Unexpected output of std::wcout << L"élève"; in Windows Shell

Question

While testing some functions to convert strings between wchar_t and utf8 I met the following weird result with Visual C++ express 2008

std::wcout << L"élève" << std::endl;

prints out "ÚlÞve:" which is obviously not what is expected.

This is obviously a bug. How can that be ? How am I suppose to deal with such "feature" ?

Dave Van den Eynde · Accepted Answer · 2009-04-06T14:12:20.627

12

The C++ compiler does not support Unicode in code files. You have to replace those characters with their escaped versions instead.

Try this:

std::wcout << L"\x00E9l\x00E8ve" << std::endl;

Also, your console must support Unicode as well.

UPDATE:

It's not going to produce the desired output in your console, because the console does not support Unicode.

edited Apr 06 '09 at 14:12

answered Apr 06 '09 at 13:09

Dave Van den Eynde

17,020
7
59
90

Unfortunately, using Dave's code yields exactly the same output. So I guess it means that the shell doesn't support unicode. – chmike Apr 06 '09 at 13:30
It seems I should be able to activate UTF-8 support in the shell by issuing the command chcp 65001. How can I do this from within a program before writing out things ? – chmike Apr 06 '09 at 13:42
It's not going to output the full UTF-16. You're lucky if you get ANSI output, because the high order bytes are knocked off. But the characters are ANSI page 1252 compatible. – Dave Van den Eynde Apr 06 '09 at 14:04

score 2 · Answer 2 · edited May 23 '17 at 12:19

2

I found these related questions with useful answers Is there a Windows command shell that will display Unicode characters? How can I embed unicode string constants in a source file?

edited May 23 '17 at 12:19

Community

1
1

answered Apr 06 '09 at 13:31

chmike

20,922
21
83
106

score 1 · Answer 3 · answered Aug 21 '10 at 02:25

This is obviously a bug. How can that be?

While other operating systems have dispensed with legacy character encodings and switched to UTF-8, Windows uses two legacy encodings: An "OEM" code page (used at the command prompt) and an "ANSI" code page (used by the GUI).

Your C++ source file is in ANSI code page 1252 (or possibly 1254, 1256, or 1258), but your console is interpreting it as OEM code page 850.

score 1 · Answer 4 · edited May 23 '17 at 12:12

1

You might also want to take a look at this question. It shows how you can actually hard-code unicode characters into files using some compilers (I'm not sure what the options would be got MSVC).

edited May 23 '17 at 12:12

Community

1
1

answered Apr 06 '09 at 13:12

jkp

78,960
28
103
104

score 0 · Answer 5 · answered Nov 11 '09 at 08:42

0

You IDE and the compiler use the ANSI code page. The console uses the OEM code page.

It also matter what are you doing with those conversion functions.

answered Nov 11 '09 at 08:42

Mihai Nita

5,547
27
27

Unexpected output of std::wcout << L"élève"; in Windows Shell

5 Answers5