Re: O(1) performance for insertions/deletions at the front of an Array/String

--- Nikolai Weibull
<[e-mail]> wrote:

> Eric Mahurin wrote:
> 
> > They could also be easily extended to Bignum when you need
> more from
> > the encoding - an arbitrary sequence of bytes could be put
> into a
> > Bignum.
> 
> Why would you want to do this?  What gives you the idea that
> a Bignum
> would be more efficient than a String anyway?

In general, you'd put a character in an Integer, not
necessarily a Bignum.  For most common cases the character
should fit in a Fixnum (ascii and unicode?).

> > The main exception to that I can think of is when you have
> an excess
> > of zero bytes.  "\000\000\000\000" and "\000" couldn't be
> > distinguished when you put them into an Integer.
> 
> So perhaps they shouldn't be?,

I'm not sure what character encodings are going to be used.  As
long as there are not both some encodings that start with zero
bytes and some that end with zero bytes, you should be OK.  You
can handle one or the other by making sure that the storage
order is such that the zero bytes are at the least significant
byte.  Even if you do have those cases, you could choose to do
those encodings differently to represent the same character.



		
____________________________________________________ 
Yahoo! Sports 
Rekindle the Rivalries. Sign up for Fantasy Football 
http://football.fantasysports.yahoo.com