Why oh why c++/cx and not c++11

Why is it that Microsoft always has to make all kinds of non-standard stuff before implementing the standard stuff, this just makes me so sad. Perhaps it's time to go back to roots again a do unix stuff or (I did not say this), FreeBSD / Apple

what do you mean?  Myself, I am warming to shared_ptr and unique_ptr. And auto helps a lot.  That is all standard, right?

 

c++11 standard overview

Sure they have added a few things, lambdas, auto, shared ptrs, rvalue references, but if the spent the time they used on their second hat (^) c++ extension, aka c++/cx and added stuff like variadic templates, data member initializers, etc... that would have been a lot better.

Also why not make a clean library c++ library for WinRT instead of some non-standard proprietary c++ extension.

1 hour ago, tdctaz wrote

c++11 standard overview

thanks for the link. enum class is a nice improvement on enum.

http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2347.pdf

Refactoring the C++ compiler must be a huge undertaking. And you have to wonder how much the windows platform folks dislike all the hat clr stuff that the .NET folks put in when they had sway. Is the hat syntax for CLR different than hat syntax for winRT? Is the hat part of the c++ standard?

 

Also why not make a clean library c++ library for WinRT instead of some non-standard proprietary c++ extension.

They did.  The Windows Runtime C++ Template Library (WRL) allows for creating/consuming WinRT apps/components with standard C++.

 

13 hours ago, tdctaz wrote

Sure they have added a few things, lambdas, auto, shared ptrs, rvalue references, but if the spent the time they used on their second hat (^) c++ extension, aka c++/cx and added stuff like variadic templates, data member initializers, etc... that would have been a lot better.

The people who did the C++/CX extensions and the core C++ language people aren't the same. Basically, it has been explained that the reason there are no new C++11 features in VC2012 is because they decided to focus on variadic templates, and this turned out to be much more complex than expected, which means they couldn't finish it or any other C++11 features in time for this release.

I do not follow why attributes support is a seemingly low priority.
http://blogs.msdn.com/b/vcblog/archive/2011/09/12/10209291.aspx
http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2008/n2761.pdf

http://social.msdn.microsoft.com/Forums/en-US/vcgeneral/thread/caeede2a-7a2c-410b-93b1-f5ec108765ad

Without attributes you can't have MEF and I guess ASP.NET MVC.

I don't have any ideas on it, but I wonder if COM would be easier if C++ had attributes. Why would you need to implement QueryInterface if a DLL exported attributes that declared what interfaces it supported and where they were located?

 

 

1 day ago, Charles wrote

The VC++ team has been very transparent here...

Watch Herb's keynote from GoingNative 2012, for example. Listen to what he has to say, too...

http://channel9.msdn.com/Events/GoingNative/GoingNative-2012/C-11-VC-11-and-Beyond

C

Thanks for that, seeing that helps a bit and knowing that its not going to be X-years till next VS release before we get the standard is nice. C++11 is needed and more c++ libraries. Not some new Microsoft c++ language extension.