Errors when compiling with Multi-Byte Character Set about winreg HOT 5 CLOSED

I don't have interest in supporting MBCS in 2018 (and complicating my code for that purpose). If you want to support both Unicode and MBCS, you can modify my code and adopt the TCHAR model. That would mean using TCHAR instead of wchar_t, replacing std::wstring with std::basic_string<TCHAR> (or use the CString class, which is already TCHAR-enabled), etc.

from winreg.

I've forked repository and added support for MBCS. It doesn't complicate anything as I added two aliases at the beginning, replaced string and char classes with those and wrapped most string literals with _T(). It compiles and passes all tests with both Unicode and MBCS. I can create pull request if you are willing to merge it to this repository.

https://github.com/mersadk/WinReg

from winreg.

@mersadk: I'm glad you were able to modify my code and adapt it to your MBCS needs. I don't feel I want to merge this TCHAR-based code in this repository, as my preference for today's code is to use Unicode instead of MBCS. I tend to reserve std::string for pure ASCII or UTF-8 (note that ASCII is a proper subset of UTF-8, so this works fine). I feel that those MBCS code bases become "bug factories" when some portions of code assume a given std::string object contains text in code page X, but then you pass it to another section of code that assumes code page Y. It's a dangerous design, while trying to keep the whole code base Unicode, and immediately convert from a given code page to Unicode, feels much safer and higher-quality to me. These days I highly recommend moving legacy MBCS code bases to Unicode: I believe this is a very good and wise investment in code quality.

from winreg.

@GiovanniDicanio Thank you for info.

from winreg.

@mersadk I'm glad of being of some help. Thank you for your interest in this repository.

from winreg.