How to partition 8 bits into two set of 4 bits each using bitset<>?

[abhinuke]

So I have been working on this code where I have to partition 8 bits into two set of 4 bits each.So far I have been trying my luck with bitset<> container class in C++.But it doesnt give me any ability copy first 4 bits of bitset<8> to first bitset<4> and remaining four bits to another one...

Here is the code that I have written so far...I want to convert a character to 8 bits and the divides these 8 bits into two sets of 4 bits each.

FILE * fil;
    fil = fopen("test", "r");
    do {
            ch = fgetc(fil);
            str[counter++] = ch;
        } while (ch != EOF);
    fclose(fil);

    struct binary {
        bitset < 8 > bitval;
        char chr;
        bitset < 4 > lhs;
        bitset < 4 > rhs;
    } bin[counter];


    for (i = 0; i < counter; i++) {
        bin[i].bitval = str[i];
        bin[i].chr = str[i];
    }
        for (i = 0; i < counter; i++) {    
            cout << bin[i].chr << "   " << bin[i].bitval<< endl;
        }

    for (i = 0; i < counter; i++) {
        for(j=0;j<4;j++)
        {
            bin[i].lhs[j]=bin[i].bitval[j];
        }
    }
    for (i = 0; i < counter; i++) {
        for(j=4;j < 8;j++)
        {
            bin[i].rhs[j]=bin[i].bitval[j];
        }
    }

This thing displays only the correct values of LHS and all values of RHS as 0000.How do I get the correct values of RHS bitset<4>?
Kindly help.
Thanks in advance.

[Oralloy]

Try changing line 34:

           bin[i].rhs[j-4]=bin[i].bitval[j];

Luck!
Oralloy

[abhinuke]

Thanks a ton...it worked fine...I hope I will be able to move ahead using Bitset<> container class.
By the way can you tell me what exactly was going wrong? Is it because of numbering conventions of bits in Bitset<> container?
Thanks again!

[Oralloy]

abhinuke,

I suppose you could call it a numbering convention problem. I don't believe there is a way with the bitset container to specify the lower and upper array bounds, so you're stuck with the zero based array model.

By the way, you could have also used simple bit fields unioned with a [icode]char[icode], something like this:

union BITZ
{
  char cValue;
  struct
  {
    unsigned  lhs : 4;
    unsigned  rhs : 4;
  }
};

Good Luck!
Oralloy

[Rabbit]

Basically, it's because

Bit 0 1 2 3 4 5 6 7
LHS 0 1 2 3
RHS         0 1 2 3

Your RHS array doesn't have 8 items so why are you trying to assign item 5 from the bitset to item 5 of RHS?