gistfile1.py

# A mill with less than 5% UMD loss from 2005-2012 is low risk, 5-28% is medium risk, and over 28% is high risk.
# One within concessions and one within entire radius
# Use the area of the entire radius or concession as the denominator (effectively assume 100% forested)

import json
import requests  # sudo pip install requests

APIURL = "http://wri-01.cartodb.com/api/v2/sql"

def parse_concession_json(j):
  '''format json response for use with other functions'''
  return

def get_concession_geoms(mill_id):
  '''Make cartodb request to get concession info for a given mill. Parse response
  into proper format for the rest of the data workflow.
  
  SELECT mill_entity_id, palm_geom, palm_id FROM mills_concessions_50 mills
  
  Sample URL:
    http://wri-01.cartodb.com/api/v2/sql?api_key=<APIKEY>&q=SELECT mill_entity_id, palm_geom, palm_id FROM mills_concessions_50 mills
  '''
  query = 'SELECT mill_entity_id, palm_geom, palm_id FROM mills_concessions_50 mills'
  r = requests.get(APIURL, params=dict(api_key='', q=query))
  concession_info = parse_concession_json(response.json())
  return concession_info

def gfw_api_request(lat, lon, radius=None, geom=None):
  '''Make GFW API request for a given latlon and radius to calculate the
  loss in a given area. Use the radius to generate a geometry suitable for GFW
  API, or just supply a geometry.'''
  return

def calc_loss(mill_id, lat, lon, radius):
  '''For a single mill, get geometries for nearby concessions, then use GFW API to
  calculate the UMD loss in each concession. Also gets the loss inside a given radius.
  
  We may want to add up the loss if there are multiple concessions.
  '''
  geoms = get_concession_geoms(mill_id)
  loss_radius = gfw_api_request(lat, lon, radius=radius)
  
  loss_dict = {}
  for geom, id in geoms.items():
    loss = gfw_api_request(lat, lon, geom)
    loss_dict[id] = loss
    
    # {10: 50, 11, 900} id: loss

  return loss_dict, loss_radius

def get_mills(table):
  '''Make request to CartoDB to get ids and latlons for all the mills
  we're working with.
  
  Something like this might work: SELECT id, lat, lon FROM mills_table'''
  
  return

def update_mill_table(mill, loss_dict, loss_radius):
  '''Update cartodb mills table with loss figures for concessions and the radius
  around the mill.'''
  return

def main():
  '''Gets a list of mills, loop through them to get loss statistics for their nearby
  concessions and the hardcoded radius (likely 50km).'''
  ids_latlons = get_mills(table)
  for mill in ids_latlons:
    loss_dict, loss_radius = calc_loss(mill['id'], mill['lat'], mill['lon'], radius)
    update_mill_table(mill, loss_dict, loss_radius)

  return loss_dict, loss_radius