Conversion from UTF-8 encoded string to bytes and vice versa in C++

Question

In C#, we have following functions to convert a string to a UTF-8 encoded sequence of bytes and vice-versa:

Encoding.UTF8.GetString(Byte[])
Encoding.UTF8.GetBytes(Char[]) / Encoding.UTF8.GetBytes(String)

I am trying to achieve the same thing in C++, as follows:

std::string GetStringFromBytes(std::vector<uint8_t> bytes){
    std::string str(bytes.begin(), bytes.end());
    return str;
}

std::vector<uint8_t> GetBytesFromString(const std::string& str){
    std::vector<uint8_t> bytes(str.begin(), str.end());
    return bytes;
}

Is this approach correct? I'm assuming that the string that I'm converting is already in UTF-8 format.

`Is this approach correct?` - have you tried it and looked at the result? — Franz Gleichmann, Jan 22 '22 at 22:30
`std::string` already has the bytes, you don't have to copy them, just use them. You are hardly converting anything, you are copying. — KamilCuk, Jan 23 '22 at 00:26

score 4 · Answer 1 · answered Jan 23 '22 at 00:25

C# string uses UTF-16, and thus requires a charset conversion to/from UTF-8.

C++ std::string does not use UTF-16 (std::u16string does). So, if you have a UTF-8 encoded std::string, you already have the raw bytes for it, just copy them as-is. The code you have shown is doing exactly that, and is fine for UTF-8 strings. Otherwise, if you have/need std::string encoded in some other charset, you will need a charset conversion to/from UTF-8. There are 3rd party Unicode libraries that can handle that, such as libiconv, ICU, etc.

Conversion from UTF-8 encoded string to bytes and vice versa in C++

1 Answers1