Paul
I didn't say it was efficient :-) Though some could argue that there is
a lot you can do that don't require triggers for ie constraints.
I'd more likely accept the point that mssql does not have row level
triggers, this makes cascading key updates require a cursor in the
trigger yuk!. AFAIK IB doesn't have batch triggers so one could argue
that row level triggers are a lot less efficient in terms of CPU and
memory load than batch ones :-)
Neven
Paul Heinz wrote:
Neven wrote:
> - BEFORE triggers
In all triggers the state before and after is available hence the
concept of before and after triggers are irrelevant
True, but if you want to reject the operation (usually to enforce a business
rule), it's much more efficient (less CPU and memory load) to abort in the
before stage than in the after stage when the database engine has gone to
all the trouble to do the actual update/insert work when you're going to NAK
it anyway.
TTFN,
Paul.
_______________________________________________
Delphi mailing list
[email protected]
http://ns3.123.co.nz/mailman/listinfo/delphi
--
Neven MacEwan (B.E. E&E)
Ph. 09 620 1356 Mob. 027 4749 062
New Address Details
===================
MWK Computer Systems
1 Taumata Rd
Sandringham
Auckland
Ph 620 1356
Fx 620 1336
begin:vcard
fn:Neven MacEwan
n:MacEwan;Neven
email;internet:[EMAIL PROTECTED]
tel;work:649 620 1356
tel;fax:649 620 1336
tel;cell:0274 749 062
version:2.1
end:vcard
_______________________________________________
Delphi mailing list
[email protected]
http://ns3.123.co.nz/mailman/listinfo/delphi
