Hi,

Does anyone know of a simple way to do this? I want to create an
insert trigger for a table and if the record already exists based on
some criteria, I want to update the table with the values that are
passed in via the insert trigger without having to use all the 'set'
statements for each field (so if we add fields in the future I won't
have to update the trigger). In other words, I want the trigger code
to look something like this:

if exists (select * from TableA where Fld1 = inserted.Fld1) then
//don't do insert, do an update instead (would i want to rollback here?
and will I have access to the 'inserted' table still?)
Update TableA
Set TableA.<all the fields> = Inserted.<all the fields>
where Fld1 = inserted.Fld1
end if

Any help or ideas would be appreciated.
Thanks,
Teresa

UPDATE requires that you specify the columns by name. It's best
practice to do so in an INSERT statement too.

Always specify the column names. In the long run this will improve
reliability and save you development time.

--
David Portas
SQL Server MVP
--

takilroy (AT) yahoo (DOT) com wrote:

Nice hack.

A rollback is no good because you'd lose the update as well. But you could
delete the inserted row.

You may also have issues with primary keys and other constraints. If a
constraint fires before the trigger, your insert will fail on a pk
constraint and your clever trigger will never fire.

Finally, the performance issue is real. Doing the insert, deleting it, and
then updating causes only one real write to the table that has to be
committed, but carries three complete journal writes. An update only
carries two discreet write. It might be worthwhile to pump a few million
operations in each combination so you can at least speak knowledgeably
about what the real performance price is.

--
Kenneth Downs
Secure Data Software, Inc.
(Ken)nneth@(Sec)ure(Dat)a(.com)

On 12 May 2005 08:50:10 -0700, takilroy (AT) yahoo (DOT) com wrote:

Hi Teresa,

There is no way to avoid listing the columns in an UPDATE statement. If
there were, I'd recommend against it (just as I recommend against using
SELECT * or INSERT without column-list in production code).

Also, your trigger's pseudo-code will do an update for all rows that
were inserted if at least one of them exists in TableA. You could remove
the existance check; the effect will be the same (rows that are not in
TableA won't be changed, rows that are will be - and if no row in
inserted is also in TableA, nothing changes in TableA), but it will
somewhat improve performance.

Also, Kenneth is correct - constraints are checked before the trigger is
executed. The only way around that is to use an INSTEAD OF trigger that
updates rows that are already present and inserts rows that are not yet
present.

If you need help transforming this to an INSTEAD OF trigger, just
holler.

Best, Hugo
--

(Remove _NO_ and _SPAM_ to get my e-mail address)