Does Python need a '>>>' operator?

[Martin v. Loewis]

"Ken Peek" <Ken.Peek at SpiritSongDesigns.comNOSPAM> writes:

> | That is only meaningful for positive numbers - it just doesn't work
> | for twos complement.
[...]
> Ahhhh!!!  Got'cha!!!!
> 
> Python DOESN'T HAVE UNSIGNED NUMBERS!!!

Why are you shouting at me? I didn't say "unsigned numbers", I said
"positive numbers". Python has infinitely many of those - assuming you
have enough memory to represent them :-)

> Since there is no way to know this happened to or object, we either
> have to make an overly complex "Rube Goldberg" type exploration, if
> we want to be able to do a logical right shift of an int-long
> object.

Just restrict yourself to positive numbers, and the logical shift is
the same as the arithmetic shift.

> THAT is the reason for the '>>>' operator-- the designers of Java
> got rid of unsigned integers, because many programming mistakes
> happened in 'C' with them in existence-- this (in turn) necessitated
> the creation of a different operator for 'arithmetic right shift',
> and 'logical right shift'.

The other reason, of course, is that they still wanted to represent 32
logical bits in a single int value. In Python, you can represent
arbitrary many bits in a positive number, and the right-shift will be
"logical".

> Quite frankly, I don't give a damn HOW the long (or int) is represented
> internally to the machine

Good, neither do I. I wonder why you bring up internal representation
then all the time.

> Is what I want so EVIL that Python can't do things that way?

No, it is impossible to do what you want - you have not presented a
single consistent specification, yet, which is also implementable.

Regards,
Martin