Re: [ANN] Lua 5.2.0 (alpha) now available

[John Passaniti <john@...>]

On Wed, Dec 8, 2010 at 7:05 AM, Roberto Ierusalimschy
<roberto@inf.puc-rio.br> wrote:
> When we write about Lua (e.g., [1]) we try to give proper
> credit to all significant prior work, being it from the Lua
> community or not; we surely will give proper credit to
> Reuben if we write about the bit library. [...]

As far as I remember, the progression of what is now called bitlib
went something like this:

1.  I hacked together the original bit library [1] using a C
preprocessor abuse and a lack of concern for handling of the sign bit
or using the full range of the mantissa.  Hey, it worked for me, and
was my first attempt at extending Lua.  So there.

2.  Reuben took my code and improved it in important and useful ways [2], [3].

3.  Mike Pall improved matters further significantly with a functional
superset of Reuben's and other's bit libraries with better portability
and other features that made things even more sane [4].

There have been other bitwise operations libraries from various
others.  Some were written in pure Lua which means there were
somewhere between insane and pedagogical, useful only to those who
didn't mind madly burning processor cycles to twiddle bits.  Someone
else had a bitwise operations library that took a different direction
and implemented arbitrary length bit strings as userdata.  Others felt
didn't like the functional approach and patched Lua and offered such
as operators.

The current alpha code in 5.2 doesn't appear to be based on my,
Reuben's or Mike's code, so I don't know if it makes much sense to
credit any of us except in some abstract functional sense.


[1] http://lua-users.org/lists/lua-l/2000-07/msg00035.html
[2] http://lua-users.org/lists/lua-l/2000-10/msg00278.html
[3] http://lua-users.org/lists/lua-l/2001-11/msg00397.html
[4] http://lua-users.org/lists/lua-l/2008-12/msg00272.html