Most of the errors and warnings seem to be simple fixes, but I thought
it best to ask here first whether or not anyone else has successfully
compiled the library in Visual Studio 2008, especially considering the
work-in-progress nature of the library.

FLAG_SIGNED =(1)<<32,

The warning is "shift count negative or too big, undefined behavior".
Doesn't this cause an overflow on 32-bit platforms, or am I missing
something? My experience is limited to Visual Studio, so I don't know
how other compilers might treat this differently.

The errors were due to the fact that TRANSACTION_COMMIT and
TRANSACTION_ROLLBACK (defined on line 738 of language_sql.h) are
macros defined in WinNT.h. To work around it, for now, I've renamed
TRANSACTION_BEGIN, TRANSACTION_COMMIT, and TRANSACTION_ROLLBACK to
SQL_TRANSACTION_BEGIN, SQL_TRANSACTION_COMMIT, and
SQL_TRANSACTION_ROLLBACK.

I can compile now, so I'm happy. I hope the above information will
give you some insight as to what the VS2008 users might encounter
using this framework.

And in case I haven't said it, I have been drooling all over the
Reason Library since reading the documentation and examples. Simple C+
+ libraries are difficult enough to find let alone an entire
framework. I predict this library will be the next the big thing for C+
+ developers, if not the Reason (pun intended) that C++ remains the
power tool of choice.

On Mar 25, 10:45 am, KRFournier <kfourn...@srpcs.com> wrote:

In general i dont worry too much about warning messages, there are
bigger fish to fry, but the one you pointed out in the formatting code
is indeed a bug, as the bit will overflow.

Youl have to forgive me since i dont regularly compile under VC8, but
im pretty sure i checked that before releasing the code.  Ive not
encountered the macro collision you seemed to have had, perhaps it has
something to do with the VC8 SDK or a windows header included
elsewhere in your project.  In my experience, windows header files are
disastrous, and i try not to include them anywhere if possible...

Its great that you find Reason to be simple and easy to use, it was my
hope that other developers would find it a refreshing change from the
usual unneeded complexity of modern software frameworks.  And if i can
manage to encourage more adoption then i certainly hope that it will
be the next big thing in C++.  But im reliant on users such as
yourself spreading the word ;)

Ill be releasing a new version within the next couple of months, since
that last release is a year old now you can expect some pretty big
changes including full signal/slot event/delegate support, unicode,
smart pointers, an enhanced iterator idiom, more collections, and
system utilities like atomic operations, process manipulation, shared
memory, memory mapped files etc.

Thanks,

Emerson

On 3/25/08, KRFournier <kfourn...@srpcs.com> wrote: