Re: PATCH: hex numbers and bit ops

[Mike Pall <mikelu-0509@...>]

Hi,

Philippe Lhoste wrote:
> Why not a variable number of arguments?
> y = math.bitor(0x12, 0x34, 0x56)
> looks nicer than
> y = math.bitor(0x12, math.bitor(0x34, 0x56))
> Isn't it?

Well, that would be easy. I'll change it if I have
to release another bugfix version.

> >  y... = math.bitfield(x, start [, len] ...)
> >    'start' is the bit position relative to the lsb (bit 0)
> >    It's a right/left shift for a positive/negative 'start'.
> >    It's a pure shift if 'len' is omitted.
> >    Otherwise the result is masked to 'len' valid bits.
> 
> Not very user friendly, it is quite hard to figure what the examples do...

Sure, docs could be improved. The following sentence may help:

  math.bitfield(x, start, len) extracts a bitfield from the
  number 'x' starting at bit number 'start' with length 'len'.

IMHO the examples show exactly what it's doing ...

> Why not make specific functions for shifting bits?

Umm? The two-argument bitfield(x, y) _is_ a plain shift.
It's (y >= 0) ? (x >> y) : (x << -y) in C notation.

Make an alias local bitshift = math.bitfield if you want. :-)

Bye,
     Mike