Subversion Repositories wimsdev

Rev

Rev 3813 | Rev 8136 | Go to most recent revision | Blame | Compare with Previous | Last modification | View Log | RSS feed

  1. /*    Copyright (C) 2002-2003 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. int nopart=0, nonesting=0, nocomma=0, nocoord=0, _estart;
  19.  
  20. void extrout(char *pb, char *pe, int dist, int commas[], int *commacnt)
  21. {
  22.     int i;
  23.     if(pe-pb<thislinelen && nopart) return;
  24.     while(pe>pb && isspace(*(pe-1))) pe--;
  25.     if(pe<=pb) return;
  26.     if(regexcnt>0 && pe-pb<MAX_LINELEN) {
  27.         char buf[MAX_LINELEN+1];
  28.         memmove(buf,pb,pe-pb); buf[pe-pb]=0;
  29.         if(!checkregex(buf)) return;
  30.     }
  31.     if(!nocoord)
  32.       printf("%d %d %d, ",thisobjline,dist+_estart,(int)(pe-pb)+dist+_estart);
  33.     while(*commacnt>0 && commas[*commacnt-1]>pe-pb) (*commacnt)--;
  34.     if(*commacnt>0 && !nocomma) {
  35.         fwrite(pb,1,commas[0],stdout); putchar(',');
  36.         for(i=1;i<*commacnt;i++) {
  37.             fwrite(pb+commas[i-1],1,commas[i]-commas[i-1],stdout); putchar(',');
  38.         }
  39.         pb+=commas[*commacnt-1];
  40.     }
  41.     fwrite(pb,1,pe-pb,stdout); printf("\n");
  42.     *commacnt=0;
  43. }
  44.  
  45. void _extract(char *p, int dist)
  46. {
  47.     int i,l,got,lt;
  48.     char *p1, *p2, *p3, *p4;
  49.     char buf[MAX_LINELEN+1];
  50.     int commas[MAX_COMMAS];
  51.     int commacnt;
  52.  
  53.     if(dist>0 && (nonesting || nopart)) return;
  54.     commacnt=0;
  55.     if(expl1<=0) got=1; else got=0;
  56.     for(p1=find_word_start(p), p2=p1; *p2; p2=find_word_start(p3)) {
  57.         if(*p2=='.' || isdigit(*p2)) {
  58.             (void)strtod(p2,&p3); continue;
  59.         }
  60.         if(*p2=='(') {
  61.             p3=find_matching(p2+1,')'); lt=exp_paren;
  62.             paren: if(p3==NULL) error("Unmatched parentheses.");
  63.             p2++; memmove(buf,p2,p3-p2); buf[p3-p2]=0;
  64.             _extract(buf,p2-p+dist);
  65.             if(expl2==lt) extrout(p2,p3,dist+p2-p,commas,&commacnt);
  66.             p3++; continue;
  67.         }
  68.         if(*p2=='[') {
  69.             p3=find_matching(p2+1,']'); lt=exp_matrix; goto paren;
  70.         }
  71.         if(*p2=='{') {
  72.             p3=find_matching(p2+1,'}'); lt=exp_set; goto paren;
  73.         }
  74.         if(isalpha(*p2)) {
  75.             for(p3=p2; *p3=='_' || isalnum(*p3); p3++);
  76.             if(p3-p2>=16) goto notdefined;
  77.             memmove(buf,p2,p3-p2); buf[p3-p2]=0;
  78.             for(i=0;i<opalphano && strcmp(buf,opalpha[i].name)!=0; i++);
  79.             if(i<opalphano) {
  80.                 l=opalpha[i].lvl; if(l>expl2) {
  81.                     if(got) extrout(p1,p2,dist+p1-p,commas,&commacnt);
  82.                     if(expl1>0) got=0; p1=find_word_start(p3);
  83.                 }
  84.                 if(l>=expl1 && l<=expl2) {
  85.                     got=1;
  86.                     if(l>0 && commacnt<MAX_COMMAS-2) {
  87.                         commas[commacnt++]=p2-p1;
  88.                         commas[commacnt++]=p3-p1;
  89.                     }
  90.                 }
  91.                 continue;
  92.             }
  93.             notdefined: p4=find_word_start(p3);
  94.             if(*p4=='(') {
  95.                 p3=find_matching(p4+1,')');
  96.                 if(p3==NULL) error("Unmatched parentheses.");
  97.                 p4++; memmove(buf,p4,p3-p2); buf[p3-p4]=0;
  98.                 _extract(buf,p2-p+dist);
  99.                 p3++;
  100.                 if(expl2==exp_fn) extrout(p2,p3,dist+p2-p,commas,&commacnt);
  101.             }
  102.             else if(expl2==exp_variable) extrout(p2,p3,dist+p2-p,commas,&commacnt);
  103.             continue;
  104.         }
  105.         for(i=0;i<oppunctno && strncmp(p2,oppunct[i].name,strlen(oppunct[i].name))!=0; i++);
  106.         if(i>=oppunctno) error("Unknown operator.");
  107.         p3=p2+strlen(oppunct[i].name); l=oppunct[i].lvl;
  108.         if(l>expl2) {
  109.             if(got) extrout(p1,p2,dist+p1-p,commas,&commacnt);
  110.             if(expl1>0) got=0; p1=find_word_start(p3);
  111.         }
  112.         if(l>=expl1 && l<=expl2) {
  113.             got=1;
  114.             if(l>0 && commacnt<MAX_COMMAS-2) {
  115.                 commas[commacnt++]=p2-p1;
  116.                 commas[commacnt++]=p3-p1;
  117.             }
  118.         }
  119.     }
  120.     if(got) extrout(p1,p1+strlen(p1),dist+p1-p,commas,&commacnt);
  121. }
  122.  
  123. void req_extract(void)
  124. {
  125.     int i;
  126.     char *p;
  127.  
  128.     if(objlinecnt<2) return;
  129.     if(*reqtype==0) error("Missing extraction type.");
  130.     for(i=0;i<exptypeno && strcmp(reqtype,exptype[i].name)!=0; i++);
  131.     if(i>=exptypeno) error("Bad extraction type.");
  132.     if(logdir!=0) nopart=1;
  133.     while((p=wordchr(objline[0],"nopart"))!=NULL) {
  134.         nopart=1; ovlstrcpy(p,p+strlen("nopart"));
  135.     }
  136.     while((p=wordchr(objline[0],"nonesting"))!=NULL) {
  137.         nonesting=1; ovlstrcpy(p,p+strlen("nonesting"));
  138.     }
  139.     while((p=wordchr(objline[0],"nocomma"))!=NULL) {
  140.         nocomma=1; ovlstrcpy(p,p+strlen("nocomma"));
  141.     }
  142.     getregex(objline[0]);
  143.     expl1=exptype[i].lvl1; expl2=exptype[i].lvl2;
  144.     for(i=1;i<objlinecnt;i++) {
  145.         thisobjline=i; p=find_word_start(objline[i]);
  146.         linelogdir=0;
  147.         if(*p=='>') {
  148.             if(logdir<0) continue;
  149.             p=find_word_start(p+1); linelogdir=1;
  150.         }
  151.         else if(*p=='<') {
  152.             if(logdir>0) continue;
  153.             p=find_word_start(p+1); linelogdir=-1;
  154.         }
  155.         thislinelen=strlen(p); if(thislinelen<=0) continue;
  156.         _estart=p-objline[i]; _extract(p, 0);
  157.     }
  158. }
  159.  
  160. void req_cut(void)
  161. {
  162.     nocoord=1; req_extract();
  163. }
  164.  
  165.