Hecademical

[Tomheza]

Hello guys, I have a task that requires convertation between hecademical to 8th 10th and binary. And I face a problem that A = 10; B = 11 and so on. I cannot use integer and as far as i used string i cant really use string since if I try to make mathematical action it takes ASCII number instead of number[i]. Could you give me some ideas or hints?

[SakurasouBusters]

int toHex(char H)
{
    if(isxdigit(H))
    return (H >= '0' && H <= '9') ? (H - '0') : (10 + H - 'A');
    return 0;
}

[keskiverto]

Where does the data come from?
User input / file? There is formatted input option: http://www.cplusplus.com/reference/ios/ios_base/fmtflags/
Computed string? http://www.cplusplus.com/reference/string/stoi/

[SakurasouBusters]

This one is better :

int toHex(char H)
{
    if(isxdigit(H))
    return (H <= '9') ? ('0' + -H) : (std::toupper(H) - 'A' + 10);
    return 0;
}

[Tomheza]

The user will have to choose from which system to which he wants and then he will input it.

[SakurasouBusters]

The user will have to choose from which system to which he wants and then he will input it.

What is that?

[whitenite1]

@SakurasouBusters

I take it, that the user will be asked, "What is the original number in? ( Binary, Octal, Decimal, or Hexidecimal)".
Then they tell what to convert that number, into.

[keskiverto]

http://www.cplusplus.com/reference/ios/hex/

#include <iostream>

int main () {
  int n = 0;
  std::cout << "hex ";
  std::cin >> std::hex >> n;
  // whatever the user did type, was read as a hex value
  return 0;
}

[mbozzi]

Here is a snippet of code which will perform numeric conversions to an arbitrary base, digit by digit.

Floating point accuracy will become an issue if the values you put in are large. I haven't done the math to determine how large, though.

# include <iostream>
# include <cmath>

int main (int, char **) {
  unsigned N = 10;  /* Input value. */

  /* Radix of the output. */
  constexpr uint64_t radix = 2;

  // When radix > 9 you'll not see your additional symbols because there's no
  // logic for printing more than 0-9.  For instance, 0xA will echo as 10 -- referring to the
  // eleventh symbol in your alphabet.  If you want, you can use an array or map
  // to perform the conversion.
  //
  // Extra points for getting this to work for negative and fractional
  // values.  More extra points for doing the math to figure out when IEEE754
  // binary64 precision problems causes the conversion to be inexact.
  for (int place = ceil (log(N) / log(radix)); place >= 0; place--)
    std:: cout << static_cast <uint64_t> (N / pow(radix, place)) % radix
	       << " ";
}

[Tomheza]

And how should i cope with it if I need to transform float numbers as well? I tried to make the div and mod parts two different integers but couldnt use % function since its only suitable for integers.

[mbozzi]

std::fmod() will compute the modulus for you.