[postgis-devel] Death to Pointless Operators

[Paragon Corporation]

> I'm not sure that "overloading =" is going to be possible. We already
> have a 'geometry = geometry' operator defined, in the btree opclass.
> When I try to define another 'geometry = geometry' under gist, it says
> there is already an '=' operator. And we need the '=' under btree if
> we are to get 'group by' working the way we want, right? Or will any
> '=' operator do for 'group by'?

> P.

Happy  Holidays all!

Paul,

I'm trying to figure out how the ltree folks get away with it.  I'm looking
at a 9.0 
Database I have that has ltree installed.  I can do group by and order by
an ltree object fine.

Hope this helps.

It looks like they just have one = operator defined like so:


CREATE OPERATOR =(
  PROCEDURE = ltree_eq,
  LEFTARG = ltree,
  RIGHTARG = ltree,
  COMMUTATOR = =,
  RESTRICT = eqsel,
  JOIN = eqjoinsel,
  MERGES);

But both there gist and btree classes contain that operator:

CREATE OPERATOR CLASS gist_ltree_ops DEFAULT
   FOR TYPE ltree USING gist AS
   OPERATOR 1  <,
   OPERATOR 2  <=,
   OPERATOR 3  =,
   OPERATOR 4  >=,
   OPERATOR 5  >,
   OPERATOR 10  @>,
   OPERATOR 11  <@,
   OPERATOR 12  ~(ltree, lquery),
   OPERATOR 13  ~(lquery, ltree),
   OPERATOR 14  @(ltree, ltxtquery),
   OPERATOR 15  @(ltxtquery, ltree),
   OPERATOR 16  ?(ltree, _lquery),
   OPERATOR 17  ?(_lquery, ltree),
   FUNCTION 1  ltree_consistent(internal, internal, smallint, oid,
internal),
   FUNCTION 2  ltree_union(internal, internal),
   FUNCTION 3  ltree_compress(internal),
   FUNCTION 4  ltree_decompress(internal),
   FUNCTION 5  ltree_penalty(internal, internal, internal),
   FUNCTION 6  ltree_picksplit(internal, internal),
   FUNCTION 7  ltree_same(internal, internal, internal)
   STORAGE ltree_gist;


CREATE OPERATOR CLASS contribs.ltree_ops DEFAULT
   FOR TYPE ltree USING btree AS
   OPERATOR 1  <,
   OPERATOR 2  <=,
   OPERATOR 3  =,
   OPERATOR 4  >=,
   OPERATOR 5  >,
   FUNCTION 1  ltree_cmp(ltree, ltree);

-- they also have this curious one which I assume is so they can do compares
between ltree query expressions and ltree objects
-- so this one is probably not relevant for our purposes

CREATE OPERATOR CLASS gist__ltree_ops DEFAULT
   FOR TYPE _ltree USING gist AS
   OPERATOR 10  <@(_ltree, ltree),
   OPERATOR 11  @>(ltree, _ltree),
   OPERATOR 12  ~(_ltree, lquery),
   OPERATOR 13  ~(lquery, _ltree),
   OPERATOR 14  @(_ltree, ltxtquery),
   OPERATOR 15  @(ltxtquery, _ltree),
   OPERATOR 16  ?(_ltree, _lquery),
   OPERATOR 17  ?(_lquery, _ltree),
   FUNCTION 1  _ltree_consistent(internal, internal, smallint, oid,
internal),
   FUNCTION 2  _ltree_union(internal, internal),
   FUNCTION 3  _ltree_compress(internal),
   FUNCTION 4  ltree_decompress(internal),
   FUNCTION 5  _ltree_penalty(internal, internal, internal),
   FUNCTION 6  _ltree_picksplit(internal, internal),
   FUNCTION 7  _ltree_same(internal, internal, internal)
   STORAGE ltree_gist;

Hope that helps,
Regina