At 12:08 PM 22/04/2003 -0400, Tom Lane wrote:
>regression=# select cast(8 as bit(32));
> bit
>----------------------------------
> 00000000000000000000000000001000
>(1 row)
That's also a little scary. "select cast(8 as bit(64))" is not what one
would expect either, except if you view it as a cast to bit(32) followed by
bit(64).
It's also sad that substring (which is I think the only get_bit-like
function) starts at the left. I would have expected the low order bits in
this model to be at the left, so Cast(X as bit(N)) would work consistently.
We still seem to be lacking > 64 bit numeric support (which is where I was
coming from with the varbit<->numeric suggestion).
Is there a view as to which bit should be considered low order?
Or is there a better data type for arbitrary length bit masks?
----------------------------------------------------------------
Philip Warner | __---_____
Albatross Consulting Pty. Ltd. |----/ - \
(A.B.N. 75 008 659 498) | /(@) ______---_
Tel: (+61) 0500 83 82 81 | _________ \
Fax: (+61) 03 5330 3172 | ___________ |
Http://www.rhyme.com.au | / \| | --________--
PGP key available upon request, | /
and from pgp5.ai.mit.edu:11371 |/