Jean-Marc,
thanks for your posting.
Our developers are certainly aware of the UNICODE and _T() macros, but the thing is that porting a piece of code from ANSI to UNICODE is by far not as simple as it seems. There are many internal problems, such as different buffer sizes that have to be passed between functions (usually measured by the sizeof operator), trickier file I/O if one needs to maintain compatibility with ANSI versions, and so on... I absolutely don't agree that a mere search/replace operation can reliably convert a project from ANSI to UNICODE.
On the other hand, we are planning a fully Unicode version - we realize that for almost half of the world (East Asia, India, Arab countries etc.) it is a must.
Thanks again
Vlk