A combination of the Cheerio and Request libraries to make a nodejs scraper that doesn't have the same memory leak issues as node-scraper, which uses jsdom.
Scrape a number of pages asynchronously and write out your data in one file when done. It generally assumes that you are scraping a series of webpages whose urls vary by a different parameter.
json (default) or
csv if json is properly formatted, i.e. one dimensional.
See the example for an example setup. Better documentation to come.
I used to use node-scraper but its dependency on jsdom made it heavy and had memory-leak issues on large jobs. Cheerio is a much lighter alternative.