Subversion Repositories wimsdev

Rev

Rev 8094 | 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. /* Check type of a file */
  19.  
  20. /*************** Customization: change values hereafter ****************/
  21.  
  22. /* limit of data buffers */
  23. #define buflim 1024*1024*16
  24.  
  25. /***************** Nothing should need change hereafter *****************/
  26. #include "../wims.h"
  27. #include "../Lib/libwims.h"
  28. #include "../hmname.c"
  29.  
  30. char fn1[1024]="", fn2[1024]="";
  31. char mathbuf[MAX_LINELEN+1];
  32. char *filebuf;
  33. int filelen=0;
  34. int latex2html=0;
  35. FILE *outf;
  36.  
  37. struct {
  38.     char *name, *trans;
  39. } backtrans[]={
  40.     {"\\ge\\", " >= "},
  41.     {"\\geq\\", " >= "},
  42.     {"\\le\\", " <= "},
  43.     {"\\leq\\", " <= "},
  44.     {"\\to\\", " -> "},
  45.     {"\\rightarrow\\", " -> "},
  46.     {"\\longrightarrow\\", " --> "},
  47.     {"\\Rightarrow\\", " => "},
  48.     {"\\Longrightarrow\\", " ==> "},
  49.     {"\\Leftrightarrow\\", " <=> "},
  50.     {"\\Longleftrightarrow\\", " <==> "},
  51.     {"\\Longleftarrow\\", " <== "},
  52. };
  53.  
  54. #define backtransno (sizeof(backtrans)/sizeof(backtrans[0]))
  55.  
  56. /*void *xmalloc(size_t n)
  57. {
  58.     void *p;
  59.     p=malloc(n);
  60.     if(p==NULL) exit(1);
  61.     return p;
  62. }
  63. */
  64.  
  65. /* Points to the end of the word */
  66. /*
  67. char *find_word_end(char *p)
  68. {
  69.     int i;
  70.     for(i=0;!isspace(*p) && *p!=0 && i<MAX_LINELEN; p++,i++);
  71.     return p;
  72. }
  73. */
  74. /* Strips leading spaces */
  75. /*char *find_word_start(char *p)
  76. {
  77.     int i;
  78.     for(i=0; isspace(*p) && i<MAX_LINELEN; p++,i++);
  79.     return p;
  80. }
  81. */
  82. char *find_tag_end(char *p)
  83. {
  84.     char *pp, *old;
  85.     pp=p; if(*pp=='<') pp++;
  86.     for(; *pp && *pp!='>'; pp++) {
  87.       if(*pp=='"') {
  88.           pp=strchr(pp+1,'"');
  89.           if(pp==NULL) {pp=p+strlen(p); break;} else continue;
  90.       }
  91.     }
  92. /* this is probably an syntax error of the page */
  93.     if(*pp==0 && pp>p+2048) {
  94.       old=p; if(*old=='<') old++;
  95.       pp=strchr(old,'>');
  96.       if(pp==NULL) pp=strchr(old,'<');
  97.       if(pp==NULL) pp=find_word_end(find_word_start(old));
  98.     }
  99.     if(*pp=='>') pp++; return pp;
  100. }
  101.  
  102. char *find_tag(char *p, char *tag)
  103. {
  104.     char *pp;
  105.     int len;
  106.     len=strlen(tag);
  107.     for(pp=strchr(p,'<'); pp!=NULL && *pp; pp=strchr(pp+1,'<')) {
  108.       if(strncasecmp(pp+1,tag,len)==0 && !isalnum(*(pp+1+len))) return pp;
  109.     }
  110.     return p+strlen(p);
  111. }
  112.  
  113. /* modify a string. Bufferlen must be at least MAX_LINELEN */
  114. void string_modify3(char *start, char *bad_beg, char *bad_end, char *good,...)
  115. {
  116.     char buf[MAX_LINELEN+1];
  117.     va_list vp;
  118.  
  119.     va_start(vp,good);
  120.     vsnprintf(buf,sizeof(buf),good,vp); va_end(vp);
  121.     if(strlen(start)-(bad_end-bad_beg)+strlen(buf)>=MAX_LINELEN) {
  122.       return;
  123.     }
  124.     strcat(buf,bad_end);
  125.     ovlstrcpy(bad_beg,buf);
  126. }
  127.  
  128. void cutamp(char *p)
  129. {
  130.     char *pp;
  131.     for(pp=strchr(p,'&'); pp; pp=strchr(pp+1,'&')) {
  132.       if(strncmp(pp,"&amp;",5)==0) {
  133.           ovlstrcpy(pp+1,pp+5); continue;
  134.       }
  135.       if(strncmp(pp,"&lt;",4)==0) {
  136.           *pp='<'; ovlstrcpy(pp+1,pp+4); continue;
  137.       }
  138.       if(strncmp(pp,"&gt;",4)==0) {
  139.           *pp='>'; ovlstrcpy(pp+1,pp+4); continue;
  140.       }
  141.  
  142.     }
  143. }
  144.  
  145. /* get the file */
  146. void prepare_file(void)
  147. {
  148.     FILE *f;
  149.     long int flen;
  150.  
  151.     filelen=0;
  152.     f=fopen(fn1,"r"); if(f==NULL) return;
  153.     fseek(f,0,SEEK_END);flen=ftell(f); fseek(f,0,SEEK_SET);
  154.     if(flen>buflim) return;
  155.     filebuf=xmalloc(2*flen+1024);flen=fread(filebuf,1,flen,f);
  156.     fclose(f);
  157.     if(flen<0 || flen>=buflim) flen=0; filebuf[flen]=0;
  158.     filelen=flen;
  159.     outf=fopen(fn2,"w"); if(outf==NULL) return;
  160. }
  161.  
  162. void getmath(char *p)
  163. {
  164.     char *pt, *pv;
  165.     char *p1, *p2, buf[256];
  166.  
  167.     mathbuf[0]=0;
  168.     pt=find_word_start(p);
  169.     if(strncmp(pt,"\\begin{displaymath}",
  170.                strlen("\\begin{displaymath}"))==0) {
  171.       pt=strchr(pt,'}')+1;
  172.       pv=strstr(pt,"\\end{displaymath}");
  173.       if(pv==NULL) return;
  174.       goto insmath;
  175.     }
  176.     if(*pt=='%') pt=strchr(pt,'$'); if(pt==NULL) return;
  177.     if(*pt!='$') return; do pt++; while(*pt=='$');
  178.     pv=strchr(pt,'$'); if(pv==NULL) return;
  179.     insmath: if(pv-pt>=MAX_LINELEN-256) return;
  180.     memmove(mathbuf,pt,pv-pt); mathbuf[pv-pt]=0;
  181.     if(strstr(mathbuf,"...\n...")!=NULL) {
  182.       ovlstrcpy(mathbuf,"......"); return;
  183.     }
  184.     cutamp(mathbuf); latex2html=1;
  185.     for(p1=strstr(mathbuf,"\\mathbb");p1;p1=strstr(p1+1,"\\mathbb")) {
  186.       char c,*d;
  187.       p2=find_word_start(p1+strlen("\\mathbb")); c=0;
  188.       if(strchr("NZQRC",*p2)!=NULL) c=*p2;
  189.       else if(*p2=='{' && *(p2+2)=='}' && strchr("NZQRC",*(p2+1))!=NULL) {
  190.             c=*(p2+1); p2+=2;
  191.       }
  192.       if(c) {
  193.           p2=find_word_start(++p2);
  194.           if(isalnum(*p2)) d=" "; else d="";
  195.           string_modify(mathbuf,p1,p2,"\\%c%c%s",c,c,d);
  196.       }
  197.     }
  198.     for(p1=strstr(mathbuf,"{\\"); p1; p1=strstr(p1+1,"{\\")) {
  199.       if(p1>mathbuf && isalpha(*(p1-1))) continue;
  200.       for(p2=p1+2; p2<p1+24 && isalpha(*p2); p2++);
  201.       if(*p2!='}' || isalnum(*(p2+1))) continue;
  202.       memmove(buf,p1+1,p2-p1-1); buf[p2-p1-1]='\\'; buf[p2-p1]=0;
  203.       if(strstr(hmsame,buf)==NULL) continue;
  204.       ovlstrcpy(p2,p2+1); ovlstrcpy(p1,p1+1);
  205.     }
  206.     if(strstr(mathbuf,"\\begin{")!=NULL) return;
  207.     for(p1=strchr(mathbuf,'{'); p1; p1=strchr(p1+1,'{')) {
  208.       if((p1>mathbuf && isalpha(*(p1-1))) ||
  209.          !isalnum(*(p1+1)) || *(p1+2)!='}') continue;
  210.       *p1=*(p1+1); ovlstrcpy(p1+1,p1+3);
  211.     }
  212.     if(strchr(mathbuf,'[')!=NULL) {
  213.         char mbuf[MAX_LINELEN+1];
  214.       snprintf(mbuf,sizeof(mbuf),"{%s}",mathbuf);
  215.       ovlstrcpy(mathbuf,mbuf);
  216.     }
  217. /* try to simplify */
  218.     if(strchr(mathbuf,'{')==NULL && strchr(mathbuf,'\\')!=NULL) {
  219.       int i, tt;
  220.       tt=0;
  221.       for(p1=strchr(mathbuf,'\\'); p1; p1=strchr(p1+1,'\\')) {
  222.           for(p2=p1+1;isalpha(*p2);p2++);
  223.           if(p2==p1+1 || p2>p1+24) {tt=1; break;}
  224.           memmove(buf,p1,p2-p1);buf[p2-p1]='\\';buf[p2-p1+1]=0;
  225.           for(i=0;i<backtransno && strcmp(buf,backtrans[i].name)!=0;i++);
  226.           if(i>=backtransno && strstr(hmsame,buf)==NULL) {
  227.             tt=1; break;
  228.           }
  229.       }
  230.       if(tt==0) {
  231.           for(p1=strchr(mathbuf,'\\'); p1; p1=strchr(p1+1,'\\')) {
  232.             for(p2=p1+1;isalpha(*p2);p2++);
  233.             if(p2==p1+1 || p2>p1+24) break;
  234.             memmove(buf,p1,p2-p1);buf[p2-p1]='\\';buf[p2-p1+1]=0;
  235.             for(i=0;i<backtransno && strcmp(buf,backtrans[i].name)!=0;i++);
  236.             if(i<backtransno)
  237.               string_modify(buf,p1,p2,backtrans[i].trans);
  238.             else *p1=' ';
  239.           }
  240.       }
  241.     }
  242. }
  243.  
  244. void output(void)
  245. {
  246.     char *p, *pp, *p2, *pt;
  247.     char buf[MAX_LINELEN+1];
  248.     p=filebuf;
  249.     restart:
  250.     pp=find_tag(p,"body"); if(*pp!=0) {
  251.       p=find_tag_end(pp); goto restart;
  252.     }
  253.     pp=find_tag(p,"html"); if(*pp!=0) {
  254.       p=find_tag_end(pp); goto restart;
  255.     }
  256.     *find_tag(p,"/body")=0; *find_tag(p,"/html")=0;
  257.     for(pp=strstr(p,"\n\n"); pp; pp=strstr(pp+1,"\n\n")) *pp=' ';
  258.     for(pp=strchr(p,'<');pp!=NULL;pp=strchr(find_tag_end(pp),'<')) {
  259.       if(pp>p) {fwrite(p,1,pp-p,outf); p=pp;}
  260.       if(latex2html && strncasecmp(pp,"<br><hr>",8)==0 &&
  261.          *find_word_start(pp+8)==0) break;
  262.       if(strncasecmp(pp+1,"!-- MATH",8)==0) {
  263.           p2=strstr(pp+8,"-->"); if(p2==NULL) continue;
  264.           *p2=0; getmath(pp+9); *p2='-';
  265.           p=p2+3; pt=find_word_start(p);
  266.           if(mathbuf[0] && strncasecmp(pt,"<IMG",4)==0 && isspace(*(pt+4))) {
  267.             p=find_tag_end(pt); pp=pt;
  268.             fprintf(outf,"\\(%s\\)",mathbuf);
  269.           }
  270.           continue;
  271.       }
  272.       if(strncasecmp(pp+1,"a",1)==0 && isspace(*(pp+2))) {
  273.  
  274.  
  275.  
  276.           continue;
  277.       }
  278.       if(strncasecmp(pp+1,"img",3)==0 && isspace(*(pp+4))) {
  279.           p2=find_tag_end(pp);
  280.           if(p2-pp>=MAX_LINELEN-256) continue;
  281.           memmove(buf,pp+1,p2-pp-2); buf[p2-pp-2]=0;
  282.           pt=strstr(buf,"ALT=\""); if(pt==NULL) pt=strstr(buf,"alt=\"");
  283.           if(pt!=NULL) {
  284.             pt+=strlen("ALT=\"");
  285.             getmath(pt); if(mathbuf[0]) {
  286.                 fprintf(outf,"\\(%s\\)",mathbuf); p=p2;
  287.             }
  288.           }
  289.       }
  290.     }
  291.     if(pp==NULL) fprintf(outf,"%s",p);
  292. }
  293.  
  294. int main(int argc, char *argv[])
  295. {
  296.     char *p, *pp;
  297.     char *mod;
  298.  
  299.     mod=getenv("w_module");
  300.     if(mod!=NULL && strncmp(mod,"adm/",4)!=0 && strcmp(mod,"home")!=0) return 1;
  301.     if(mod==NULL) p=argv[1]; else p=getenv("wims_exec_parm");
  302.     if(p==NULL || *p==0) return 1;
  303.     p=find_word_start(p); pp=find_word_end(p);
  304.     if(pp<=p || pp-p>sizeof(fn1)-1) return 1;
  305.     memmove(fn1,p,pp-p); fn1[pp-p]=0;
  306.     p=find_word_start(pp); pp=find_word_end(p);
  307.     if(pp<=p || pp-p>sizeof(fn2)-1) ovlstrcpy(fn2,fn1);
  308.     else {memmove(fn2,p,pp-p); fn2[pp-p]=0;}
  309.     prepare_file();
  310.     output();
  311.     fclose(outf);
  312.     return 0;
  313. }
  314.