On 29 Dec 2008, at 04:45, Chris Anderson wrote:
On Sun, Dec 28, 2008 at 7:41 PM, Antony Blakey <[email protected]
> wrote:
So given that descending=true swaps the start/end parameters, I
think the inclusiveness would have to swap as well.
I think you are correct about this, unless it'd be better to change
the interaction of descending=true with start/end key.
Swapping of startkey & endkey is a major WTF in the API that is
otherwise overly intuitive. It makes sense from the implementation
point of view, but for the user it does not. If I remember correctly,
it should be too hard to keep all the low-level code and make the
swap in the HTTP API, transparent for the user.
This is unrelated to the open vs. closed interval discussion. Any
interaction with that should be clearly thought out as well.
Cheers
Jan
--
