How to delete rows in a Teradata table that are not in another table?

sql teradata
Paul Hooper picture Paul Hooper · Dec 1, 2010 · Viewed 31.8k times · Source

What makes my situation tricky is that I don't have a single column key, with a simple list of primary keys to delete (for instance, "delete from table where key in ([list])"). I have multiple columns together as the primary key, and would need to join on all of them.

Using what I know of other databases, I thought this might be done as:

DELETE FROM
    table1 t1
  LEFT OUTER JOIN
      table2 t2
    ON
      t2.key1 = t1.key1 AND
      t2.key2 = t1.key2
  WHERE
    t2.key1 IS NULL;

But Teradata (v12) responds with error number 3706, saying "Syntax error: Joined Tables are not allowed in FROM clause."

Answer

Paul Hooper picture Paul Hooper · Dec 1, 2010

Found this is done by:

DELETE FROM
    table1
  WHERE
    (key1, key2) NOT IN (
      SELECT UNIQUE key1, key2 FROM table2
    );

Related questions

SQL SELECT multi-columns INTO multi-variable

I'm converting SQL from Teradata to SQL Server in Teradata, they have the format SELECT col1, col2 FROM table1 INTO @variable1, @variable2 In SQL Server, I found SET @variable1 = ( SELECT col1 FROM table1 ); That only allows a single column/variable …

Read More
What is ROWS UNBOUNDED PRECEDING used for in Teradata?

I am just starting on Teradata and I have come across an Ordered Analytical Function called "Rows unbounded preceding" in Teradata. I tried several sites to learn about the function but all of them uses a complicated example explaining the …

Read More
Find which rows have different values for a given column in Teradata SQL

I am trying to compare two addresses from the same ID to see whether they match. For example: Id Adress Code Address 1 1 123 Main 1 2 123 Main 2 1 456 Wall 2 2 456 Wall 3 1 789 Right 3 2 100 Left I'm just trying to figure out whether the address for each ID …

Read More