I'm trying to use a FOR loop:
create or replace function update_revisions() returns trigger as
$$
begin
declare col_name declare col_name information_schema.columns%ROWTYPE;
for col_name in
select column_name from information_schema.columns
where table_name='t'
loop
insert into debug_table values (col_name);
end loop;
end;
$$
language plpgsql;
But it always says:
syntax error at or near 'for'
Could someone please give me a hint what's wrong with it?
Immediate problem
Invalid syntax. Untangled:
create or replace function update_revisions()
returns trigger as
$$
declare
col_name information_schema.columns%ROWTYPE;
begin
for col_name in
select column_name from information_schema.columns
where table_name='t'
loop
insert into debug_table values (col_name);
end loop;
end;
$$ language plpgsql;
More problems
Table names are not unique in a Postgres database. More in this recent answer:
Behaviour of NOT LIKE with NULL values
The whole approach is inefficient. Use a single INSERT statement instead:
INSERT INTO debug_table (target_column) -- with column definition list!
SELECT column_name
FROM information_schema.columns
WHERE table_name = 't'
AND table_schema = 'public'; -- your schema
Related
I'm trying to create a dynamic audit trigger that reads a column_name from the information schema and inserts into an audit table which column has been changed. So far I have the following code:
CREATE OR REPLACE FUNCTION dynamic_column_audit()
RETURNS trigger
LANGUAGE 'plpgsql'
COST 100
VOLATILE NOT LEAKPROOF AS $BODY$
<<main_function>>
declare
table_name_tr text;
table_schema_tr text;
column_name_trigger text;
old_column_name_trigger record;
new_column_name_trigger record;
begin
-- variables
table_name_tr := TG_TABLE_NAME;
table_schema_tr := TG_TABLE_SCHEMA;
<<loop_sql>>
FOR column_name_trigger IN (SELECT c.column_name FROM information_schema.columns c WHERE c.table_schema = table_schema_tr and c.table_name = table_name_tr ORDER BY ordinal_position asc)
LOOP
RAISE INFO E'\n Column name: %, table_name: %', column_name_trigger , table_schema_tr||'.'||table_name_tr;
IF (TG_OP = 'UPDATE') THEN
INSERT INTO my_loggingtable (operation, table_schema, table_name, column_name, old_value, new_value)
VALUES (TG_OP, table_name_tr, table_schema_tr, column_name_trigger, OLD.||column_name_trigger, NEW.||column_name_trigger);
end if;
END LOOP loop_sql;
return new;
end main_function;
$BODY$;
I'm trying to see whether it's possible to do something like NEW. + column_name or OLD> + column_name from the loop rather than hardcode the column name value for each table.
The idea behind this, which it could not be possible is to do an audit trigger for around 20 tables without adding the name all the time.
You can use a construct like this:
DECLARE
newval text;
BEGIN
EXECUTE format(
'SELECT ($1::%I).%I',
table_name,
column_name
)
INTO newval
USING NEW;
END;
I am trying to create a dynamic function to use for setting up triggers.
CREATE OR REPLACE FUNCTION device_bid_modifiers_count_per()
RETURNS TRIGGER AS
$$
DECLARE
devices_count INTEGER;
table_name regclass := TG_ARGV[0];
column_name VARCHAR := TG_ARGV[1];
BEGIN
LOCK TABLE device_types IN EXCLUSIVE MODE;
EXECUTE format('LOCK TABLE %s IN EXCLUSIVE MODE', table_name);
SELECT INTO devices_count device_types_count();
IF TG_OP = 'DELETE' THEN
SELECT format(
'PERFORM validate_bid_modifiers_count(%s, %s, OLD.%s, %s)',
table_name,
column_name,
column_name,
devices_count
);
ELSE
SELECT format(
'PERFORM validate_bid_modifiers_count(%s, %s, NEW.%s, %s)',
table_name,
column_name,
column_name,
devices_count
);
END IF;
RETURN NEW;
END;
$$ LANGUAGE plpgsql;
My issue is with the execution of the dynamic function validate_bid_modifiers_count(). Currently it throws:
ERROR: query has no destination for result data
HINT: If you want to discard the results of a SELECT, use PERFORM instead.
CONTEXT: PL/pgSQL function device_bid_modifiers_count_per() line 21 at SQL statement
I can't really wrap my head around this. I understand that format() returns the correct string of function call with arguments. How do I fix this and make it work?
This should do it:
CREATE OR REPLACE FUNCTION device_bid_modifiers_count_per()
RETURNS TRIGGER AS
$func$
DECLARE
devices_count int := device_types_count();
table_name regclass := TG_ARGV[0];
column_name text := TG_ARGV[1];
BEGIN
LOCK TABLE device_types IN EXCLUSIVE MODE;
EXECUTE format('LOCK TABLE %s IN EXCLUSIVE MODE', table_name);
IF TG_OP = 'DELETE' THEN
PERFORM validate_bid_modifiers_count(table_name
, column_name
, (row_to_json(OLD) ->> column_name)::bigint
, devices_count);
ELSE
PERFORM validate_bid_modifiers_count(table_name
, column_name
, (row_to_json(NEW) ->> column_name)::bigint
, devices_count);
END IF;
RETURN NEW;
END
$func$ LANGUAGE plpgsql;
The immediate cause for the error message was the outer SELECT. Without target, you need to replace it with PERFORM in plpgsql. But the inner PERFORM in the query string passed to EXECUTE was wrong, too. PERFORM is a plpgsql command, not valid in an SQL string passed to EXECUTE, which expects SQL code. You have to use SELECT there. Finally OLD and NEW are not visible inside EXECUTE and would each raise an exception of their own the way you had it. All issues are fixed by dropping EXECUTE.
A simple and fast way to get the value of a dynamic column name from the row types OLD and NEW: cast to json, then you can parameterize the key name like demonstrated. Should be a bit simpler and faster than the alternative with dynamic SQL - which is possible as well, like:
...
EXECUTE format('SELECT validate_bid_modifiers_count(table_name
, column_name
, ($1.%I)::bigint
, devices_count)', column_name)
USING OLD;
...
Related:
Get values from varying columns in a generic trigger
Trigger with dynamic field name
Aside: Not sure why you need the heavy locks.
Aside 2: Consider writing a separate trigger function for each trigger instead. More noisy DDL, but simpler and faster to execute.
As I pointed out in the comment to Erwin Brandstetter's answer, initially I have an almost identical solution.
But issue was that I was getting the error
ERROR: record "new" has no field "column_name"
CONTEXT: SQL statement "SELECT validate_bid_modifiers_count(table_name, column_name, NEW.column_name, devices_count)"
PL/pgSQL function device_bid_modifiers_count_per() line 15 at PERFORM
This is why I thought I needed a way to dynamically evaluate things.
Currently got this working with the following still ugly looking to me solution (ugly because I don't like 2 IF statements, I would like it to be super dynamic, but maybe I am asking for too much):
CREATE OR REPLACE FUNCTION device_bid_modifiers_count_per()
RETURNS TRIGGER AS
$func$
DECLARE
row RECORD;
table_name regclass := TG_ARGV[0];
column_name text := TG_ARGV[1];
devices_count INTEGER;
BEGIN
LOCK TABLE device_types IN EXCLUSIVE MODE;
EXECUTE format('LOCK TABLE %s IN EXCLUSIVE MODE', table_name);
devices_count := device_types_count();
IF TG_OP = 'DELETE' THEN
row := OLD;
ELSE
row := NEW;
END IF;
IF column_name = 'campaign_id' THEN
PERFORM validate_bid_modifiers_count(table_name, column_name, row.campaign_id, devices_count);
ELSIF column_name = 'adgroup_id' THEN
PERFORM validate_bid_modifiers_count(table_name, column_name, row.adgroup_id, devices_count);
ELSE
RAISE EXCEPTION 'invalid_column_name %', column_name;
END IF;
RETURN NEW;
END;
$func$ LANGUAGE plpgsql;
I am open to more robust solution suggestions.
Basically, the second condition kind'a almost defeats the purpose of having a single function, I could have at this point as well split it into two functions. Because the goal is to define multiple (2) triggers using this function (providing arguments to it).
I am looking to automate each table update with an automatic update of the updated_at column. I am able to make this work for a specific table using a trigger. But my main goal, which I can't find anywhere, is to create a function that dynamically grabs all the tables in the schema, and creates that same trigger and only changing the table name that the trigger is referencing to. For the life of me I can't figure it out.
I believe this shouldn't be as tricky as I'm making it as ever table in our schema will have the exact same column name of 'updated_at'.
One solution that I tried and thought would work was turning the table schema into an array, and iterating through that to invoke/create the trigger each iteration. But I don't have a ton of psql experience so I am finding myself googling for hours to solve this one little thing.
SELECT ARRAY (
SELECT
table_name::text
FROM
information_schema.tables
WHERE table_schema = 'public') as tables;
I have also tried:
DO $$
DECLARE
t text;
BEGIN
FOR t IN
SELECT table_name FROM information_schema.columns
WHERE column_name = 'updated_at'
LOOP
EXECUTE format('CREATE TRIGGER update_updatedAt
BEFORE UPDATE ON %I
FOR EACH ROW EXECUTE PROCEDURE updated_at()',
t);
END loop;
END;
$$ language 'plpgsql';
Procedure:
CREATE OR REPLACE FUNCTION updated_at()
RETURNS TRIGGER AS $$
BEGIN
NEW.updated_at = now();
RETURN NEW;
END;
$$ language 'plpgsql';
Your DO block works. The only problem with it is we can't have same Trigger name for multiple triggers. So, you can either add a table_name suffix/prefix for the Trigger name.
DO $$
DECLARE
t text;
BEGIN
FOR t IN
SELECT table_name FROM information_schema.columns
WHERE column_name = 'updated_at'
LOOP
EXECUTE format('CREATE TRIGGER update_updatedAt_%I
BEFORE UPDATE ON %I
FOR EACH ROW EXECUTE PROCEDURE updated_at()',
t,t);
END loop;
END;
$$ language 'plpgsql';
Additionally you may add a check to see if the trigger already exists in information_schema.triggers to be safe.
IF NOT EXISTS ( SELECT 1 from information_schema.triggers
where trigger_name = 'update_updatedat_'|| t)
THEN
I'm trying to create a PostgreSQL stored procedure that must set default value 'now()' to each columns named 'CreationDate' in every tables.
CREATE OR REPLACE FUNCTION set_creation_date() RETURNS
void AS $$
DECLARE
t pg_tables%ROWTYPE;
BEGIN
FOR t IN SELECT "tablename" FROM pg_tables WHERE "schemaname" = 'public' LOOP
IF EXISTS (select * from information_schema.columns where table_name = t."tablename"
and column_name = 'CreationDate') THEN
EXECUTE FORMAT('ALTER TABLE %I ALTER COLUMN "CreationDate" SET DEFAULT now()', t."tablename");
END IF;
END LOOP;
END;
$$ LANGUAGE plpgsql;
But no columns are affected. What's wrong?
It looks like it's trying to alter t."tablename" every time. Try the following instead:
EXECUTE FORMAT('ALTER TABLE %I ALTER COLUMN "CreationDate" SET DEFAULT now()', t."tablename");
I don't have a PostgreSQL server available to test, so please let me know if the syntax isn't quite right.
I finally resolved. The error was in pg_tables%ROWTYPE: using debugger i realized that the field 'tablename' doesn't exist, so i declared 't' as text.
I created a function with parameters, where you have to pass schema name and column name. I ll post it here for anyone who ll need it
DROP FUNCTION IF EXISTS set_creation_date(character varying, character varying);
CREATE OR REPLACE FUNCTION set_creation_date(_schema_name character varying, _column_date character varying)
RETURNS void AS
$BODY$
DECLARE
t TEXT;
BEGIN
FOR t IN SELECT "tablename" FROM pg_tables WHERE "schemaname" = _schema_name LOOP
IF EXISTS (select * from information_schema.columns where table_name = t and column_name = _column_date) THEN
EXECUTE FORMAT('ALTER TABLE %I ALTER COLUMN %I SET DEFAULT now()', t, _column_date);
END IF;
END LOOP;
END;
$BODY$
LANGUAGE plpgsql;
Hi all,
I wrote a function in pl/pgsql and I have this problem:
I want use the values of NEW and OLD trigger record but I don't know the column name and number if the table.
for example:
CREATE OR REPLACE FUNCTION tt() RETURNS trigger AS $$
DECLARE
text1 text;
text2 text;
orig_name text =tg_argv[0];
orig_schema text =tg_argv[1];
log_name text =tg_argv[2];
log_schema text =tg_argv[3];
col pg_attribute.attname%TYPE;
[...]
BEGIN
orig_comp := quote_ident(orig_schema)||'.'||quote_ident(orig_name);
log_comp := quote_ident(log_schema)||'.'||quote_ident(log_name);
IF(trigger_mode='INSERT')
THEN
-- I want know the names of column
FOR colonna in
SELECT attname
FROM pg_attribute
WHERE attrelid = orig_comp::regclass
AND attnum > 0
AND NOT attisdropped
loop --for each column I want save the value like a string
text1=NEW||'.'||colonna; -- this don't work: error: record NEW don't have colonna like values
text2:=text2||' , '||text1;
end loop;
[...]
END IF;
[...]
RETURN NEW;
END;
$$ LANGUAGE plpgsql;
you know a original name - it is a variable TG_TABLE_NAME. And dynamic access to record fields is possible with EXECUTE USING statement.
CREATE OR REPLACE FUNCTION dynamic_trigger()
RETURNS TRIGGER
LANGUAGE plpgsql
AS $$
DECLARE
ri RECORD;
t TEXT;
BEGIN
RAISE NOTICE E'\n Operation: %\n Schema: %\n Table: %',
TG_OP,
TG_TABLE_SCHEMA,
TG_TABLE_NAME;
FOR ri IN
SELECT ordinal_position, column_name, data_type
FROM information_schema.columns
WHERE
table_schema = quote_ident(TG_TABLE_SCHEMA)
AND table_name = quote_ident(TG_TABLE_NAME)
ORDER BY ordinal_position
LOOP
EXECUTE 'SELECT ($1).' || ri.column_name || '::text' INTO t USING NEW;
RAISE NOTICE E'Column\n number: %\n name: %\n type: %\n value: %.',
ri.ordinal_position,
ri.column_name,
ri.data_type,
t;
END LOOP;
RETURN NEW;
END; $$;
This code is wrote by Tom Lane and it is from postgresql tricks pages Iteration over RECORD variable inside trigger.