Eric Niebler schrieb:
The change in reverse_view.hpp causes fusion::reverse_view to always be a bidirectional access sequence, even if the underlying sequence is a random access one. This is correct behaviour according to the documentation.
What is the motivation for this change? It seems to me that a reversed random-access range can and should be random-access as well.
On the one hand, that's correct. On the other hand, the documentation states that fusion::reverse_view models the bidirectional sequence concept. The random access intrinsic sequence backend implementation is also missing for fusion::reverse_view_tag. BTW, the upcoming C++0x-port of fusion has this issue fixed for all views.
The problem in proto is probably caused by a bug in the VC9/VC10 compiler. The non-random-access implementation of fusion::detail::fold wrongly causes the compiler to instantiate nested template parameters of the functor.
I attached a minimal sample which reproduces this problem.
You should file a bug with Microsoft and attach the preprocessed source code. I can do this if you don't know how.
I will file a bug myself. Thanks for the offer though :)
Interestingly, the random-access fold implementation works fine in your particular usecase, whereas my sample reproduces the problem for forward and random access sequences. I also attached a temporary workaround-patch for proto.
Proto's fold transform should work with any forward fusion sequence, so I'm afraid I cannot apply your patch because it may break third-party code.
Unfortunately I have no idea of how to "fix" fusion in the first place.
Is reverting the change to reverse_view and option?
No problem! I will have the change reverted and the documentation updated by tomorrow. -Christopher