def get_data(self):
""" Get the houses for rent """
try:
data = urllib2.urlopen(self.url)
#Tricky part
result = json.load(data)
aux = json.dumps(result)
aux2 = json.dumps(json.loads(aux))
#Data Frame
data_frame = pd.read_json(aux2)
data_frame = data_frame.dropna()
distances = []
#Compute the distance between two points useing harvesine formule
#print(data_frame)
for i, r in data_frame.iterrows():
ans = haversine(r['longitude'], r['latitude'])
distances.append(ans)
data_frame['distance'] = pd.Series(distances,
index=data_frame.index)
return data_frame
except Exception as e:
print("error", str(e))
return []
print("================ Extracting gps locations===================")
max_latitude = -500
min_latitude = 500
max_longitude = -500
min_longitude = 500
lon_lat_list = []
for file in gpx_files:
gpx_file = open(file,'r')
gpx = gpxpy.parse(gpx_file)
lon = []
lat = []
last_point = (0,0)
for track in gpx.tracks:
for segment in track.segments:
for gps_point in segment.points:
if(functions.haversine(last_point, (gps_point.longitude, gps_point.latitude)) > 2):#to avoid bias in heat in uphills, when you move slower, the haversine distance between two points has to be at least 2 meters
lon.append(gps_point.longitude)
lat.append(gps_point.latitude)
if(max_latitude < gps_point.latitude):
max_latitude = gps_point.latitude
if(min_latitude > gps_point.latitude):
min_latitude = gps_point.latitude
if(max_longitude < gps_point.longitude):
max_longitude = gps_point.longitude
if(min_longitude > gps_point.longitude):
min_longitude = gps_point.longitude
last_point = (gps_point.longitude, gps_point.latitude)
lon_lat_list.append((lon,lat))
print("Done in: ", time.time() - lap, " s")
lap = time.time()
lon1 = 283.90
lat1 = -14.37 # SAm 1
lon2 = 286.76
lat2 = -33.01 # SAm 2
lon3 = 143.07
lat3 = 25.33 # Mariana
rad_km = 6378.
thresh_dist = 200.
# every-so-often case
num = 0
for n in range(len(dips)):
lon = dips[n, 0]
lat = dips[n, 1]
dist1 = haversine(lon, lat, lon1, lat1, rad_km)
dist2 = haversine(lon, lat, lon2, lat2, rad_km)
dist3 = haversine(lon, lat, lon3, lat3, rad_km)
if dist1 < thresh_dist or dist2 < thresh_dist or dist3 < thresh_dist:
pass
else:
num = num + 1
dips_new = np.zeros((num, 3))
num = 0
for i in range(len(dips)):
lon = dips[i, 0]
lat = dips[i, 1]
dist1 = haversine(lon, lat, lon1, lat1, rad_km)
dist2 = haversine(lon, lat, lon2, lat2, rad_km)
dist3 = haversine(lon, lat, lon3, lat3, rad_km)
num_obs = len(dips_lall)
slab2_dips_LallLocations = np.zeros((num_obs, 4))
for i in range(num_obs):
lon_obs = dips_lall[i, 0]
lat_obs = dips_lall[i, 1]
dist_max = 5000.0
dist_ind = 1e6
for j in range(len(dips_slab2_full)):
lon_calc = dips_slab2_full[j, 0]
lat_calc = dips_slab2_full[j, 1]
dist = haversine(lon_obs, lat_obs, lon_calc, lat_calc, 6378.)
if dist < dist_max:
dist_max = dist
dist_ind = j
if dist_ind != 1e6:
slab2_dips_LallLocations[i, 0] = dips_slab2_full[dist_ind, 0]
slab2_dips_LallLocations[i, 1] = dips_slab2_full[dist_ind, 1]
slab2_dips_LallLocations[i, 2] = dips_slab2_full[dist_ind, 2]
slab2_dips_LallLocations[i, 3] = dist_max
slab2cont_dips_LallLocations = np.zeros((num_obs, 4))
for i in range(num_obs):
lon_obs = dips_lall[i, 0]
lat_obs = dips_lall[i, 1]
seg_dist = np.zeros((num_obs))
is_trench_fixed = np.zeros((num_obs))
is_just_upper_mant = np.zeros((num_obs))
num_good_subds = 0
for i in range(num_obs):
lon_obs = dips_obs_tmp[i, 0]
lat_obs = dips_obs_tmp[i, 1]
dist_max = 1000.0
for j in range(num_DPs):
lon_mod = DPs_mod[j, 0]
lat_mod = DPs_mod[j, 1]
dist = haversine(lon_obs, lat_obs, lon_mod, lat_mod, 6378.)
if dist < dist_max and iwall[j] == 1:
dist_max = dist
dist_ind = j
lon_mod = DPs_mod[dist_ind, 0]
lat_mod = DPs_mod[dist_ind, 1]
DP = DPs_mod[dist_ind, 4]
age = sp_ages[dist_ind, 2]
vc = DPs_mod[dist_ind, 5]
vt = DPs_mod[dist_ind, 6]
crust_density = 3450 # rho_eclogite [kg/m3], C-T Lee, W-P Chen, 2007, EPSL
crust_thick = 7.5e3 # m
# fill array...
# get age coordinates if iwall[i] == 1: if polarity[i] == 1: # dipping to left age_azim = alpha[i] + (np.pi/2.) else: # dipping to right age_azim = alpha[i] - (np.pi/2.) lon_age_rad , lat_age_rad = project_to_point(np.deg2rad(lono[i]), np.deg2rad(lato[i]), age_azim, age_len_ratio) if lon_age_rad < 0.: lon_age_rad = (2.*np.pi) + lon_age_rad sp_ages[i,0] = math.degrees(lon_age_rad) sp_ages[i,1] = math.degrees(lat_age_rad) dist_max = 1.e9 for j in range(0,len(age_txt)): dist_pt = haversine(sp_ages[i,0], sp_ages[i,1], age_txt[j,0], age_txt[j,1],6378.) if dist_pt < dist_max: dist_max = dist_pt closest_ind = j print i sp_ages[i,2] = age_txt[closest_ind,2] print "getting rid of nans..." for i in range(len(sp_ages)): if np.isnan(sp_ages[i,2]): dist_max = 1.e9 for k in range(0,len(sp_ages)): dist_pt = haversine(sp_ages[i,0], sp_ages[i,1], sp_ages[k,0], sp_ages[k,1],6378.)
import json
sys.path.append("..")
from functions import generate_random_string, haversine
from app import app
from models import db, Stops, Links, Edges, Routes
index = 0.0
with app.app_context():
stops = Stops.query.all()
num_stops = len(stops) * 1.0
for begin_stop in stops:
index += 1.0
print(begin_stop.id)
print(str(index / num_stops))
for end_stop in stops:
if begin_stop.id != end_stop.id:
distance = haversine(begin_stop.lng, begin_stop.lat,
end_stop.lng, end_stop.lat)
existing_edge = Edges.query.filter_by(
beginning_stop=begin_stop.id).filter_by(
ending_stop=end_stop.id).first()
if existing_edge == None:
if distance < 1:
new_edge = Edges(generate_random_string(10, 20),
begin_stop.id, end_stop.id, distance)
db.session.add(new_edge)
db.session.commit()
print("Finished")