Trying to output a lot of records:

Question

I've got trickey problem, I have a user that gets sent a number of files each week, these need to be compared.
As an example file1 and file2 are sent from an external source on disk. These files are in a text format and both contain email addresses. One of the files normally contains approx 600k email address records, where as the other normally contains approx 40k email address records.
There needs to be a comparison between these which I have done by creating 2 tables and using sqlldr to get the data in.
I then run a script like this:
select email_address
from file1
where email_address not in (select email_address from file2);
The problem in when I try to get these results out it takes an age... TOAD has never completed and using spool in sqlplus takes forever - infact I had a look at the expected runtime in Enterprise Manager and it said something like 336 hours...
Is there a better way to:
a) Compare the data in these files.
b) Get this data back out to a text file (csv file) for use by the user.
Is there any benefit to be had from using the 9i functionality of an external table?

Answer 1

Brilliant thank you.
select email_address from file1 f1 --643618-
where not exists (select email_address
     from file2 f2
where f2.email_address = f1.email_address);
Works like a rocket!