Martijn van Oosterhout <[EMAIL PROTECTED]> writes:
> That's very strange.
Not really. In the dummy example, the two indexes are exactly the same
size and so there is no reason for the planner to prefer one over the
other. (Note that the two plans have exactly the same estimated cost.)
In a real
Hiroshi Inoue <[EMAIL PROTECTED]> writes:
> What kind of expression is allowed as the predicate now ?
Anything that doesn't involve an aggregate or a subselect.
(But what the planner can recognize as matching a query's
WHERE clause is currently much more restricted, just ANDs
and ORs of simple V
Tom Lane wrote:
>
> Mike Mascari <[EMAIL PROTECTED]> writes:
> > To enforce uniqueness because
> > deactive is NULL, I cannot just create an index like:
>
> > CREATE UNIQUE INDEX i_foo (value, deactive);
>
> It's not clear to me what you are really after here. You *can* create a
> unique index
Mike Mascari <[EMAIL PROTECTED]> writes:
> To enforce uniqueness because
> deactive is NULL, I cannot just create an index like:
> CREATE UNIQUE INDEX i_foo (value, deactive);
It's not clear to me what you are really after here. You *can* create a
unique index, even though 'deactive' is allowed
I just wrote:
> What are the limits of the WHERE expression? Must they be composed of
> constant expressions or can they be more complex?
[snip]
Sorry. I see the SGML in your patch:
"Each element can only consist of ATTR OP CONST and these can only be
joined by
AND and OR operators."
Mike Mas
Martijn van Oosterhout wrote:
>
> On Wed, Jul 11, 2001 at 04:09:51AM -0400, Mike Mascari wrote:
> > Hello,
> >
> > I have created table/view pairs like:
>
> [snip]
>
> Yes, creating a unique partial index should be possible and will do what you
> want I think, (I couldn't totally follow what yo
On Wed, Jul 11, 2001 at 04:09:51AM -0400, Mike Mascari wrote:
> Hello,
>
> I have created table/view pairs like:
[snip]
Yes, creating a unique partial index should be possible and will do what you
want I think, (I couldn't totally follow what you meant).
However, partial indicies will not supp
Mike Mascari <[EMAIL PROTECTED]> writes:
>> Are Partial Indices still supported into latest release of PostgreSQL?
> I think they have suffered some serious bit-rot over the years
> and are no longer functional, although some of the core code
> still exists in the backend.
Actually, quite a LOT
