On Mon, Sep 19, 2005 at 04:33:27PM -0400, Tom Lane wrote:
> This is a bad idea, because it translates "x IS NULL" into "x = NULL"
> which is under no circumstances the same thing. It might coincidentally
> fail to malfunction for btree indexes, depending on the specifics of the
> "=" operator in u
Martijn van Oosterhout writes:
> Actually, it's two changes:
> - In the btree index code, if SK_ISNULL is set, do the right thing
> - If the query has col IS NULL, expand that to col =3D NULL in the index
> quals
This is a bad idea, because it translates "x IS NULL" into "x = NULL"
which is under
Here is a patch that enables IS NULL to use a btree index. After
mentioning it was really hard, I had a brainwave and here are the 80
lines of code necessary to make it work. What surprised me is that in
various important parts most of the work had actually been done already.
Actually, it's two ch
