Unicode Characters Size

[Gobi Sakthivel]

std::string str ="����������";
std::cout<<str.length()<<std::endl;
 

this code returns 20, hence each unicode character is considered as 4 bytes, but the same code in java returns 10, where each unicode character is considered as 2 bytes.

My Question why java and c++ treats different size for the same kind of character, also if it is because of their own implementation, then give me a way to convert the c++ unicode strings to java readable unicode strings.

Eg:

std::string cpluscplustoJava(std::string str)
{
/** 
 * This function should return the same 
 * string with length as 10 byes..:P, It is possible??
 */
}
 

[Gobi Sakthivel]

Actually my string contains only 5 characters which is of square shape (unicode character), the bytes.com viewer shows like 10 characters as '?'. Don't get confused by it.

[weaknessforcats]

You can't use string for Unicode. Unicode characters are 16 bits (2 bytes).

Instead of char you use wchar_t and instead of string and cout you use wstring and wcout. the w is for "wide character".

Only if the 16 bit pattern in the wchar_t matches a valid Unicode character do you have a Unicode character. Be advised that mapping (called encoding) varies in Unicode so some characters are 8 while others are 16.

With Windows you convert using WideCharToMultiByte().

You usually need to call OS functions to create Unicode strings.

Thus is a bigger topic than can be addressed in a thread like this.

[Gobi Sakthivel]

@weeknessforcats Thanks for your reply, can you give me a link to understand the unicode much better in C++.

[weaknessforcats]

Try this: http://www.cprogramming.com/tutorial/unicode.html