On Mon, 17 Jan 2005 20:15:35 +0200, Antti Mäkelä wrote:
If there is some easy way to get a bit more "verbose" view on what the program was exactly trying to do with the Calendarinfo call, that might be something I could at least *try* to patch up myself. At least if it's something simple.
Your best bet is to put some ERR statements (so they are always printed) in the stub code itself. You're right, kernel32 functions are trickier to test, you need to install the mingw cross compiler (RPMs are available somewhere, I forget where) then cross-compile a test case that can be run on any random Windows box.
thanks -mike