I've seen contradictory answers on the internet with regard to whether C++/CLI is a superset of C++ or not.
The accepted answer on this question claims that "technically no", but doesn't provide an examples of non-C++/CLI code that conforms to ISO C++. Another answer on that question cites a book that says the opposite.
So, can you please provide accurate answers with example code that fails on C++/CLI or cite a trusted source (MSDN for example) on this matter?
I had someone this to开发者_高级运维pic come up today and thought I would like to inform myself, but I didn't find any clear answer elsewhere!
Valid C++, invalid C++/CLI:
int main()
{
int gcnew = 42;
}
gcnew
, generic
, and nullptr
are all reserved words in C++/CLI; nullptr
isn't really an issue in C++0x, of course. Herb Sutter blogged about C++/CLI keywords: Under the hood in 2003.
For what it's worth, when I was working on a large mixed codebase (both C++ and C++/CLI), I never ran into syntax issues or trivial issues like this. The giant issue with C++/CLI (in my opinion, of course) is that native code and managed code have completely different resource management paradigms, and it's really easy when writing mixed code to forget which paradigm you need for certain things. I fixed a lot of errors that were due either to an assumption that managed resources were destroyed deterministically or that native resources would be cleaned up by the garbage collector.
I am wondering about the plethora of the super-set/extension polemics. C++/CLI is a new language designed to act as a binding between ISO C++ and the ISO CLI. I heartily recommend to read the C++/CLI Design Rationale which clearly states the design goals at the very beginning of the document:
C++/CLI’s mission is to provide direct access for C++ programmers to use existing CLI libraries and create new ones, with little or no performance overhead, with the minimum amount of extra notation, and with full ISO C++ compatibility.
精彩评论