I want to insert a new record if the record is not present in the table
For that I am using below query in Teradata
INSERT INTO sample(id, name) VALUES('12','rao')
WHERE NOT EXISTS (SELECT id FROM sample WHERE id = '12');
When I execute the above query I am getting below error.
WHERE NOT EXISTS
Failure 3706 Syntax error: expected something between ')' and the 'WHERE' keyword.
Can anyone help with the above issue. It will be very helpful.
You can use INSERT INTO ... SELECT ... as follows:
INSERT INTO sample(id,name)
select '12','rao'
WHERE NOT EXISTS (SELECT id FROM sample WHERE id = '12');
You can also create the primary/unique key on id column to avoid inserting duplicate data in id column.
I would advise writing the query as:
INSERT INTO sample (id, name)
SELECT id, name
FROM (SELECT 12 as id, 'rao' as name) x
WHERE NOT EXISTS (SELECT 1 FROM sample s WHERE s.id = x.id);
This means that you do not need to repeat the constant value -- such repetition can be a cause of errors in queries. Note that I removed the single quotes. id looks like a number so treat it as a number.
The uniqueness of ids is usually handled using a unique constraint or index:
alter table sample add constraint unq_sample_id unique (id);
This makes sure that the database ensures uniqueness. Your approach can fail if two inserts are run at the same time with the same id. An attempt to insert a duplicates returns an error (which the exists can then avoid).
In practice, id columns are usually generated automatically by the database. So the create table statement would look more like:
id integer generated by default as identity
And the insert would look like:
insert into sample (name)
values (name);
If id is the Primary Index of the table you can use MERGE:
merge into sample as tgt
using VALUES('12','rao') as src (id, name)
on src.id = tgt.id
when not matched
then insert (src.id,src.name)
