Fredrik Lundh wrote: > Colin J. Williams wrote: > >> One of the little irritants of Python is that the range syntax is rather >> long-winded: >> [Dbg]>>> range(3, 20, 6) >> [3, 9, 15] >> [Dbg]>>> >> It would be nice if one could have something like 3:20:6. > > if you find yourself using range a lot, maybe you should check if you > couldn't use custom iterators more often. > > or use the R helper: > > >>> R[3:20:6] > [3, 9, 15] > >>> R[:20] > [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19] > >>> R[0:20:2] > [0, 2, 4, 6, 8, 10, 12, 14, 16, 18] > >>> R[1:20:2] > [1, 3, 5, 7, 9, 11, 13, 15, 17, 19] > > where R is defined as: > > >>> class R: > ... def __getitem__(self, slice): > ... return range(*slice.indices(slice.stop)) > ... > >>> R = R() > > </F> > Thanks, this appears to be a bit neater than the numpy version. On the other hand, the numpy version provides a wider functionality.
R[start, stop, increment] is certainly a little simpler than range(start, stop, increment). However, could we not achieve that with: >>> R= range >>> R(2, 20, 3) [2, 5, 8, 11, 14, 17] >>> Your point about iterators is well taken, but it seems that the range is used sufficiently frequently that some syntactic form would be helpful. Colin W. -- http://mail.python.org/mailman/listinfo/python-list
