I want to execute a SQL script in a client using JDBC (not the postgreSQL pslq client). In this script I would like to do something like that:
skip errors on;
alter table foo ...;
skip errors off;
Is there a way to do this with PostgreSQL >= 9.1?
I found this thread with a good solution usind DO blocks and error codes:
How to continue sql script on error?
And creating a function for my problem:
create or replace function continueOnError(v_sqlStatement text)
returns void
language plpgsql
as '
BEGIN
execute v_sqlStatement;
EXCEPTION
WHEN invalid_schema_name
OR undefined_table
OR undefined_column
THEN RAISE WARNING ''Continued on error sql statement: %'', v_sqlStatement;
END;
';
...use it:
select continueOnError('alter table test add constraint fk_test_valueid foreign key (valueid) references value(id)');
Related
as we start to migrate our Application from using Oracle to PostgreSQL we ran into the following problem:
A lot of our Oracle scripts create triggers that work on Oracle specific tables which dont exist in PostgreSQL. When running these scripts on the PG database they will not throw an error.
Only when the trigger is triggered an error is thrown.
Example code:
-- Invalid query under PostgreSQL
select * from v$mystat;
-- Create a view with the invalid query does not work (as expected)
create or replace view Invalid_View as
select * from v$mystat;
-- Create a test table
create table aaa_test_table (test timestamp);
-- Create a trigger with the invalid query does(!) work (not as expected)
create or replace trigger Invalid_Trigger
before insert
on aaa_test_table
begin
select * from v$mystat;
end;
-- Insert fails if the trigger exists
insert into aaa_test_table (test) values(sysdate);
-- Select from the test table
select * from aaa_test_table
order by test desc;
Is there a way to change this behavior to throw an error on trigger creation instead?
Kind Regards,
Hammerfels
Edit:
I was made aware, that we actually dont use basic PostgreSQL but EDB instead.
That would probably explain why the syntax for create trigger seems wrong.
I'm sorry for the confusion.
It will trigger an error, unless you have configured Postgres to postpone validation when creating functions.
Try issuing this before creating the trigger:
set check_function_bodies = on;
Creating the trigger should show
ERROR: syntax error at or near "trigger"
LINE 1: create or replace trigger Invalid_Trigger
I am wondering if it's possible to use the result of a subquery as database name in a PostgreSQL (9.5.1) DDL statement.
For example, I wanted to alter the current database with something like:
ALTER DATABASE (SELECT current_database()) SET a_var TO 'a_value';
If I run this, an error occurs:
ERROR: syntax error at or near "("
LINE 1: ALTER DATABASE (SELECT current_database()) SET ...
What's the correct way to use the sub-query (if possible)?
You need dynamic SQL for that:
DO
$do$
BEGIN
EXECUTE format($f$ALTER DATABASE %I SET x.a_var TO 'a_value'$f$, current_database());
END
$do$;
Using format() to escape the db name safely while being at it.
BTW, to unset:
ALTER DATABASE your_db RESET x.a_var;
To see the current setting:
SELECT current_setting('x.a_var');
(The DB default is not active before you start a new session.)
Related:
Table name as a PostgreSQL function parameter
Error when setting n_distinct using a plpgsql variable
I´m creating a trigger on PGAdminIII where I want to delete the rows that have the foreign key on the other table. However I´m getting a Syntax error and I can´t find where the problem is:
CREATE TRIGGER clienteDelete
BEFORE DELETE ON cliente
FOR EACH ROW
BEGIN
DELETE FROM contacto WHERE contacto.id = OLD.contacto_idcontacto;
END;
ERROR: syntax error at or near "BEGIN"
LINE 4: BEGIN
^
********** Error **********
ERROR: syntax error at or near "BEGIN"
SQL state: 42601
Character: 68
I´m not used to the syntax of triggers on Postgres but that´s what I know according to the SQL standard. Any help will be highly apreciated
So Postgresql triggers has some limitations, for instance
PostgreSQL only allows the execution of a user-defined function for
the triggered action.
So to accomplish what you want you need to define a function and make the trigger fire that. Something like this should work:
CREATE FUNCTION clienteDelete() RETURNS TRIGGER AS $_$
BEGIN
DELETE FROM contacto WHERE contacto.id = OLD.contacto_idcontacto;
RETURN OLD;
END $_$ LANGUAGE 'plpgsql';
And the trigger:
CREATE TRIGGER delete_contacto
BEFORE DELETE ON cliente
FOR EACH ROW
EXECUTE PROCEDURE clienteDelete();
I'm no Postgresql expert though so expect the code above to not be perfect.
Using SQLFiddle, PostgreSQL 9.3.1.
I am learning to define triggers in PostgreSQL, and after doing some research I've found out the following:
Triggers in Postgres are different from MYSQL. Where in Postgres you must create a function that RETURNS TRIGGER, in MySQL you can just create a trigger. So this is what I've come up with:
On Employee Insert, we want to update Departments Total Salary.
CREATE FUNCTION update_sal() RETURNS TRIGGER AS $$
BEGIN
IF NEW.dno IS NOT NULL THEN
UPDATE Department SET Total_sal = total_sal + NEW.salary
WHERE department.dno = NEW.dno;
END IF;
RETURN NULL;
END;
$$ Language plpgsql;
CREATE TRIGGER updateInsert
AFTER INSERT ON Employee
FOR EACH ROW
EXECUTE PROCEDURE update_sal();
And I'm getting the following error:
Schema Creation Failed: ERROR: unterminated dollar-quoted string at or near "$$
BEGIN IF NEW.dno IS NOT NULL THEN UPDATE Department
SET Total_sal = total_sal +NEW.salary WHERE department.dno = NEW.dno":
I've solved the issue thanks to Database Function giving an error - Postgresql
It seems just changing the query terminator at the bottom of the Scheme Window solves this issue.
If you copy-pasted the code, then you've got a simple syntax error: ENDl should be END; in the last-but-one line of the function definition.
Otherwise, it looks good to me.
With SQL Server, I can execute code ad hoc T-SQL code with full procedural logic through SQL Server Management Studio, or any other client. I've begun working with PostgreSQL and have run into a bit of a difference in that PGSQL requires any logic to be embedded in a function.
Is there a way to execute PL/PGSQL code without creating an executing a function?
Postgres 9
DO $$
-- declare
BEGIN
/* pl/pgsql here */
END $$;
No, not yet. Version 9.0 (still alpha) will have this option (do), you a have to wait until it's released.
I struggled to get this working because it's fairly strict about adding semi colons in exactly the right places. But once you get used to that it works well. Besides the inability to return records of course, however you can raise notices & exceptions and do the other workarounds like using temp tables as #ErwinBrandstetter pointed out in a comment above.
e.g.:
DO
$$
BEGIN
IF EXISTS(SELECT 'any rows?'
FROM {your_table}
WHERE {your_column} = 'blah')
THEN
RAISE NOTICE 'record exists';
ELSE
RAISE EXCEPTION 'record does not exist';
END IF;
DROP TABLE IF EXISTS foo;
CREATE TEMP TABLE foo AS
SELECT 'bar'::character varying(5) as baz;
END
$$;
SELECT * FROM foo;