s2t2
commented
9 years ago presence of these socrata datasets (see bottom of the page at https://data.medicare.gov/data/nursing-home-compare) gives us more design flexibility.
for example we might be able to get away with not having a database, passing requests directly to the SODA api endpoints:
- providers - https://data.medicare.gov/resource/hq9i-23gr.json
- quality measures - https://data.medicare.gov/resource/djen-97ju.json etc.
I'm thinking of writing a script to download and parse the CSV data from source (to replicate the manual steps I've taken to import the data into a database). It would be easy for me to do this in Ruby and optionally also to use Active Record to save the results in the database. Let me know if those technologies work, and if there is any preference between postgresql and mysql for the database.