I want to get the codepoint for each character in this string "عربى" so I write this code but it always output 63 which is the codepoint of the question mark character "?"
TCHAR myString[50] = _T("عربى");
int stringLength=_tcslen(_T(myString));
for(int i=0;i<stringLength;i++)
{
unsigned int number =myString[i];
cout<<number<<endl;
}
any suggestions ? :)
Here's code that uses only the standard library and iterates the string by 32-bit wide code units. In the latest UTF-32 this matches up with code points.
using namespace std;
const auto str = u8"عربى";
wstring_convert<codecvt_utf8<char32_t>, char32_t> cv;
auto str32 = cv.from_bytes(str);
for(auto c : str32)
cout << uint_least32_t(c) << '\n';
If your standard library hasn't implemented these features yet, you should probably use an external library.
I copied your code, and by removing the _T(myString) cast into simply myString, it worked. Here is the full program.
#include <afxwin.h>
#include <iostream>
int main() {
using namespace std;
TCHAR myString[50] = _T("عربى");
int stringLength = _tcslen(myString); // <----- edit here
for(int i=0;i<stringLength;i++)
{
unsigned int number =myString[i];
cout<<number<<endl;
}
}
Output:
1593
1585
1576
1609
TCHAR is 16 bits wide (if unicode is enabled). A 32 bit wide code point will be represented by 2 code units in UTF-16. This code still iterates the string by (16 bits wide) code unit rather than code point. For example this character: 𐌂 (isthisthingon.org/unicode/…), the code would print 55296 57090 instead of 66306. (at least I think so, the code doesn't compile in my compiler). I've added an answer that works with all current unicode but requires c++11.