Re: [PATCH] Support for foreign keys with arrays

On 6 April 2012 07:21, Peter Eisentraut <peter_e@gmx.net> wrote:
> On lör, 2012-03-24 at 10:01 +0000, Gianni Ciolli wrote:
>> ON (DELETE | UPDATE) actions for EACH foreign keys
>> ==================================================
>>
>> ------------------ ----------- -----------
>>                   |    ON     |    ON     |
>> Action            |  DELETE   |  UPDATE   |
>> ------------------ ----------- -----------
>> CASCADE           |    Row    | Forbidden |
>> SET NULL          |    Row    |    Row    |
>> SET DEFAULT       |    Row    |    Row    |
>> EACH CASCADE      |  Element  |  Element  |
>> EACH SET NULL     |  Element  |  Element  |
>> EACH SET DEFAULT  | Forbidden | Forbidden |
>> NO ACTION         |     -     |     -     |
>> RESTRICT          |     -     |     -     |
>> ------------------ --------- -------------
>>
> I took another fresh look at this feature after not having looked for a
> month or two.  I think the functionality is probably OK, but I find the
> interfaces somewhat poorly named.  Consider, "PostgreSQL adds EACH
> foreign keys" -- huh?  I think they key word ELEMENT would be more
> descriptive and precise, and it also leaves the door open to other kind
> of non-atomic foreign key relationships outside of arrays.  EACH has no
> relationship with arrays.  It might as well refer to each row.
>
> On the matter of the above chart, there has been a long back and forth
> about whether the row or the element case should be the default.  Both
> cases are probably useful, but unfortunately you have now settled on
> making maximum destruction the default.  Additionally, we would now have
> the case that sometimes, depending on some configuration elsewhere, an
> ON DELETE CASCADE deletes more than what was actually involved in the
> foreign key.  What I'd suggest is to make both cases explicit.  That is,
> forbid ON DELETE CASCADE altogether and make people write ON DELETE
> CASCADE ROW or ON DELETE CASCADE ELEMENT.  In addition to making things
> more explicit and safer, it would again leave the door open to other
> kinds of relationships later.

+1

--
 Simon Riggs                   http://www.2ndQuadrant.com/
 PostgreSQL Development, 24x7 Support, Training & Services