Sorry to reply to myself. By reading the whole document I see he does give a few different ways one can hadle the lack of padding (including adding it back in.) Both the specs. seem to fudge a bit on this...I'll have to think more about it.
On Apr 6, 5:40 pm, "[EMAIL PROTECTED]" <[EMAIL PROTECTED]> wrote: > Gabriel... > > I looked at Zooko's encoding. I didn't really like that the first 16 > characters weren't the same as for base 16, [0-9a-f]. > > I hadn't considered the need for padding. Zooko doesn't seem to use > it either. How does he get around it? > > Thanks > > p.s. Paul...yes it is different. :) > > On Apr 6, 7:15 am, "Gabriel Genellina" <[EMAIL PROTECTED]> wrote: > > > En Sun, 06 Apr 2008 06:07:18 -0300, Petite Abeille > > <[EMAIL PROTECTED]> escribió: > > > > On Apr 6, 2008, at 9:20 AM, [EMAIL PROTECTED] wrote: > > > >> Anyone know of aPythonimplementation of this: > > >>http://www.crockford.com/wrmg/base32.html > > > > Not sure aboutCrockford'sBase32encoding itself, but here is an > > > implementation of Bryce "Zooko" Wilcox-O'Hearn's "human-oriented > > > base-32 encoding": > > > >https://zooko.com/repos/z-base-32/base32/ > > >https://zooko.com/repos/z-base-32/base32/DESIGN > > > The design and rationale looks better. TheCrockfordversion is > > ill-defined in the sense that you can't recover the exact input string > > length in some cases; by example both "\x00"*4 and "\x00"*5 share the same > > encoding. > > base-64 encoding, by example, uses '=' as pad bytes at the end to avoid > > this problem. > > > -- > > Gabriel Genellina -- http://mail.python.org/mailman/listinfo/python-list
