Rewrite conditional logic to use tabular data #8
Conversation
|
Hi Joe, I am really happy this repo captured your attention. As you probably noticed, you are the first contributor. So I am stoked this project is getting some traction. Moreover, your input is major. Making the app data-driven is one of the main goal of this project. Eventually, if the app could use HL7-compliant datasets pulled from the world's healthcare systems as input files, I think the app's output could be very interesting. The only comments I have is regarding aesthetics. I have published the app from your PR onto www.youlldie.com so you can see what I am talking about. Specifically, I find that the black bar with the title "YOU'LL DIE" at the top of the grey area could be removed. Also, I am wondering if the graph could be better positioned in relation to the block "Based on the information entered in the grey area, statistically, you have:". I mean, on desktop it looks OK but on mobile, the graph is not very telling. I am happy to approve this merge if you don't want to make additional updates. Thanks, Adam |
I got nerd-sniped by the discussion between you and @gadenbuie and couldn't resist seeing if the logic could be rewritten to be data-driven instead of entirely encoded in logic.
The first commit does this, using a transform.R script to grep and parse your R code to generate CSVs (:grimacing:). Then I left your original
cod_reactlogic intact, but also added a second data frame driven way of calculating the same data, and asserted that the two results were the same; then tried playing with all the options in the app. Indeed the results were the same.The second commit removes your original implementation, leaving just the new one.
The third commit tries to make the app easier to navigate using bslib's new sidebar/accordion. I also shortened some repeated code with loops.