Crashpad

Finding an apartment in SF is rapidly turning into a chore. Craigslist is spammy, and other not so spammy services primarily host corporate listings which are out of my price range. Being a good little nerd, the natural thing to do is to simply take the noisy data stream (craigslist) and filter it down to something more usable.

About

This is single entry point app which I run by hand at a REPL or via a cronjob.

Global state:

• Output .org file (emacs org-mode is awesome and easy to printf)
• "blacklist" .edn file containing listing URLs, titles and preview images for listings which have already been considered and added to the .org output.

Pipeline:

• Use arrdem/crajure to interface with craigslist.
• Pull down a whole bunch of listings.
• Use the blacklist to filter out listings by title, preview and URL which have been visited in a previous crawl.
• Update the blacklist file with the titles, previews and URLs of the surviving listings.
• Append a new header and elements to the org file.

License

Copyright Reid 'arrdem' McKenzie 2016, all rights reserved.