/* 4. zadatak - Tijana Radovanovic - 158/2000 */


 /* program se poziva sa:  DEKODER Ulaz.htm Izlaz.htm
    gde je:
	   DEKODER - ime izvrsnog programa
	   Ulaz(.htm) - ime ulazne datoteke tj. seminarskog
	   Izlaz(.htm) - ime izlazne datoteke
    napomena:
    izlazna datoteka ce sadrzati tekst preveden u ISO-8859-1 (trazenu kodnu semu)
    i to samo ako program preko META-taga uspe da identifikuje kodnu semu kojom je tekst unet */





#include <stdio.h>


   /* Uporedjivanje dva slova nezavisno od njihove velicine */

int uchar(char a, char b)
{
 if ((a>64) && (a<91))
  a=a+32;
 if ((b>64) && (b<91))
  b=b+32;
 if (a==b)
  return 1;
 else
  return 0;
}


   /* Uporedjivanje dva stringa preko uchar funkcije */

int ustring(char *a, char *b)
{
 int l;
 int e=1;
 int lx=0;
 while (a[lx]!='\0')
 {
  lx++;
 }

 for (l=0;l<lx;l++)
 {
  int ec=uchar(a[l],b[l]);
  if (!ec)
  {
   e=0;
  }
 }
 return e;
}


   /* Pretraga u fajlu do prvog pojavljivanja "charset" reci i vracanje njene pozicije */

long gotocharset(FILE *ulaz)
{
 char c;
 int st;
 long poz=0;

 int exit=1;

 char te[7];
  te[0]=' ';
  te[1]=' ';
  te[2]=' ';
  te[3]=' ';
  te[4]=' ';
  te[5]=' ';
  te[6]=' ';
  te[7]='\0';

 while ((!feof(ulaz)) && (exit!=0))
 {
  te[0]=te[1];
  te[1]=te[2];
  te[2]=te[3];
  te[3]=te[4];
  te[4]=te[5];
  te[5]=te[6];
  c=fgetc(ulaz);
  if (c<32)
   te[6]=' ';
  else
   te[6]=c;
  st=ustring(te,"charset");
  if (st)
  {
   exit=0;
  }
  poz++;
 }
 if ((feof(ulaz)) && (exit==1))
  poz=-2;

 c=fgetc(ulaz);
 poz++;
 return poz;
}


   /* Funkcija koja ispituje kojom kodnom semom je pisan tekst i vracanje te vrednosti nazad */

int getcharset(FILE *ulaz)
{
 char c;
 int st;
 int rr=0;
 char is[10];
 char wi[12];
 char w;
 long poz;

 int g;

 poz=gotocharset(ulaz);
 if (poz==-1)
 {
  return -1;
 }

 wi[0]=fgetc(ulaz);
 wi[1]=fgetc(ulaz);
 wi[2]=fgetc(ulaz);
 wi[3]=fgetc(ulaz);
 wi[4]=fgetc(ulaz);
 wi[5]=fgetc(ulaz);
 wi[6]=fgetc(ulaz);
 wi[7]=fgetc(ulaz);
 wi[8]=fgetc(ulaz);
 wi[9]=fgetc(ulaz);
 wi[10]=fgetc(ulaz);
 w=fgetc(ulaz);

 for (g=0;g<10;g++)
  is[g]=wi[g];

 st=ustring(is,"iso-8859-2");
 if (st)
 {
  rr=1;
 }

 st=ustring(is,"iso-8859-5");
 if (st)
 {
  rr=2;
 }

 st=ustring(is,"windows-12");
 if ((st) && (wi[10]=='5') && (w=='0'))
 {
  rr=3;
 }

 st=ustring(is,"windows-12");
 if ((st) && (wi[10]=='5') && (w=='1'))
 {
  rr=4;
 }

 return rr;
}


   /* Konvertovanje teksta u ISO-8859-1
      (u META-tagu 'prevedenog' teksta ce biti oznaka za ovu kodnu semu)
	cs - kodna sema
	poz - pozicija stringa stare kodne seme */

void konvertuj(FILE *ulaz, FILE *izlaz, int cs, long poz)


{
 char str;
 int j,bra,prom;
 long p=-1;

 int znak;
 char tb[60];
 char *chr[60];
 znak=0;

 if (cs==1)
 {
  znak=9;
  tb[0] = 0xC8; chr[0] = "&Cx;";
  tb[1] = 0xD0; chr[1] = "&Dx;";
  tb[2] = 0xAE; chr[2] = "&Zx;";
  tb[3] = 0xC6; chr[3] = "&Cy;";
  tb[4] = 0xA9; chr[4] = "&Sx;";
  tb[5] = 0xE8; chr[5] = "&cx;";
  tb[6] = 0xF0; chr[6] = "&dx;";
  tb[7] = 0xBE; chr[7] = "&zx;";
  tb[8] = 0xE6; chr[8] = "&cy;";
  tb[9] = 0xB9; chr[9] = "&sx;";
 }

 if (cs==2)
 {
  znak=59;
  tb[ 0] = 0xB0; chr[ 0] = "A";
  tb[ 1] = 0xB1; chr[ 1] = "B";
  tb[ 2] = 0xB2; chr[ 2] = "V";
  tb[ 3] = 0xB3; chr[ 3] = "G";
  tb[ 4] = 0xB4; chr[ 4] = "D";
  tb[ 5] = 0xA2; chr[ 5] = "&Dx;";
  tb[ 6] = 0xB5; chr[ 6] = "E";
  tb[ 7] = 0xB6; chr[ 7] = "&Zx;";
  tb[ 8] = 0xB7; chr[ 8] = "Z";
  tb[ 9] = 0xB8; chr[ 9] = "I";
  tb[10] = 0xA8; chr[10] = "J";
  tb[11] = 0xBA; chr[11] = "K";
  tb[12] = 0xBB; chr[12] = "L";
  tb[13] = 0xA9; chr[13] = "&Lx;";
  tb[14] = 0xBC; chr[14] = "M";
  tb[15] = 0xBD; chr[15] = "N";
  tb[16] = 0xAA; chr[16] = "&Nx;";
  tb[17] = 0xBE; chr[17] = "O";
  tb[18] = 0xBF; chr[18] = "P";
  tb[19] = 0xC0; chr[19] = "R";
  tb[20] = 0xC1; chr[20] = "S";
  tb[21] = 0xC2; chr[21] = "T";
  tb[22] = 0xAB; chr[22] = "&Cy;";
  tb[23] = 0xC3; chr[23] = "U";
  tb[24] = 0xC4; chr[24] = "F";
  tb[25] = 0xC5; chr[25] = "H";
  tb[26] = 0xC6; chr[26] = "C";
  tb[27] = 0xC7; chr[27] = "&Cx;";
  tb[28] = 0xAF; chr[28] = "&Dy;";
  tb[29] = 0xC8; chr[29] = "&Sx;";
  tb[30] = 0xD0; chr[30] = "a";
  tb[31] = 0xD1; chr[31] = "b";
  tb[32] = 0xD2; chr[32] = "v";
  tb[33] = 0xD3; chr[33] = "g";
  tb[34] = 0xD4; chr[34] = "d";
  tb[35] = 0xF2; chr[35] = "&dx;";
  tb[36] = 0xD5; chr[36] = "e";
  tb[37] = 0xD6; chr[37] = "&zx;";
  tb[38] = 0xD7; chr[38] = "z";
  tb[39] = 0xD8; chr[39] = "i";
  tb[40] = 0xF8; chr[40] = "j";
  tb[41] = 0xDA; chr[41] = "k";
  tb[42] = 0xDB; chr[42] = "l";
  tb[43] = 0xF9; chr[43] = "&lx;";
  tb[44] = 0xDC; chr[44] = "m";
  tb[45] = 0xDD; chr[45] = "n";
  tb[46] = 0xFA; chr[46] = "&nx;";
  tb[47] = 0xDE; chr[47] = "o";
  tb[48] = 0xDF; chr[48] = "p";
  tb[49] = 0xE0; chr[49] = "r";
  tb[50] = 0xE1; chr[50] = "s";
  tb[51] = 0xE2; chr[51] = "t";
  tb[52] = 0xFB; chr[52] = "&cy;";
  tb[53] = 0xE3; chr[53] = "u";
  tb[54] = 0xE4; chr[54] = "f";
  tb[55] = 0xE5; chr[55] = "h";
  tb[56] = 0xE6; chr[56] = "c";
  tb[57] = 0xE7; chr[57] = "&cx;";
  tb[58] = 0xFF; chr[58] = "&dy;";
  tb[59] = 0xE8; chr[59] = "&sx;";
 }

 if (cs==3)
 {
  znak=9;
  tb[0] = 0xC8; chr[0] = "&Cx;";
  tb[1] = 0xD0; chr[1] = "&Dx;";
  tb[2] = 0x8E; chr[2] = "&Zx;";
  tb[3] = 0xC6; chr[3] = "&Cy;";
  tb[4] = 0x8A; chr[4] = "&Sx;";
  tb[5] = 0xE8; chr[5] = "&cx;";
  tb[6] = 0xF0; chr[6] = "&dx;";
  tb[7] = 0x9E; chr[7] = "&zx;";
  tb[8] = 0xE6; chr[8] = "&cy;";
  tb[9] = 0x9A; chr[9] = "&sx;";
 }

 if (cs==4)
 {
  znak=59;
  tb[ 0] = 0xC0; chr[ 0] = "A";
  tb[ 1] = 0xC1; chr[ 1] = "B";
  tb[ 2] = 0xC2; chr[ 2] = "V";
  tb[ 3] = 0xC3; chr[ 3] = "G";
  tb[ 4] = 0xC4; chr[ 4] = "D";
  tb[ 5] = 0x80; chr[ 5] = "&Dx;";
  tb[ 6] = 0xC5; chr[ 6] = "E";
  tb[ 7] = 0xC6; chr[ 7] = "&Zx;";
  tb[ 8] = 0xC7; chr[ 8] = "Z";
  tb[ 9] = 0xC8; chr[ 9] = "I";
  tb[10] = 0xA3; chr[10] = "J";
  tb[11] = 0xCA; chr[11] = "K";
  tb[12] = 0xCB; chr[12] = "L";
  tb[13] = 0x8A; chr[13] = "&Lx;";
  tb[14] = 0xCC; chr[14] = "M";
  tb[15] = 0xCD; chr[15] = "N";
  tb[16] = 0x8C; chr[16] = "&Nx;";
  tb[17] = 0xCE; chr[17] = "O";
  tb[18] = 0xCF; chr[18] = "P";
  tb[19] = 0xD0; chr[19] = "R";
  tb[20] = 0xD1; chr[20] = "S";
  tb[21] = 0xD2; chr[21] = "T";
  tb[22] = 0x8E; chr[22] = "&Cy;";
  tb[23] = 0xD3; chr[23] = "U";
  tb[24] = 0xD4; chr[24] = "F";
  tb[25] = 0xD5; chr[25] = "H";
  tb[26] = 0xD6; chr[26] = "C";
  tb[27] = 0xD7; chr[27] = "&Cx;";
  tb[28] = 0x8F; chr[28] = "&Dy;";
  tb[29] = 0xD8; chr[29] = "&Sx;";
  tb[30] = 0xE0; chr[30] = "a";
  tb[31] = 0xE1; chr[31] = "b";
  tb[32] = 0xE2; chr[32] = "v";
  tb[33] = 0xE3; chr[33] = "g";
  tb[34] = 0xE4; chr[34] = "d";
  tb[35] = 0x90; chr[35] = "&dx;";
  tb[36] = 0xE5; chr[36] = "e";
  tb[37] = 0xE6; chr[37] = "&zx;";
  tb[38] = 0xE7; chr[38] = "z";
  tb[39] = 0xE8; chr[39] = "i";
  tb[40] = 0xBC; chr[40] = "j";
  tb[41] = 0xEA; chr[41] = "k";
  tb[42] = 0xEB; chr[42] = "l";
  tb[43] = 0x9A; chr[43] = "&lx;";
  tb[44] = 0xEC; chr[44] = "m";
  tb[45] = 0xED; chr[45] = "n";
  tb[46] = 0x9C; chr[46] = "&nx;";
  tb[47] = 0xEE; chr[47] = "o";
  tb[48] = 0xEF; chr[48] = "p";
  tb[49] = 0xF0; chr[49] = "r";
  tb[50] = 0xF1; chr[50] = "s";
  tb[51] = 0xF2; chr[51] = "t";
  tb[52] = 0x9E; chr[52] = "&cy;";
  tb[53] = 0xF3; chr[53] = "u";
  tb[54] = 0xF4; chr[54] = "f";
  tb[55] = 0xF5; chr[55] = "h";
  tb[56] = 0xF6; chr[56] = "c";
  tb[57] = 0xF7; chr[57] = "&cx;";
  tb[58] = 0x9F; chr[58] = "&dy;";
  tb[59] = 0xF8; chr[59] = "&sx;";
 }


 str=fgetc(ulaz);
 bra=0;
 while (feof(ulaz)==0)
 {
  p++;
  if (p==poz)
  {
   if ((cs==1) || (cs==2))
   {
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
   }
   if ((cs==3) || (cs==4))
   {
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
    str=fgetc(ulaz);
   }
   fputs("iso-8859-1",izlaz);
  }
  if (bra==1)
  {
   fputc(str,izlaz);
   if (str=='>')
    bra=0;
  }
  if (bra==0)
  {
   prom=0;
   j=0;
   while (j<=znak)
   {
    if (str==tb[j])
    {
     prom=1;
     fputs(chr[j],izlaz);
     if (str=='<')
      bra=1;
    }
    j++;
   }
   if (prom==0)
    fputc(str,izlaz);
  }
  str=fgetc(ulaz);
 }
}




int main(int brarg, char **argum)
{

 FILE *ulaz;
 FILE *izlaz;

 int cs;
 long po;

 if (brarg<3)
 {
  printf("Morate navesti ime ulazne i ime izlazne datoteke! \n");
  return 1;
 }


 ulaz = fopen(argum[1], "rt");
 izlaz = fopen(argum[2], "wt");

 cs=getcharset(ulaz);

 switch(cs)
 {
  case -1:printf("Ne postoji <META> tag ! \n(informacija o kodnoj semi bi trebalo da se nalazi u zaglavlju datoteke u okviru META-taga)");
   break;
  case 0:printf("Koriscen je ISO-8859-1 ili nepostojeca kodna sema");
   break;
  case 1:printf("Tekst je kodiran prema semi ISO-8859-2");
   break;
  case 2:printf("Tekst je kodiran prema semi ISO-8859-5");
   break;
  case 3:printf("Tekst je kodiran prema semi CP-1250");
   break;
  case 4:printf("Tekst je kodiran prema semi CP-1251");
   break;
 }


 fclose(ulaz);
 ulaz = fopen(argum[1], "rt");

 po=gotocharset(ulaz);
 fclose(ulaz);
 ulaz = fopen(argum[1], "rt");

 if (cs>0)
 {
  konvertuj(ulaz,izlaz,cs,po);
 }


 fcloseall();

 return 0;
}