[AD] 4.0 release candidate 1

[Peter Wang <tjaden@xxxxxxxxxx>]

Ok, 24 hours after expected, the files are here:

http://alleg.sf.net/future/all400.zip                    (2621022 bytes)
http://alleg.sf.net/future/all400_bin.zip                (1332026 bytes)
http://alleg.sf.net/future/all400_diff.zip               ( 853058 bytes)
http://alleg.sf.net/future/allegro-4.0.0.tar.gz          (2260322 bytes)
http://alleg.sf.net/future/allegro-enduser-4.0.0.tar.gz  ( 703492 bytes)
http://alleg.sf.net/future/allegro-4.0.0-1.src.rpm       (2236518 bytes)

Don't go announcing these anywhere ;-)

Some of them contain demo.dat, which is why they're bigger than usual.
If nothing's wrong with them (wishful thinking), they will be uploaded
as 4.0.0 (the date would say 4/12 instead of 10/12, but *shrug*).

The Windows DLLs are now named all40.dll, alld40.dll, allp40.dll
(instead of all400.dll ...).  I think this is the right thing, but I'm
not sure.

The DLLs in all400_bin.zip were built with Mingw 1.1.  I tested
building with MSVC5, so if one more person says it doesn't work, I'm
gonna start saving up to buy a chainsaw.

The source RPM is a v4 RPM.  The spec file inside wouldn't work with
rpm 3.x anyway (I think), so it doesn't matter.  We just warn people
on the download page.

Eric, was there no better way to fix makedoc.c for BCC?  Now makedoc.c
depends on Allegro's headers.  I took out makedoc.c from the
allegro-tools RPM.

Don't touch the tree.  There are some changes which I've made but
haven't committed yet.  I'll send those later or tomorrow.