WebSVN – wimsdev – Diff – /trunk/wims/src/Misc/modind.c

 char *dicdir=           "public_html/bases";
 char *outdir=           "public_html/bases/site2";
 char *maindic=          "sys/words";
 char *groupdic=         "sys/wgrp/wgrp";
 char *suffixdic=        "sys/suffix";
+char *domaindic=        "sys/domaindic";
 char *ignoredic=        "sys/indignore";
 char *conffile=         "log/wims.conf";
 char *mlistbase=        "list";
 char lang[MAX_LANGS][4]={
         /* Find first occurrence of word */
 char *wordchr(char *p, char *w)
+{
     char *r;
     for(r=strstr(p,w);r!=NULL &&
         ( (r>p && !isspace(*(r-1))) || (!isspace(*(r+strlen(w))) && *(r+strlen(w))!=0) );
         r=strstr(r+1,w));
     return r;
+}
         /* modify a string. Bufferlen must be ast least MAX_LINELEN */
 void string_modify(char *start, char *bad_beg, char *bad_end, char *good,...)
+{
     char buf[MAX_LINELEN+1];
     va_list vp;
     va_start(vp,good);
     vsnprintf(buf,sizeof(buf),good,vp); va_end(vp);
     if(strlen(start)-(bad_end-bad_beg)+strlen(buf)>=MAX_LINELEN)
       return;
     strcat(buf,bad_end);
 void comma(char *p)
+{
     char *pp;
     for(pp=strchr(p,','); pp; pp=strchr(pp+1,','))
       string_modify(p,pp,pp+1,", ");
 }
 void _getdef(char buf[], char *name, char value[])
+{
     char *p1, *p2, *p3;
 void getdef(char *fname, char *name, char value[])
+{
     FILE *f;
     char *buf;
     int l;
     value[0]=0;
     f=fopen(fname,"r"); if(f==NULL) return;
     fseek(f,0,SEEK_END); l=ftell(f); fseek(f,0,SEEK_SET);
     buf=xmalloc(l+256); l=fread(buf,1,l,f);
     fclose(f);
     free(buf);
+}
 #include "translator_.c"
-char *mdicbuf, *gdicbuf;
+char *mdicbuf, *gdicbuf, *ddicbuf;
-char gentry[sizeof(entry)], mentry[sizeof(entry)];
+char gentry[sizeof(entry)], mentry[sizeof(entry)], dentry[sizeof(entry)];
-int gentrycount, mentrycount;
+int gentrycount, mentrycount, dentrycount;
         /* Preparation of data */
 void prep(void)
+{
     char buf[MAX_LINELEN+1];
     char *p1,*p2,*s,*old;
     int i,l,thislang,t;
     FILE *f;
     s=getenv("modind_outdir"); if(s!=NULL && *s!=0) outdir=s;
     s=getenv("modind_sheetdir"); if(s!=NULL && *s!=0) sheetdir=s;
     snprintf(buf,sizeof(buf),"%s/addr",outdir);
     addrf=fopen(buf,"w");
     snprintf(buf,sizeof(buf),"%s/serial",outdir);
 void sprep(void)
+{
     char *p1,*p2,*s;
     int i,l,thislang;
     modcnt=0;
     s=getenv("slist"); if(s==NULL) return;
     l=strlen(s); if(l<0 || l>100*MAX_LINELEN) return;
     mlist=xmalloc(l+16); ovlstrcpy(mlist,s);
     for(p1=find_word_start(mlist); *p1 && modcnt<MAX_MODULES; p1=find_word_start(p2)) {
     fclose(langf); fclose(titf); fclose(descf); fclose(robotf);
     fclose(authorf); fclose(versionf);
+}
 char *sheetindex[]={
       "title", "description",
       "duration", "severity",
       "level", "domain",
       "keywords", "reserved1", "reserved2", "remark"
 };
 #define SHEETINDEX_NO (sizeof(sheetindex)/sizeof(sheetindex[0]))
       s_keywords, s_reserved1, s_reserved2,
       s_remark
 };
 char *modindex[]={
       "title", "description",
       "author", "address", "copyright",
       "version", "wims_version", "language",
       "category", "level", "domain", "keywords",
       "keywords_ca", "keywords_en", "keywords_fr", "keywords_it", "keywords_nl",
       "title_ca", "title_en", "title_fr", "title_it", "title_nl",
       "require"
 };
 #define MODINDEX_NO (sizeof(modindex)/sizeof(modindex[0]))
+{
     char nbuf[MAX_LINELEN+1], buf[MAX_LINELEN+1];
     int i, ll;
     char *p;
     FILE *f;
     if(!isalnum(*word) || (ll=strlen(word))<2 ||
        wordchr(taken,word)!=NULL ||
        wordchr(ignore[lind],word)!=NULL ||
        takenlen>=MAX_LINELEN-ll-16)
       return;
         f=fopen(nbuf,"a");
         if(f!=NULL) {fputs(buf,f); fclose(f);}
+    }
+}
+void appenditem1 (char *buf, int lind, int serial, int weight, char *l )
+{
+  char *p1, *p2 ;
+  for(p1=find_word_start(buf); *p1;
+        p1=find_word_start(p2)) {
+        p2=strchr(p1,',');
+        if(p2!=NULL) *p2++=0; else p2=p1+strlen(p1);
+        if(strlen(p1)<=0) continue;
+        appenditem(p1,lind,serial,weight,module_language);
+  }
+}
+void appenditem2 (char *buf, int lind, int serial, int weight, char *l )
+{
+  char *p1, *p2 ;
+  for(p1=find_word_start(buf);*p1;
+        p1=find_word_start(p2)) {
+        p2=find_word_end(p1); if(*p2) *p2++=0;
+        appenditem(p1,lind,serial,weight,module_language);
+  }
+}
 void onemodule(const char *name, int serial, int lind)
+{
     int i;
     unsigned char trlist[]={
         i_title,i_description,i_category,i_domain,i_keywords,
           i_title_ca,i_title_en,i_title_fr,i_title_it,i_title_nl
     };
     #define trcnt (sizeof(trlist)/sizeof(trlist[0]))
     char *p1, *p2, *pp, *q, buf[MAX_LINELEN+1], lbuf[16];
     FILE *f;
     if(module_index(name)) return;
     towords(indbuf[i_category]);
         /*  list the categories (among A=all,X=eXercise,O,D,...) corresponding to this module  */
     for(i=catcnt=0;i<catno && catcnt<16;i++) {
         if(wordchr(indbuf[i_category],cat[i].name)!=NULL)
           categories[catcnt++]=cat[i].typ;
+    }
     if(catcnt==0) return;
     if(categories[0]!=cat[0].typ)
       categories[catcnt++]=cat[0].typ;
     fprintf(langf,"%d:%s\n",serial,module_language);
     fprintf(titf,"%d:%s\n",serial,indbuf[i_title]);
     fprintf(descf,"%d:%s\n",serial,indbuf[i_description]);
     fprintf(authorf,"%d:%s\n",serial,indbuf[i_author]);
     fprintf(versionf,"%d:%s\n",serial,indbuf[i_version]);
         /*  add module's information in html page for robots  */
     snprintf(buf,sizeof(buf),"%s",indbuf[i_description]);
     for(pp=strchr(buf,','); pp; pp=strchr(pp,','))
       string_modify(buf,pp,pp+1,"&#44;");
     if(strcmp(module_language,lang[lind])==0)
       fprintf(robotf,"%s ,%s,%s,%s,%s\n",name,module_language,name,
               indbuf[i_title], buf);
-        /*  Normalize the information, using main dictionary bases/sys/words.xx */
-    entrycount=mentrycount; dicbuf=mdicbuf;
-    memmove(entry,mentry,mentrycount*sizeof(entry[0]));
-    unknown_type=unk_leave;  /* used in translator_.c */
+        /*  Normalize the information, using dictionary
+        -- bases/sys/domain.xx without suffix (--> english version)
+        -- bases/sys/words.xx with suffix */
+    entrycount=dentrycount; dicbuf=ddicbuf;
+    memmove(entry,dentry,dentrycount*sizeof(entry[0]));
+    unknown_type=unk_leave;
     for(i=0;i<trcnt;i++) {
         detag(indbuf[trlist[i]]);
         deaccent(indbuf[trlist[i]]);
         comma(indbuf[trlist[i]]);
         singlespace(indbuf[trlist[i]]);
+        translate(indbuf[trlist[i]]);
+    }
+    entrycount=mentrycount; dicbuf=mdicbuf;
+    memmove(entry,mentry,mentrycount*sizeof(entry[0]));
+    unknown_type=unk_leave;  /* used in translator_.c */
+    for(i=0;i<trcnt;i++) {
         suffix_translate(indbuf[trlist[i]]);
         translate(indbuf[trlist[i]]);
+    }
-        /*  append words of title  */
+/* taken contains all words already seen in the module index */
     taken[0]=0; takenlen=tweight=0;
+/*  append words of title  */
     ovlstrcpy(buf,indbuf[i_title]); towords(buf);
-    for(p1=find_word_start(buf);*p1;
-        p1=find_word_start(p2)) {
-        p2=find_word_end(p1); if(*p2) *p2++=0;
-        appenditem(p1,lind,serial,4,module_language);
+    appenditem2(buf,lind,serial,4,module_language);
-    }
-        /*  append words of every other information except level  */
+/*  append words of every other information except level  */
     snprintf(buf,sizeof(buf),"%s %s %s %s %s %s %s %s %s %s %s %s %s %s %s",
              indbuf[i_description],indbuf[i_keywords],
              indbuf[i_keywords_ca],indbuf[i_keywords_en],indbuf[i_keywords_fr],
              indbuf[i_keywords_it],indbuf[i_keywords_nl],
              indbuf[i_title_ca],indbuf[i_title_en],indbuf[i_title_fr],
              indbuf[i_title_it],indbuf[i_title_nl],
              indbuf[i_domain],indbuf[i_require],indbuf[i_author]);
     towords(buf);
-    for(p1=find_word_start(buf);*p1;
-        p1=find_word_start(p2)) {
-        p2=find_word_end(p1); if(*p2) *p2++=0;
-        appenditem(p1,lind,serial,2,module_language);
+    appenditem2(buf,lind,serial,4,module_language);
-    }
-        /*  this time the dictionary is the group dictionary  sys/wgrp/wgrpwith a g (=global ? general ?), not an m . see below main,suffix,group.
+        /*  this time the dictionary is the group dictionary  sys/wgrp/wgrp
+         with a g (groupdic), not an m (maindic) . see below main, suffix, group.
         and delete unknown ?? and translate  */
     entrycount=gentrycount; dicbuf=gdicbuf;
     memmove(entry,gentry,gentrycount*sizeof(entry[0]));
+/*  append words (?) of every other information except level  */
+    ovlstrcpy(buf,indbuf[i_title]);
     unknown_type=unk_delete;
+    translate(buf);
+    appenditem1(buf,lind,serial,2,module_language);
-        /*  append words (?) of every other information except level  */
+/*  append words (?) of information of description except level  */
-    ovlstrcpy(buf,indbuf[i_title]); translate(buf);
+    snprintf(buf,sizeof(buf),"%s", indbuf[i_description]);
-    for(p1=find_word_start(buf); *p1;
+    unknown_type=unk_delete;
-        p1=find_word_start(p2)) {
-        p2=strchr(p1,',');
+    translate(buf);
-        if(p2!=NULL) *p2++=0; else p2=p1+strlen(p1);
-        if(strlen(p1)<=0) continue;
-        appenditem(p1,lind,serial,4,module_language);
+    appenditem1(buf,lind,serial,4,module_language);
-    }
-        /*  append words (?) of every other information except level  */
+/*  append words (or group of words) of keywords and domain level  */
-    snprintf(buf,sizeof(buf),"%s, %s, %s, %s, %s, %s, %s, %s",
+    snprintf(buf,sizeof(buf),"%s, %s, %s, %s, %s, %s, %s",
-             indbuf[i_description],indbuf[i_keywords],
+             indbuf[i_domain],indbuf[i_keywords],
              indbuf[i_keywords_ca], indbuf[i_keywords_en],indbuf[i_keywords_fr],
-             indbuf[i_keywords_it], indbuf[i_keywords_nl],
+             indbuf[i_keywords_it], indbuf[i_keywords_nl]);
-             indbuf[i_domain]);
+        unknown_type=unk_leave;
     translate(buf);
-    for(p1=find_word_start(buf); *p1;
-        p1=find_word_start(p2)) {
-        p2=strchr(p1,',');
-        if(p2!=NULL) *p2++=0; else p2=p1+strlen(p1);
-        if(strlen(p1)<=0) continue;
-        appenditem(p1,lind,serial,2,module_language);
+    appenditem1(buf,lind,serial,2,module_language);
-    }
         /*  append level information, with weight 2 */
     snprintf(buf,sizeof(buf),"%s",indbuf[i_level]);
     ovlstrcpy(lbuf,"level");
     for(p1=buf; *p1; p1++) if(!isalnum(*p1)) *p1=' ';
     q=buf+strlen(buf);
     for(p1=find_word_start(buf); (*p1) && (p1 < q) ;
         p1=find_word_start(p2)) {
         p2=find_word_end(p1);
         if(p2!=NULL) *p2++=0; else p2=p1+strlen(p1);
         if(!isalpha(*p1) ||
            (!isdigit(*(p1+1)) && *(p1+1)!=0) ||
            (*(p1+1)!=0 && *(p1+2)!=0))
           continue;
         *p1=tolower(*p1);
         ovlstrcpy(lbuf+strlen("level"),p1);
 void modules(void)
+{
     int i,j,k,d;
     char namebuf[MAX_LINELEN+1];
-    char mdic[MAX_LINELEN+1], sdic[MAX_LINELEN+1], gdic[MAX_LINELEN+1];
+    char mdic[MAX_LINELEN+1], sdic[MAX_LINELEN+1], gdic[MAX_LINELEN+1], ddic[MAX_LINELEN+1];
     for(j=0;j<langcnt;j++) {
         snprintf(namebuf,sizeof(namebuf),"%s/weight.%s",outdir,lang[j]);
         weightf=fopen(namebuf,"w");
         snprintf(mdic,sizeof(mdic),"%s/%s.%s",dicdir,maindic,lang[j]);
         snprintf(sdic,sizeof(sdic),"%s/%s.%s",dicdir,suffixdic,lang[j]);
         snprintf(gdic,sizeof(gdic),"%s/%s.%s",dicdir,groupdic,lang[j]);
+        snprintf(ddic,sizeof(ddic),"%s/%s.%s",dicdir,domaindic,lang[j]);
         suffix_dic(sdic); prepare_dic(gdic);
         gdicbuf=dicbuf; gentrycount=entrycount;
         memmove(gentry,entry,gentrycount*sizeof(entry[0]));
         prepare_dic(mdic);
         mdicbuf=dicbuf; mentrycount=entrycount;
         memmove(mentry,entry,mentrycount*sizeof(entry[0]));
+        prepare_dic(ddic);
+        ddicbuf=dicbuf; dentrycount=entrycount;
+        memmove(dentry,entry,dentrycount*sizeof(entry[0]));
         unknown_type=unk_leave; translate(ignore[j]);
         for(i=0;i<modcnt;i++) {
             if(mod[i].langcnt>0) {
                 for(d=k=0;k<mod[i].langcnt;k++)
                   if(mod[i].langs[k]<mod[i].langs[d]) d=k;
+            }
+        }
         if(mentrycount>0) free(mdicbuf);
         if(gentrycount>0) free(gdicbuf);
         if(suffixcnt>0) free(sufbuf);
+        if(dentrycount>0) free(ddicbuf);
         if(weightf) fclose(weightf);
+    }
+}
+/* FIXME ? differences with appenditem - use fprintf instead of  snprintf */
 void sappenditem(char *word, int lind, int serial, int weight)
+{
     int ll;
     char *p;
     if(!isalnum(*word) || (ll=strlen(word))<2 ||
        wordchr(taken,word)!=NULL ||
        wordchr(ignore[lind],word)!=NULL ||
        takenlen>=MAX_LINELEN-ll-16)
       return;
     unsigned char trlist[]={
         s_title,s_description,s_domain,s_keywords,s_remark
     };
     #define trcnt (sizeof(trlist)/sizeof(trlist[0]))
     char *p1, *p2, buf[MAX_LINELEN+1];
     if(sheet_index(serial)) return;
     fprintf(listf,"%s\n",mod[serial].name+3);
     fprintf(titf,"%d:%s\n",serial,sindbuf[s_title]);
     fprintf(descf,"%d:%s\n",serial,sindbuf[s_description]);
-    entrycount=mentrycount; dicbuf=mdicbuf;
+    entrycount=dentrycount; dicbuf=ddicbuf;
-    memmove(entry,mentry,mentrycount*sizeof(entry[0]));
+    memmove(entry,dentry,dentrycount*sizeof(entry[0]));
     unknown_type=unk_leave;
     for(i=0;i<trcnt;i++) {
         detag(sindbuf[trlist[i]]);
         deaccent(sindbuf[trlist[i]]);
         comma(sindbuf[trlist[i]]);
         singlespace(sindbuf[trlist[i]]);
+        translate(sindbuf[trlist[i]]);
+    }
+    entrycount=mentrycount; dicbuf=mdicbuf;
+    memmove(entry,mentry,mentrycount*sizeof(entry[0]));
+    unknown_type=unk_leave;
+    for(i=0;i<trcnt;i++) {
         suffix_translate(sindbuf[trlist[i]]);
         translate(sindbuf[trlist[i]]);
+    }
     taken[0]=0; takenlen=tweight=0;
     ovlstrcpy(buf,sindbuf[s_title]); towords(buf);
     for(p1=find_word_start(buf);*p1;
         p1=find_word_start(p2)) {
         p2=find_word_end(p1); if(*p2) *p2++=0;
         sappenditem(p1,lind,serial,4);
+    }
     snprintf(buf,sizeof(buf),"%s %s %s %s",
              sindbuf[s_description],sindbuf[s_keywords],
              sindbuf[s_domain],sindbuf[s_remark]);
              sindbuf[s_description],sindbuf[s_keywords],
              sindbuf[s_domain],sindbuf[s_remark]);
     translate(buf);
     for(p1=find_word_start(buf); *p1;
         p1=find_word_start(p2)) {
         p2=strchr(p1,',');
         if(p2!=NULL) *p2++=0; else p2=p1+strlen(p1);
         if(strlen(p1)<=0) continue;
         sappenditem(p1,lind,serial,2);
+    }
     fprintf(weightf,"%d:%d\n",serial,tweight);
+}
 void sheets(void)
+{
     int i,j;
     char mdic[MAX_LINELEN+1], sdic[MAX_LINELEN+1], gdic[MAX_LINELEN+1];

Subversion Repositories wimsdev

(root)/trunk/wims/src/Misc/modind.c – Rev 6819 → 6881