I have a table let's call it movie that is partitioned by genre.
I have a trigger defined on one of the partitions as follows:
create trigger refresh after insert or update or delete
on movie_comedy
for each statement execute procedure refresh();
This trigger is used to update a materialized view.
If I do an update on movie_comedy the trigger is called but if I update the movie table the trigger doesn't get called. The updated row is on movie_comedy as well.
Is it possible to make this happen?
You'd have to create a trigger on the partitioned table itself for that.
Related
When the following transaction is run concurrently on different connections it sometimes errors with
trigger "my_trigger" for relation "my_table" already exists
What am I doing wrong?
BEGIN;
DROP TRIGGER IF EXISTS my_trigger ON my_table;
CREATE TRIGGER my_trigger
AFTER INSERT ON my_table
REFERENCING NEW TABLE AS new_table
FOR EACH STATEMENT EXECUTE PROCEDURE my_function();
COMMIT;
I am trying to set up a system where I can add triggers to notify about data changes in specific tables. If a table already has such a trigger then skip it. Otherwise CREATE all CRUD triggers. This logic needs to run sequentially in case of concurrent requests.
After trying ISOLATION LEVEL SERIALIZABLE I noticed that any conflicting transactions are failed and dropped (I would need to manually check sql status and retry). But what I want is to queue up these transactions and run afterwards one by one in the order they're sent.
At the moment I am trying to achieve this by having a my_triggers (table_name TEXT) table that has a BEFORE INSERT OR DELETE trigger. Within this trigger I do the actual table trigger upsert logic. Inserts or deletes on my_triggers are made with LOCK TABLE my_triggers IN ACCESS EXCLUSIVE MODE ... which should queue up conflicting CRUD transactions ?!
What happens is following:
BEGIN....DROP TRIGGER IF EXISTS....CREATE TRIGGER....COMMIT;
..BEGIN....DROP TRIGGER IF EXISTS....CREATE TRIGGER--------EXCEPTION.
Both transactions starts when trigger is not present.
Both succeed in drop trigger because of "IF EXISTS" statement.
First transaction starts creating a trigger. For that a SHARE ROW EXCLUSIVE lock is placed on table my_table. The lock SHARE ROW EXCLUSIVE conflicts with it self so no other transaction is allowed to create a trigger until the first one completes.
Second transaction blocks on CREATE TRIGGER.
First transaction completes.
Second transaction proceeds with CREATE TRIGGER but it already exists. Exception is raised.
What you need is adding a LOCK before DROP TRIGGER statement. This way you will ensure the trigger is dropped and not created in concurrent transaction.
BEGIN;
LOCK TABLE my_table IN SHARE ROW EXCLUSIVE MODE ;
DROP TRIGGER IF EXISTS my_trigger ON my_table;
CREATE TRIGGER my_trigger
AFTER INSERT ON my_table
REFERENCING NEW TABLE AS new_table
FOR EACH STATEMENT EXECUTE PROCEDURE my_function();
COMMIT;
I am trying to trigger a refresh for a materialised view after an Insert,update or delete on another materialised view. I am trying to use a triggered function to achieve this. Script is as follows;
CREATE FUNCTION aza_ods_version1.populate_fact_churn_m()
RETURNS TRIGGER AS $$
BEGIN
REFRESH MATERIALIZED VIEW aza_ods_version1.fact_churn_monthly;
RETURN NULL;
END;
$$ LANGUAGE plpgsql;
CREATE TRIGGER refresh_fact_churn_monthly
AFTER INSERT OR UPDATE OR DELETE OR TRUNCATE
ON aza_ods_version1.fact_transactions
FOR EACH ROW
WHEN (OLD.* IS DISTINCT FROM NEW.*)
EXECUTE PROCEDURE aza_ods_version1.populate_fact_churn_m();
On attempting to create the trigger, I get this error;
ERROR: "fact_transactions" is not a table or view
SQL state: 42809
I think its because fact_transactions is a materialized view and not a view or table.
Is there a way i can achieve what i am trying to??
You can't.
The documentation for CREATE TRIGGER doesn't mention materialized views:
table_name The name (optionally schema-qualified) of the table, view, or foreign table the trigger is for.
You cannot put a trigger on a materialized view, you will need to put the trigger on each table references in the query that generated the materialized. Further for TRUNCATE you will need to create a separate trigger:
In addition, triggers may be defined to fire for TRUNCATE, though only
FOR EACH STATEMENT.
So I am new to using procedures and triggers and it is really confusing me
I have used temporal tables and want to basically create a history table of records inserted,updated or deleted.
Infact I have created my history table and works fine when I use this trigger sql
DROP TRIGGER if exists versioning_trigger on mytable;
CREATE TRIGGER versioning_trigger BEFORE INSERT OR UPDATE OR DELETE ON mytable FOR EACH ROW EXECUTE PROCEDURE versioning('sys_period', 'table_history', true);
This creates records of the rows updated or deleted,precisely copies the old row record from mytable into table_history table and updates the record in mytable.But I want to insert the updated record from mytable to table_history also so that it has records of all types('current active record'and 'record before updation').Also insert some other fields in table_history when the trigger is executed.
I want to ask
How is it possible to have different trigger events(BEFORE or AFTER) together in one CREATE TRIGGER query in temporal_tables?
Is it possible to insert new field values in table_history on trigger execution? How can I accomplish this?
https://www.postgresql.org/docs/current/static/plpgsql-trigger.html
A trigger procedure is created with the CREATE FUNCTION command,
declaring it as a function with no arguments and a return type of
trigger
and also
same trigger can't fire both before and after event - just create two triggers if you really need it
https://www.postgresql.org/docs/current/static/sql-createtrigger.html
Determines whether the function is called before, after, or instead of
the event.
use NEW instead of OLD for new values
https://www.postgresql.org/docs/current/static/plpgsql-trigger.html
NEW
Data type RECORD; variable holding the new database row for
INSERT/UPDATE operations in row-level triggers. This variable is
unassigned in statement-level triggers and for DELETE operations.
Situation
I have a database in PostgreSQL 9.5 used to store object locations by time.
I have a main table named "position" with the columns (only relevant):
position_id
position_timestamp
object_id
It is partitioned into 100 child tables on object_id with the condition:
CREATE TABLE position_object_id_00
( CHECK object_id%100 = 0 )
INHERITS ( position );
And so on for the others children. I partitioned with a modulus relation to distribute equally the objects.
Each child is indexed on position_id and object_id (two different indexes).
The trigger to redirect inserts on children is:
CREATE TRIGGER insert_position_trigger
BEFORE INSERT ON position
FOR EACH ROW EXECUTE PROCEDURE insert_position();
And the procedure insert_position() looks for the right child table to insert the data, inserts it and then return the NEW object:
CREATE OR REPLACE FUNCTION insert_position() RETURNS TRIGGER AS $insert_position$
DECLARE
BEGIN
--Look for child table
[...]
--Insert data in right child table
[...]
RETURN NEW;
END;
$insert_position$ LANGUAGE plpgsql;
I have a summary table object_last_known_position with the same columns that is updated with the trigger:
CREATE TRIGGER update_object_last_known_position
AFTER INSERT OR UPDATE ON position
FOR EACH ROW
EXECUTE PROCEDURE update_object_last_known_position();
The procedure update_object_last_known_position() basically checks if the position_timestamp is more recent, then delete the older entry and create a new entry with the data passed on the INSERT or UPDATE query (NEW).
Issue
So these two triggers react to the same event: insert on position, one is before, the other is after
Returning new for insert_position() allows me to use NEW in the trigger update_object_last_known_position(), and this is absolutely necessary. But doing that, it also insert the data on the master table position. So my data is now duplicated.
I tried to put the two triggers before, they both execute when I insert data if I let it like that, but if I remove the "return new" from the procedure insert_position(), update_object_last_known_position() is not executed.
I am stuck with this issue and I didn't find a way to execute both of these triggers without filling the master table position when I insert data.
So if you have any ideas, I will really appreciate :)
Thank you for your help!
EDIT
Solution
Thanks to the answer
I "merged" my two triggers: insert_position() now calls update_object_last_known_position directly. For that, I modified update_object_last_known_position to a stored procedure with a parameter. The parameter is the id of the position insert_position() just created, so I am able to find it and retrieve information.
(Calling update_object_last_known_position inside the other trigger means we cannot use NEW anymore)
And obviously return type for insert_position() is now NULL, and everything works fine :)
If I understand you correctly you are trying to:
Stop the insert, and replace it with an insert into another table (determined by the trigger)
Update a summary table (delete/insert) to point to the new row.
Your problem is that 1 stops 2 from happening? That's logical because you've stopped the insert so you've stopped any processing on the insert as well.
So to solve this you have a couple of options (options 1 and 2 are similar)
Call update_object_last_known_position() from insert_position() and only have one trigger
Create a wrapper method for both insert_position() and update_object_last_known_position() and have only one trigger.
Put the trigger for update_object_last_known_position() on all of the tables that insert_position() might insert into.
I want to call a stored procedure in three different cases: after an insert, after an update and before a delete. Is it possible to combine these cases into one create trigger statement? Like so (I know, this won't work):
CREATE TRIGGER update_pntzzz3
AFTER INSERT OR UPDATE OR BEFORE DELETE ON onedee
FOR EACH ROW
EXECUTE PROCEDURE update_pntzzz3();
Right now I create a new trigger for the delete case.
CREATE TRIGGER update_pntzzz3
AFTER INSERT OR UPDATE ON onedee
FOR EACH ROW
EXECUTE PROCEDURE update_pntzzz3();
CREATE TRIGGER del_pntzzz3
BEFORE DELETE ON onedee
FOR EACH ROW
EXECUTE PROCEDURE update_pntzzz3();