This project scraps live pollution readings for cities and stores them in mongodb.

Pollution readings are currently:

• @BeijingAir
• British Columbia government air quality readings

• A reading scraped from Twitter is saved as follows (structure is not final):

  {u'data': {u'concentration': 84.0,
             u'display': {u'en': u'Unhealthy (at 24-hour exposure at this level)', 
                          u'fr': u''},
             u'index': [166, u'AQI'],
             u'missing': False,
             u'pollutant': u'PM2.5'},
   u'raw': u'02-11-2014 08:00; PM2.5; 84.0; 166; Unhealthy (at 24-hour exposure at this level)',
   u'reading_id': 433031445540372480L,
   u'source': {u'display_name': u'BeijingAir',
               u'screen_name': u'BeijingAir',
               u'type': u'twitter',
               u'user_id': 15527964},
   u'type': {u'hour_from': datetime.datetime(2014, 2, 11, 8, 0),
             u'hour_to': None,
             u'nature': u'hourly'}}

• During deployment, you need to set several environment variables:
  • CONSUMER_KEY
  • CONSUMER_SECRET
  • OAUTH_TOKEN
  • OAUTH_TOKEN_SECRET
• This project is deployed using docker. See project airquality-docker for detals.

• Add support for postgreSQL.
• Change code in setup.py back to original checking for git projects.

• Gavin Bong