Обсуждение: = is not always defined as equality is bad

Vadim wrote:
> but this will be "known bug": this breaks OO-nature of Postgres,
because of
> operators can be overrided and '=' can mean  s o m e t h i n g (not
equality).
> Example: box data type. For boxes, = means equality of _areas_ and =~
> means that boxes are the same ==> =~ ANY should be used for IN.

Ok, here I think there should be a restriction to have the = operator
always be defined as equality operator. Because in the long run it will
be hard
to write equality restrictions.   a = a1 and b =~ b1 and c +*#~ c1.
Also =, >, <, >= and the like will allways be candidates for use by the
optimizer
(boolean math to simplify restriction or to make an existing index
usable could be used).

I vote for:  = must always be defined as equality in user defined types.

(if such comparison is not possible for a special type the = should not
be defined for it)
I therefore also suggest changing the box ops =~ to = and the area = to
some other sign.

Andreas

>
> Vadim wrote:
> > but this will be "known bug": this breaks OO-nature of Postgres,
> because of
> > operators can be overrided and '=' can mean  s o m e t h i n g (not
> equality).
> > Example: box data type. For boxes, = means equality of _areas_ and =~
> > means that boxes are the same ==> =~ ANY should be used for IN.
>
> Ok, here I think there should be a restriction to have the = operator
> always be defined as equality operator. Because in the long run it will
> be hard
> to write equality restrictions.   a = a1 and b =~ b1 and c +*#~ c1.
> Also =, >, <, >= and the like will allways be candidates for use by the
> optimizer
> (boolean math to simplify restriction or to make an existing index
> usable could be used).

I think each operator in pg_operator has a 'commutative' field for this:

| oprcom                           | oid                              |    4 |


--
Bruce Momjian
maillist@candle.pha.pa.us