AWS CUR format CSV and Parquet

AWS offers the CUR files to report on spending. The quality of the data is way better than the old Detailed Billing Files and it come in different formats.

I use the CUR files to generate really good looking Dashboards with AWS QuickSight

Today for the first time I checked CUR formatted as Parquet files, and there is a problem for anyone trying to reuse code, while the data seems to be very much the same as in the CSV, the headers are different. My first observations:

  • We have underscores added in the middle of words - bill/billingperiodstartdate (in CSV) becomes bill_ billing_period_start_date
  • We have underscores added instead of / bill/billingperiodstartdate (in CSV) becomes bill _ billing_period_start_date

Come on, I want the same headers to create one and only one query into Athena that works with CSV and Parquet

a bientôt

comments powered by Disqus