Created
April 30, 2012 22:48
-
-
Save boundsj/2563386 to your computer and use it in GitHub Desktop.
Find and count all the things nearby a place
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
import math | |
import pymongo | |
from pymongo import Connection | |
earth_radius_in_miles = 3959.0 | |
radians_to_degrees = 180.0 / math.pi | |
radius = (0.0189393939 / earth_radius_in_miles) * radians_to_degrees # 100 foot radius | |
bbox_earth_radius = 3959.0 # miles | |
bbox_dist_from_center = 0.1 # miles | |
def sum_by_distinct_type(distinct_type, distinct_key, lat, lon): | |
nearby_requests = db.windygrid.find({ | |
"where.location": { | |
"$within": {"$center": [[lat, lon], radius ]}}, | |
"type": aggregate_type, | |
distinct_key: distinct_type}) | |
return {distinct_type: nearby_requests.count()} | |
def sum_types_by_locations_in_query(search_type_query, search_mask, mask_name_key, distinct_key, aggregate_type, db, table): | |
cursor = db.windygrid.find(search_type_query, search_mask) | |
result = {} | |
location_counts = [] | |
print "running against query results of %d records" % cursor.count() | |
for doc in cursor: | |
if doc['where']['location'] == [-1, 1]: | |
print "invalid coordinates (-1, 1) for %s" % doc['_id'] | |
continue | |
lat = doc['where']['location'][0] | |
lon = doc['where']['location'][1] | |
# within circle | |
#nearby_requests = db.windygrid.find({ | |
# "where.location": { | |
# "$within": {"$center": [[lat, lon], radius ]}}, | |
# "type": aggregate_type}) | |
# | |
# within bounding box | |
ll_x = lon - math.degrees(bbox_dist_from_center / bbox_earth_radius / math.cos(math.radians(lat))) | |
ll_y = lat - math.degrees(bbox_dist_from_center / bbox_earth_radius) | |
ur_x = lon + math.degrees(bbox_dist_from_center / bbox_earth_radius / math.cos(math.radians(lat))) | |
ur_y = lat + math.degrees(bbox_dist_from_center / bbox_earth_radius) | |
print "bbox = lowerleft:[%f, %f], upperright[%f, %f]" % (ll_x, ll_y, ur_x, ur_y) | |
box = [[ll_y, ll_x], [ur_y, ur_x]] | |
nearby_requests = db.windygrid.find({"where.location": {"$within": {"$box" : box}}, "type": aggregate_type}) | |
count = nearby_requests.count() | |
distinct_types = db.windygrid.find({"type": aggregate_type}) \ | |
.distinct(distinct_key) | |
counts = [sum_by_distinct_type(t, distinct_key, lat, lon) \ | |
for t in distinct_types] | |
type_counts = [] | |
for c in counts: | |
keys = c.keys() | |
for key in keys: | |
if c[key] > 0: | |
type_counts.append({key: c[key]}) | |
print "processed %s; it had %d records near it" % \ | |
(doc[mask_name_key], count) | |
location_counts.append({"what": doc[mask_name_key], | |
"location": doc['where']['location'], | |
"total_count": count, | |
"type_counts": type_counts}) | |
result["location_counts"] = location_counts | |
return result | |
connection = Connection() | |
db = connection["test"] | |
search_type_query = {"type" : "business_licenses"} | |
search_mask = {"_id" : 0, "what.legal_name" : 1, "where.location" : 1} | |
mask_name_key = "what" | |
distinct_key = "what.type_code" | |
aggregate_type = "311" | |
table = "windygrid" | |
result = sum_types_by_locations_in_query(search_type_query, search_mask, mask_name_key, distinct_key, aggregate_type, db, table) | |
#could save to a db here | |
print result |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment