Claudio Valderrama C. wrote:

>Ann W. Harrison wrote:
>
>
>> INSERT INTO <whatever> SELECT <list> FROM <data source> YIELDING
>>...
>>
>>A question is whether the ability to return a transient data set from
>>an update or delete has value, since those operations are not based
>>on a full query expression.
>>
>>
>
>I think the important need is for INSERT, not for deletions and updates,
>unless somebody shows a compelling case.
>
>When RETURNING or YIELDING produce rows, are those rows the ones that are
>stored in the engine? I ask to be sure, since I may want to insert (a,b,c)
>in a row, but pre modify triggers change those values before being stored.
>
>
>

It's not difficult to construct a case for update and delete. Suppose
somebody wanted to do mass delete based on a range of time stamps and
wanted to know what records (primary key or any other information) got
deleted. Or if records were automatically timestamped by an update
trigger what the <primary key, timestamp> value pairs were for logging.
Symmetry is a good thing.