On Dec 16, 10:31 am, Ivan Pavlov <ivan.pav...@xxxxxxxxx> wrote: > I can't answer your question but I think you may have a serious > database design issue at hand. > Why not try to accomplish your goals in a simpler way? > > Regards, > Ivan Pavlov > > On Dec 15, 12:49 pm, ket...@xxxxxxxxxx (Ketema Harris) wrote: > > > I am interested in finding out the pros, cons, pitfalls of using the > > following design: > > > Manual insert into Table A. > > Table A has a BEFORE INSERT trigger that causes an insert to table B. > > Table B has an AFTER INSERT trigger that causes an insert back to > > table A (With different criteria not an endless loop) > > > Table A will have its Before Trig fire again and this time the > > criteria causes it to finish with a return new. > > > Will the second insert into table A commit before the first insert > > into table A? What order does the insert into table B finish up? > > > Ketema J. Harriswww.ketema.net > > ket...@xxxxxxxxxx > > ketemaj on iChat > > > -- > > Sent via pgsql-general mailing list (pgsql-gene...@xxxxxxxxxxxxxx) > > To make changes to your subscription:http://www.postgresql.org/mailpref/pgsql-general I am all for simple, but some times there is not a simple answer. Complex business rules don't always have a simple solution. And complex design is not necessarily bad design. I am tasked with creating a transaction system that has a lot of things occur automatically after certain input. In analyzing the task I saw two main paths. Use the trigger and rule system that Pg provides or construct external methods to control the logic and issue "simple" commands to insert data when appropriate. I chose to use the trigger and rule system because the DB is built for transactional applications. Isn't that the point of ACID and atomicity and all those other buzzwords? I did not want to have to recreate what the database already can do. I just want to make sure that my understanding of what I think is going to happen is on target, and I'm looking for experience from others, as well as the tests I am performing. So far it is working as expected, I'd appreciate any feedback from anyone who has done something similar to avoid stepping in the same potholes others may have discovered. Thanks -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
