range or xrange disallowed for big numbers

Chad Netzer cnetzer at mail.arc.nasa.gov
Tue Oct 8 21:52:54 CEST 2002


On Tuesday 08 October 2002 02:21, Joerg Woelke wrote:

> Both work here on a tru64 OS.
>
> >>> range(10000000000L, 10000000000L + 1L)
>
> [10000000000]

Granted, the specific examples will work on 64 bit machines, but the 
principle remains (try this:)

range( 10L**20, 10L**20L+1L)

Of course, one could argue that number ranges that require longs, aren't that 
common in practise (especially w/ 64 bits).

However, with auto-promotion of python ints to longs, I think it makes sense 
to support these ranges from a consistency standpoint (especially since it 
seems possible to do so with relatively minor work)

In fact, I'll code up an implementation to do it (and write the PEP, if 
people approve of it)

-- 

Chad Netzer
cnetzer at mail.arc.nasa.gov




More information about the Python-list mailing list