IS DISTINCT FROM vs NOT EQUAL for non-null columns in Postgres?

I'm defining a trigger similar to the code posted in this answer.

CREATE TRIGGER trigger_update_account_details
AFTER UPDATE ON account_details
FOR EACH ROW
WHEN (OLD.email    IS DISTINCT FROM NEW.email
   OR OLD.username IS DISTINCT FROM NEW.username
   OR OLD.password IS DISTINCT FROM NEW.password) 
EXECUTE FUNCTION notify_insert_account_details();

However, let's say that email, username, and password are not NULLABLE columns.

It's my understanding that you use IS DISTINCT FROM when a column is nullable, but in my case, all of these values are NOT NULL.

In this case, is it still correct to use IS DISTINCT FROM, or should it simply be something like this instead?

WHEN (OLD.email    != NEW.email
   OR OLD.username != NEW.username
   OR OLD.password != NEW.password)

I'd imagine this latter would not only be more correct, but also faster, because no null check is performed. Is this correct?

Solution

You are correct that using IS DISTINCT FROM is not necessary for non-nullable columns, and using the != operator instead would be more appropriate and likely faster since no null check is needed.

Therefore, you can use the following code for your trigger:

CREATE TRIGGER trigger_update_account_details
AFTER UPDATE ON account_details
FOR EACH ROW
WHEN (OLD.email != NEW.email
OR OLD.username != NEW.username
OR OLD.password != NEW.password)
EXECUTE FUNCTION notify_insert_account_details();

This will ensure that the trigger only fires when the email, username, or password values are actually updated.