The best and most efficient way is to catch the "table not found" exception: this avoids the overhead of checking if the table exists twice; and doesn't suffer from the problem that if the DROP fails for some other reason (that might be important) the exception is still raised to the caller:
BEGIN
EXECUTE IMMEDIATE 'DROP TABLE ' || table_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -942 THEN
RAISE;
END IF;
END;
ADDENDUM
For reference, here are the equivalent blocks for other object types:
Sequence
BEGIN
EXECUTE IMMEDIATE 'DROP SEQUENCE ' || sequence_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -2289 THEN
RAISE;
END IF;
END;
View
BEGIN
EXECUTE IMMEDIATE 'DROP VIEW ' || view_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -942 THEN
RAISE;
END IF;
END;
Trigger
BEGIN
EXECUTE IMMEDIATE 'DROP TRIGGER ' || trigger_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -4080 THEN
RAISE;
END IF;
END;
Index
BEGIN
EXECUTE IMMEDIATE 'DROP INDEX ' || index_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -1418 THEN
RAISE;
END IF;
END;
Column
BEGIN
EXECUTE IMMEDIATE 'ALTER TABLE ' || table_name
|| ' DROP COLUMN ' || column_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -904 AND SQLCODE != -942 THEN
RAISE;
END IF;
END;
Database Link
BEGIN
EXECUTE IMMEDIATE 'DROP DATABASE LINK ' || dblink_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -2024 THEN
RAISE;
END IF;
END;
Materialized View
BEGIN
EXECUTE IMMEDIATE 'DROP MATERIALIZED VIEW ' || mview_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -12003 THEN
RAISE;
END IF;
END;
Type
BEGIN
EXECUTE IMMEDIATE 'DROP TYPE ' || type_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -4043 THEN
RAISE;
END IF;
END;
Constraint
BEGIN
EXECUTE IMMEDIATE 'ALTER TABLE ' || table_name
|| ' DROP CONSTRAINT ' || constraint_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -2443 AND SQLCODE != -942 THEN
RAISE;
END IF;
END;
Scheduler Job
BEGIN
DBMS_SCHEDULER.drop_job(job_name);
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -27475 THEN
RAISE;
END IF;
END;
User / Schema
BEGIN
EXECUTE IMMEDIATE 'DROP USER ' || user_name;
/* you may or may not want to add CASCADE */
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -1918 THEN
RAISE;
END IF;
END;
Package
BEGIN
EXECUTE IMMEDIATE 'DROP PACKAGE ' || package_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -4043 THEN
RAISE;
END IF;
END;
Procedure
BEGIN
EXECUTE IMMEDIATE 'DROP PROCEDURE ' || procedure_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -4043 THEN
RAISE;
END IF;
END;
Function
BEGIN
EXECUTE IMMEDIATE 'DROP FUNCTION ' || function_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -4043 THEN
RAISE;
END IF;
END;
Tablespace
BEGIN
EXECUTE IMMEDIATE 'DROP TABLESPACE' || tablespace_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -959 THEN
RAISE;
END IF;
END;
Synonym
BEGIN
EXECUTE IMMEDIATE 'DROP SYNONYM ' || synonym_name;
EXCEPTION
WHEN OTHERS THEN
IF SQLCODE != -1434 THEN
RAISE;
END IF;
END;
SELECT Orders.OrderNumber, LineItems.Quantity, LineItems.Description
FROM Orders
JOIN LineItems
ON LineItems.LineItemGUID =
(
SELECT TOP 1 LineItemGUID
FROM LineItems
WHERE OrderID = Orders.OrderID
)
In SQL Server 2005 and above, you could just replace INNER JOIN
with CROSS APPLY
:
SELECT Orders.OrderNumber, LineItems2.Quantity, LineItems2.Description
FROM Orders
CROSS APPLY
(
SELECT TOP 1 LineItems.Quantity, LineItems.Description
FROM LineItems
WHERE LineItems.OrderID = Orders.OrderID
) LineItems2
Please note that TOP 1
without ORDER BY
is not deterministic: this query you will get you one line item per order, but it is not defined which one will it be.
Multiple invocations of the query can give you different line items for the same order, even if the underlying did not change.
If you want deterministic order, you should add an ORDER BY
clause to the innermost query.
Example sqlfiddle
Best Answer
From what I've read, this error means that you're not referencing the table name correctly. One common reason is that the table is defined with a mixed-case spelling, and you're trying to query it with all lower-case.
In other words, the following fails:
Use double-quotes to delimit identifiers so you can use the specific mixed-case spelling as the table is defined.
Re your comment, you can add a schema to the "search_path" so that when you reference a table name without qualifying its schema, the query will match that table name by checked each schema in order. Just like
PATH
in the shell orinclude_path
in PHP, etc. You can check your current schema search path:You can change your schema search path:
See also http://www.postgresql.org/docs/8.3/static/ddl-schemas.html