At the moment we're just assuming a scraper has an id column. We should instead check for an ID column and if there isn't one then we should fail fast and give a clear error message.
This might be affected by the solution to #12, in that we'll need to make sure that the scraper contains all the columns that we want to compare against.
At the moment we're just assuming a scraper has an
id
column. We should instead check for an ID column and if there isn't one then we should fail fast and give a clear error message.