RE: Section 4.1 clarifications

["THALER,PAT (A-Roseville,ex1)" <pat_thaler@agilent.com>]

Michael,

I hadn't thought about that aspect, but it would be a computational pain.
One could assemble the binary number by doing decimal math on the decimal
encoding:

Repeat the following until the quotient is zero. 
   Divide the decimal string by 2^n
   Convert the remainder to an n-bit binary number and append to the high
order end of the binary result
   Replace the decimal string with the quotient

It is possible but not very pretty and I don't think making decimal encoding
available for long strings is worth the trouble.

Pat
  

-----Original Message-----
From: Michael Krueger [mailto:michael.krueger@windriver.com]
Sent: Thursday, April 25, 2002 2:42 PM
To: THALER,PAT (A-Roseville,ex1); Julian Satran
Actually, I'm concerned neither with storage space nor with speed, but
rather with code complexity.  Unlike hexadecimal and base-64 encoding, there
is not a simple correspondence between decimal digits in the string
representation and bit locations in the binary item.  Most implementation of
decimal-to-binary conversion functions like strtoul() do integer arithmetic
to obtain the binary representation, but this isn't feasible for binary
items larger than 4 (or, at an absolute maximum, 8) bytes.  Show me a simple
algorithm or a widely available library function for converting a
2,467-digit string to its 1024-byte binary representation, and I'll use it;
otherwise, support for decimal encoding of large binary items seems like a
needless complication that pays back very little added value.

Michael

--
Michael J. Krueger              mailto:michael.krueger@windriver.com
Wind River Networks                         http://www.windriver.com
500 Wind River Way                               phone: 510-749-2130
Alameda, CA  94501                                 fax: 510-749-2010