baseConvert() & 32-bit ops

[Dar Scott]

On Monday, June 23, 2003, at 12:19 PM, Mark Brownell wrote:

> the -1 is in regards to the bitAnd part of the f function's S-boxes.
>
>>>     f = S1[a] + S2[b]
>>>     f = bitXor( f , S3[c] )
>>>     f = bitAnd( f + S4[d], -1 )

Yeah, replace that last bitAnd with a mod 2^32.

> I'm tempted to get that highly recommended book that you referenced.

You will see that my final one line for above turns out to read just 
like the line in the book.

If you have a reference you prefer, I'm willing to let you buy me a 
copy of that!  But if you don't do that and you can afford it, get 
Schneier.  It will help in clarifying your thoughts and in 
communicating with people like me.  It covers Blowfish, block ciphers 
in general, and Freistel networks in general.  It even has C code for 
Blowfish.

Dar Scott