In the program that I have been converting from Borland C++ (which was written before Unicode appeared) to Visual C++ (about 15 years too late, I regret), a strange situation has developed:
It treats deprecations (e.g. using fopen() rather than fopen_s() ) as errors rather than warnings, and I had to wade through 26 of them to get a compilation.
I set its compilation mode character set in Debug mode and in Release mode to Multi-Byte Character Set to get Borland-type 8-bit character compilation and linking.
When in Debug mode it compiles correctly and the resulting *.exe file runs as it should, as far as I can see.
When in Release mode it compiles in as if Unicode mode, including throwing errors at illegal attempted conversions between 8-bit and 16-bit mode. The Property Pages shows Character Set :: Multi-Byte Character Set when in Debug mode and when in Release node.
The project's Release directory was automatically created and exists but has no files in it.
I switched the character compilation mode to Unicode mode and then back to Multi-Byte Character Set, and this fault remained.
It treats deprecations (e.g. using fopen() rather than fopen_s() ) as errors rather than warnings, and I had to wade through 26 of them to get a compilation.
I set its compilation mode character set in Debug mode and in Release mode to Multi-Byte Character Set to get Borland-type 8-bit character compilation and linking.
When in Debug mode it compiles correctly and the resulting *.exe file runs as it should, as far as I can see.
When in Release mode it compiles in as if Unicode mode, including throwing errors at illegal attempted conversions between 8-bit and 16-bit mode. The Property Pages shows Character Set :: Multi-Byte Character Set when in Debug mode and when in Release node.
The project's Release directory was automatically created and exists but has no files in it.
I switched the character compilation mode to Unicode mode and then back to Multi-Byte Character Set, and this fault remained.