It seems no secret that a child table will not fire a trigger defined on
it's parent table. Various posts comment on this. But nowhere could I find a
reason for this.
Now, I just wonder whether the people who request this are wrong in their
assumption that a trigger should fire on the child table, since those
requests date from 2004 and are still not implemented?
As far as I see propagation has numerous advantages and not-propagation
leads to maintenance problems resulting in data inconsistencies in case of
designs where triggers should propagate. On the other hand, do any design(s)
exist where there should be no propagation?
I think the same could be argued for FK propagation.
I read in the change logs of 8.4: "Force child tables to inherit CHECK
constraints from parents (Alex Hunsaker, Nikhil Sontakke, Tom)." So why not
with triggers and FK's?
Regards,
Davor