StravaFlyByDownloaderAndPlotter.py

#!/usr/bin/env python
# coding: utf-8

import urllib
import json
import os
import matplotlib.pyplot as plt
import datetime

import math

def distance(p1,p2):
    
    lat1, long1, lat2, long2 = [math.radians(float(x)) for x in 
                                  (p1[0], p1[1], p2[0], p2[1])]

    d_lat = lat2 - lat1
    d_long = long2 - long1
    a = math.sin(d_lat/2)**2 + math.cos(lat1) * math.cos(lat2) * math.sin(d_long/2)**2
    c = 2 * math.atan2(math.sqrt(a), math.sqrt(1-a))
    
    return 6378137 * c

def accumulator(lis):
    total = 0
    for x in lis:
        total += x
        yield total



mainActivity = 562503255 ### REPLACE BY YOUR OWN ACTIVITY HERE!! ###

maindir = str(mainActivity)

matchesUrl = "https://nene.strava.com/flyby/matches/{}".format(mainActivity)

matchesFname = '{}/matches.json'.format(maindir)

try:
    with open(matchesFname) as jsonin:
        matches = json.load(jsonin)
except:
    if not os.path.exists(maindir):
        os.makedirs(maindir)    
    matchesContent = urllib.urlopen(matchesUrl).read()
    with open(matchesFname, 'w') as jsonout:
        jsonout.write(matchesContent)
    matches = json.loads(matchesContent)



athletes = matches['athletes']

for match in matches['matches']:
    activityId = match['otherActivity']['id']
    athlId = str(match['otherActivity']['athleteId'])
    
    correlation = match['correlation']['spatialCorrelation']
    if correlation < 0.9:
        continue
    
    otherFname = '{}/{}.json'.format(mainActivity, activityId)
    try:
        with open(otherFname) as otherin:
            other = json.load(otherin)
    except:
        otherUrl = ("https://nene.strava.com"+
                    "/flyby/stream_compare/545858689/"+
                    "{}".format(activityId))
        otherContent = urllib.urlopen(otherUrl).read()
        with open(otherFname, 'w') as otherout:
            otherout.write(otherContent)
        other = json.loads(otherContent)


    stream = other['stream']

    times = []
    lats = []
    lons = []
    for point in stream:
        times.append(datetime.datetime.fromtimestamp(point['time']))
        lats.append(point['point']['lat'])
        lons.append(point['point']['lng'])

    dtimes = [0.0001] + [(times[n] - times[n-1]).total_seconds() for n in xrange(1, len(times))]
    cumtimes = [v / 3600 for v in accumulator(dtimes)]

    pts = zip(lats, lons)
    dists = [0] + [distance(pts[n], pts[n-1]) for n in xrange(1, len(pts))]
    cumdists = [v / 1000 for v in accumulator(dists)]

    averages = [x[0]/x[1] for x in zip(cumdists, cumtimes)]

    plt.figure(1)
    plt.plot(lons, lats, alpha=0.5)

    plt.figure(2)
    plt.plot(cumdists, averages, lw=2, alpha=0.5, label=athletes[athlId]['firstName'])

plt.figure(1)
plt.grid()
plt.axis('equal')

plt.figure(2)
plt.xlabel("Tempo (horas)")
plt.ylabel(u"Velocidade Média (km/h)")
plt.xlim(-5, 450)
plt.ylim(0, 35)
plt.legend(loc="lower left", prop={'size':8})
plt.grid()

plt.show()