Scripts for scraping.

#Creating JSON feeds for iBurn You may ask yourself, "How did I get here?" ##Source Data (./data)

####"Unofficial"" data

• art.json Descriptive data for art. The result of scraper.py for art installations
• camps.json Descriptive data for camps. The result of scraper.py for camps
• art-locations Art geodata from BMorg
• camp-locations Camp geodata from BMorg

####Official data from the Playa Events API

• playaevents-art.json Descriptive data for Art installations
• playaevents-camps.json Descriptive data for camps
• playaevents-events.json Descriptive data for events

##The Python Modules

###merge_playaevents-camps.py## INPUT: camp-locations.json, playaevents-camps.json, camps.json (optional)

OUTPUT: ./results/camp_data_and_locations.json, ./results/unmatched_camps.json

This module merges camps from the scraper into the offical api feed, then adds location data (lat, lon, string of polar man coords) into this merged camp list from camp-locations.

Camps for which no location was found are ALSO output to ./results/unmatched_camps.json for manual inspection.

###merge_playaevents-art.py## INPUT: art-locations.json, playaevents-art.json, art (optional)

OUTPUT: ./results/art_data_and_locations.json, ./results/unmatched_art.json

This module merges art from the scraper into the offical api feed, then adds location data (lat, lon, string of polar man coords) into this merged event list from art-locations.

Art for which no location was found are ALSO output to ./results/unmatched_art.json for manual inspection.

###merge_playaevents-events.py## INPUT: camp-locations.json, playaevents-events.json,

OUTPUT: ./results/event_data_and_locations.json, ./results/unmatched_events.json

This module merges events from the scraper into the offical api feed, then adds location data (lat, lon, string of polar man coords) into this merged event list from camp-locations.

Events for which no location was found are ALSO output to ./results/unmatched_camps.json for manual inspection.

###merge_camp_id_from_events.py## INPUT: camp_data_and_locations.json, playaevents-events.json

OUTPUT: ./results/camp_data_and_locations_ids.json, ./results/unmatched_camps_id.json

This module merges camp ids nested in the Playa Events API Event feed (some Events have a hosted_by_camp parameter, which reveals name and id of the host camp) into a JSON listing of camps.

We used OpenJUMP to create a warp vector between the boundary points from 2013->2014, and then warped all the data for 2013->2014 from that same vector. Check out the .jmp file...

As of 2015 we host the data on S3 and use Cloudflare's CDN to cache the data to allow for updates without resubmitting the app. Cloudflare only caches these file types so make sure we use those extensions to keep the S3 bill down. We append .js for the .json files and .jar to the .mbtiles file.

Clients fetch from an 'secret' S3 ENDPOINT_URL/update.json.js to check the last time the data files were updated. We use the same datetime format as the PlayaEvents API.

{
	"art": {"file": "art.json.js",
			"updated": "2015-06-25 16:30:00"},
	"camps": {"file": "camps.json.js",
			  "updated": "2015-06-25 16:30:00"},
	"events": {"file": "events.json.js",
			   "updated": "2015-06-25 16:30:00"},
	"tiles": {"file": "iburn.mbtiles.jar",
			  "updated": "2015-06-25 16:30:00"}
}

If a file has been updated since the last time the client fetched it, simply fetch the new data for each datatype from ENDPOINT_URL/file and merge it with the existing data on the client.

Currently each file describes a complete snapshot of its collection, not a "delta" update. An item previously present but absent in an update should be treated as removed.

• Cleanup this README
• Better instructions on what scripts need to be run in what order