App loads slowly about ratecomparison HOT 1 OPEN

Good points Patrick! I have a few ideas for likely speedups, in order of ease of implementation:

1. Saving the dataset as a .rda file instead of a CSV should speed up the initial load time.

2. Drew's recommendation to not calculate hypothetical revenue for historical data (only for forecasts) would speed up normal operation because it would put less load on RateParser.

3. There is probably some room for parallelization in RateParser. E.g. could chunk up the dataframe and dispatch to different cores.

4. Right now RateParser is called more times than it needs to be. This is a result of Shiny's reactive nature and the feedback between the OWRS file and the inputs. If we could only call RateParser once instead of ~3 times when the tab is first loaded we would see a bit speed boost. Tracking things through the chain of reactivity can be a nightmare though.

from ratecomparison.