[Boost-users] Is the behaviour of the binary_from_base64 intended?

11 Apr 2007

      Hello!

I've run into a strange problem with encoding/decoding base64. Let's
have a look at the following code:

/////////////////////////////////////////////////////////
#include "boost/archive/iterators/base64_from_binary.hpp"
#include "boost/archive/iterators/binary_from_base64.hpp"
#include "boost/archive/iterators/transform_width.hpp"
#include <string>
#include <iostream>

using namespace std;
using namespace boost::archive::iterators;

typedef
    base64_from_binary<
        transform_width<string::const_iterator, 6, 8>
        > base64_t;

typedef
    transform_width<
        binary_from_base64<string::const_iterator>, 8, 6
        > binary_t;

int main()
{
    string str("Hello, world!");
    cout << str << endl;
    // This works quite well.
    string enc(base64_t(str.begin()), base64_t(str.end()));
    cout << enc << endl;
    // The following call throws an exception about unacceptable
    // base64 character in the stream. Probably a bug?
    //string dec(binary_t(enc.begin()), binary_t(enc.end()));
    // Hackish workaround:
    string dec(
        binary_t(enc.begin()),
        binary_t(enc.begin() + enc.length() - 1)
        );
    cout << dec << endl;
    return 0;
}
//////////////////////////////////////////////////////////////

Does those transforming iterators work as expected? Why the tailoring
zero is included in the transformation process? Is there a better way
to workaround the problem?

Thanks in advance!

-- Anatoli Sakhnik.