Re: [lmi] a bad MSVC problem with mc_enum

[Greg Chicares]

On 2008-06-23 05:26Z, Vadim Zeitlin wrote:
> 
>  I've found what seems an amazingly bad bug in MSVC compiler: it happily
> compiles the current code in mc_enum.[cht]pp but it doesn't seem to be able
> to handle references to arrays correctly.

Wow. Can they even compile this...

  // http://www.comeaucomputing.com/techtalk/#arraybounds

...tiny C++ example? If not, they'll choke on lmi_array_size()
in 'miscellany.hpp'.

>  The only workaround I found so far is [...snip...] But,
> again, I just don't seem to be able to find any other way to make the
> compiler generate the correct code after spending close to 2 hours on it.

Are you close to being able to predict whether this compiler is
going to work at all for lmi? I understand that you can't detect
linker problems until it compiles, or run-time problems until it
links. Does it all compile and link now, with your local patches,
so that we're in the final stage? For all binaries, especially
'lmi_wx_shared$(EXEEXT)', or only for the unit tests?

>  As usual, if anybody has any ideas about how to make this work otherwise,
> I'd love to hear about them. In the meanwhile, here is the patch I'm using
> right now, just for the reference:

My position would still be as stated here:
  http://lists.nongnu.org/archive/html/lmi/2008-02/msg00029.html