlstrcpy(pp,pp+1);
  117.           if(*pp=='\n') {
  118.             pp++;
  119.             gopt: for(p2=pp; isspace(*p2) && *p2!='\n'; p2++);
  120.             if(p2>pp) ovlstrcpy(pp,p2); pp--;
  121.           }
  122.           else {
  123.             pp++; if(!isspace(*pp) || *pp=='\n') continue;
  124.             goto gopt;
  125.           }
  126.       }
  127.       else {
  128.           if(*pp!=' ') *pp=' ';
  129.           pp++; if(!isspace(*pp)) continue;
  130.           for(p2=pp;isspace(*p2);p2++);
  131.           ovlstrcpy(pp,p2); pp--;
  132.       }
  133.     }
  134. }
  135.  
  136. /* Prepare dictionary */
  137. void prepare_dic(void)
  138. {
  139.     int i;
  140.     FILE *dicf;
  141.     char *p1, *p2, *pp;
  142.     long int flen;
  143.  
  144.     entrycount=0;
  145.     dicf=fopen(dicname,"r"); if(dicf==NULL) return;
  146.     fseek(dicf,0,SEEK_END);flen=ftell(dicf); fseek(dicf,0,SEEK_SET);
  147.     if(flen>diclim) return;
  148.     dicbuf=xmalloc(2*flen+1024);flen=fread(dicbuf,1,flen,dicf);
  149.     fclose(dicf);
  150.     if(flen>0 && flen<diclim) dicbuf[flen]=0;
  151.     else return;
  152.     for(i=0,p1=dicbuf;p1!=NULL && *p1!=0 && i<entrylim;p1=p2) {
  153.       p2=strchr(p1+1,'\n'); if(p2>p1) *p2++=0;
  154.       pp=strchr(p1,sepchar); if(pp==NULL) continue;
  155.       *pp++=0;
  156.       strip_trailing_spaces2(p1); strip_trailing_spaces2(pp);
  157.       singlespace2(p1);
  158.       p1=find_word_start(p1); pp=find_word_start(pp);
  159.       if(*p1==0) continue;
  160.       entry[i].original=p1; entry[i].replace=pp; i++;
  161.     }
  162.     entrycount=i;
  163. }
  164.  
  165. void output(void)
  166. {
  167.     int i;
  168.     FILE *f;
  169.  
  170.     ocount=0;
  171.     strcat(dicname,".sorted");
  172.     f=fopen(dicname,"w"); if(f==NULL) return;
  173.     for(i=0;i<entrycount;i++) {
  174.       if(i>0 && strcmp(entry[i].original,entry[i-1].original)==0
  175.          && strcmp(entry[i].replace,entry[i-1].replace)==0)
  176.         continue;
  177.       if(grpchar!=0) {
  178.           if(i>0 && strcmp(entry[i].original,entry[i-1].original)==0)
  179.             fprintf(f,"%c%s",grpchar, entry[i].replace);
  180.           else {
  181.             if(i>0) fprintf(f,"\n");
  182.             fprintf(f,"%s%c%s",entry[i].original,sepchar,entry[i].replace);
  183.             ocount++;
  184.           }
  185.  
  186.       }
  187.       else {
  188.           fprintf(f,"%s%c%s\n",entry[i].original,sepchar,entry[i].replace);
  189.           ocount++;
  190.       }
  191.     }
  192.     if(grpchar!=0) fprintf(f,"\n");
  193.     fclose(f);
  194. }
  195.  
  196. int main(int argc, char *argv[])
  197. {
  198.     char *ss, *gr;
  199.     if(argc<2) return -1;
  200.  
  201.     ss=getenv("dicsort_separator");
  202.     if(ss!=NULL && *ss!=0) sepchar=*ss;
  203.     gr=getenv("dicsort_grouping");
  204.     if(gr!=NULL && *gr!=0) grpchar=*gr;
  205.     snprintf(dicname,sizeof(dicname)-128,"%s",argv[1]); prepare_dic();
  206.     if(argc>2) {
  207.       snprintf(suffixname,sizeof(suffixname),"%s",argv[2]);
  208.       suffix_dic(suffixname); hassuffix=1;
  209.     }
  210.     else suffixname[0]=hassuffix=0;
  211.     sortdic(); output();
  212.     printf("%s: sorted %d entries.\n",dicname, ocount);
  213.     return 0;
  214. }
  215.  
  216.