🚀 go-pugleaf

On Fri, 18 Jan 2002 09:53:30 -0500, "James White" <jamesw@avi.com>
wrote:

>1.) What does unicode compliance in a Delphi & C++ Builder application
>(backend, frontend) usually entail? Is it as simple as a compiler switch and
>replacing all chars with TCHAR's?

If you #define UNICODE, the Windows calls are translated to the
Unicode variants. Sadly, your app won't run on W9x then.

And you have to replace all your strings with Unicode variants -
AnsiString with WideString, string with wstring and char with wchar_t.

>2.) Is there any way to ensure that an application does not use any plain
>character functions when compiled for Unicode?

None that I know about. You have to check the source manually.

>3.) What does it mean when people say VCL does not support Unicode? Does

That means all strings you can put to the VCL componets are
AnsiStrings. Forget Unicode.

>that mean that data-entry in TEdit's etc can only support the basic English
>character set?

It supports the character set of the selected active language. It
supports even multibyte (Japanese...), when you have Japanese Windows
or W2000 with Japanese system locales.

>4.) What would you do?

I wrote my own Unicode aware VCL components, that use ExtTextOutW,
which works on W2000 and W9x. The text input remained restricted to
the Windows settings, but is converted to Unicode immediately after
the user typed it in.

Regards

		Petr Sulla