[Test] More errors when building Win32 Unicode app.

There are some more errors from the revised Boost.Test code when building with /D"_UNICODE" /D"UNICODE": which is the default behaviour for new projects created with the VC8 IDE. The errors are: c:\data\boost\boost\trunk\boost\test\impl\debug.ipp(794) : error C2664: 'LONG (HKEY,LPCTSTR,PHKEY)' : cannot convert parameter 2 from 'const char [53]' to 'LPCTSTR' Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast c:\data\boost\boost\trunk\boost\test\impl\debug.ipp(807) : error C2664: 'LONG (HKEY,LPTSTR,LPDWORD,LPDWORD,LPBYTE,LPDWORD)' : cannot convert parameter 2 from 'const char [9]' to 'LPTSTR' Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast c:\data\boost\boost\trunk\boost\test\impl\debug.ipp(807) : error C2664: 'LONG (HKEY,LPTSTR,LPDWORD,LPDWORD,LPBYTE,LPDWORD)' : cannot convert parameter 1 from 'LPDWORD' to 'HKEY' Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast c:\data\boost\boost\trunk\boost\test\impl\debug.ipp(843) : error C2664: 'CreateProcessW' : cannot convert parameter 2 from 'char [200]' to 'LPWSTR' Types pointed to are unrelated; conversion requires reinterpret_cast, C-style cast or function-style cast Which can likely be easily solved by adding an "A" suffix to the Win32 API calls. Whether these are all the errors, or just the first ones to get reported is another matter... John.

"John Maddock" <john@johnmaddock.co.uk> wrote in message news:006001c8171b$cea69a80$2e121452@fuji...
There are some more errors from the revised Boost.Test code when building with /D"_UNICODE" /D"UNICODE": which is the default behaviour for new projects created with the VC8 IDE. The errors are:
[...]
Which can likely be easily solved by adding an "A" suffix to the Win32 API calls. Whether these are all the errors, or just the first ones to get reported is another matter...
I've made it to compile. I am not sure it's going to work properly at runtime though. In general I wouldn't recommend using Boost.Test in unicode environment at the moment. The are number of issues related to the Boost.Test unicode support, starting with the fact that everything is progrmmed using std::ostream. It may not be trivial to make it completely "unicode proof", but I'll try at some point. Not now though. Gennadiy
participants (2)
-
Gennadiy Rozental
-
John Maddock