Re: Writing to dependent tables in a function
От | Keith Worthington |
---|---|
Тема | Re: Writing to dependent tables in a function |
Дата | |
Msg-id | 20050112165834.M73361@narrowpathinc.com обсуждение исходный текст |
Ответ на | Re: Writing to dependent tables in a function (Tom Lane <tgl@sss.pgh.pa.us>) |
Список | pgsql-novice |
> "Keith Worthington" <keithw@narrowpathinc.com> writes: > > I have written a function that moves data from a load table > > (data_transfer.tbl_sales_order_line_item) to two data tables > > (sales_order.tbl_line_item & sales_order.tbl_item_description) > > the second of which is dependent on the first. When I run > > the function manually (create it as a function returning an > > integer and execute it after loading the data with the COPY > > command) it works fine. When I convert it to a TRIGGER and > > COPY data to the load table I get an error that says a > > foreign key constraint is being violated. > > Why are you iterating over the whole table in a FOR EACH ROW trigger? > At best that's exceedingly inefficient. If you don't want to change > the function then it should probably be an AFTER STATEMENT trigger. > > The example works with no error for me in 8.0, but in 7.4 I do get a > failure. I think the difference has to do with the delayed firing of > AFTER triggers in 7.4, but I'm not entirely sure why that's affecting > anything. > > regards, tom lane As I mentioned in my previous post the short answer is ignorance. However, by reading the documentation and testing I have learned that unfortunately AFTER STATEMENT is not available in v7.3.6 anyway. I am hoping to upgrade to v8 when it is released but we are currently facing a project deadline and I can't wait. Not to mention that v8 may 'break' some of the software that we have written. What is really confusing to me is why the trigger works when a single line is COPYed into the inital table and fails when multiple rows are COPYed. Any ideas on why this happens and suggested work arounds would really be appreciated. Kind Regards, Keith ______________________________________________ 99main Internet Services http://www.99main.com
В списке pgsql-novice по дате отправления: