Need Help in Understanding HOWTO Write UNICODE String to a File

[mike@spb]

How do I write to a file using UNICODE. I have searched the archives with no real answers. What do I need to do first, second and third.

I know how to use var.WriteString("whatever");

What are the parameters do I use to open a file to write?
Do I do it this way:

sConfFile.Open((CString)m_sConfigFile,CFile::modeCreate|CFile::modeNoTruncate|CFile::modeWrite);

or is there another way.

I am still learning this VC++ stuff so I am new to a lot of things.

Thanks in advance

Mike@spb

[BMW]

I don't know much either about Unicode, but it seems that you should use the _T prefix for your string ...

[BMW]

Give a look at :

http://www.codeproject.com/cpp/unicode.asp

[hankdane]

You need to decide how to store your Unicode string. You can store it in UTF-16, or double byte. In this case, you can either use wide versions of the file writer methods, or write your strings out as raw data. (Your data length would be [wcslen(str) * sizeof(WCHAR)]). You can also convert your strings to MBCS, in which case you can use the A2W macro from ATL, or manually use WideCharToMultibyte(). Personally, I prefer writing the data out in UTF-8.

Obviously, you need to read your data back the same way you wrote it out.

[mike@spb]

OK, now I'm confused. When I try to open the original file the first line looks like this in unicode:

 [][][][][]<[]p[]r[]o[]f[]i[]l[]e[][]v[]e[]r[]s[]i[]o[]n[]=[]"[]4[]5[]8[]7[]5[]2[]"[]

The "brackets" are actually squares when I read the file using wordpad or word.

When I do a test to try in duplicating this file it is written out as text. Here is what I have

#deine _UNICODE
CFile sTest;
CString sfile = "C:\\test.prx";

sTest.Open(sfile, CFile::modeCreate|CFile::modeWrite);
CString str = _T("\"is is some test\"");
sTest.Write(str, (str.GetLength()+1) * sizeof(TCHAR));
sTest.Close();

The result of the output to the file is "This is some test"[] instead of []"[]T[]h[]i[]s[] []i[]s[] []s[]o[]m[]e[] []t[]e[]s[]t[]"[] .

What am I doing wrong?

Mike@spb

[teejbee]

Hi Mike,
You are not doing anything wrong. What you see is correct and that IS unicode text. It is a 16-bit character. In usual case, the first 8 bit contains the ascii code for the character and the second 8 bits are 0. That's the reason why you see the [] symbol if your viewer program does not support unicode strings. My notepad is Win2K supports unicode strings and I can view the string properly.

Hope this makes you clear!

[mike@spb]

More questions. When I use the sample, as you describe as being correct. It seems to be in UNICODE; however, the application that reads this UNICODE file does not recognize it. If I save the file as a Unicode Text Document in Wordpad, the application then is able to see the unicode file. The applicaiton I am referring to is the Windows Media Encoder. The profiles it uses are in UNICODE text.

If I am writing in UTF-16, maybe I should be writing it out in UTF-8. What do I need to change to write to UTF-8 so I can see if the Windows Media Encoder recognizes the profile written in UTF-8?

Examples would help.

Mike@spb