I am programatically downloading aws bill from s3 bucket,parsing all line items and processes them. I am doing this process daily by remembering a marker based on timestamp of last lineitem in the bill. So, in the next run, I will process only the lineitems after my marker. But Aws is updating cost values by adding more line items back in time which I am ignoring in current approach. I can't process all lineitems as sometimes the files can be huge. Does aws mention this problem anywhere ? I couldn't find any documentation.
It would be helpful if anyone can give a better approach to resolve this.