This happened a long time ago…

Boss’ Boss: Hey Simon. IT are testing a system. Can you give them a million call records from the finance copy? They want realistic test data.

Lowly Little Me: Er… OK. Don’t they have access already?

BB: Well, you know. It’s about being helpful.

LLM: Ah politics. Not interested. Ok anyway…. is this for that monthly reporting thing?

BB: Yes.

LLM: So, you need records distributed arbitrarily through time? Not just the last million calls?

BB: Yes. And include texts too. Everything really.

LLM: OK

Alt-Tab -> TOAD. New file.

select * from (
select caller_number, count(*)
from
	normalized_event ne
group by
	caller_numer
order
	count(*) desc
)
) where rownum < 1000

Hmm… these add up to about a million…. click… click… click…

Huh, how come there are so many calls for these top numbers? .. peek … ah… porn lines! Wow! I guess that’s realistic. Maybe I’ll throw in some from further down the list. click… click…

select * from
	normalized_event ne 
where 
	caller_number in ( .. pasted list .. )
 and rownum < 1000000

… wait…

Export to CSV…