Experience in compiling code adapted to Unicode and non-Unicode compiling Environments

The program we compile sometimes has two versions that support Unicode and do not support Unicode. In vs2005, you can set the project attributes. After changing the settings, we will find that the two versions are completely different when compiling the program. How can we avoid modifying the source code as much as possible? Therefore, we need to use the string functions that can be used in both environments during programming.

The following are my experiences:

1. For constant strings, use the text () Macro or _ T () macro for processing. For example:
   Regkey. querydwordvalue (text ("column count"), ntmp );
   Regkey. querydwordvalue (_ T ("column count"), ntmp );
2. For string copy programs, use _ tcscpy_s to replace strcpy. For example:
   _ Tcscpy_s (TND. sztip, STIP );