-
Notifications
You must be signed in to change notification settings - Fork 172
/
Tomtom.py
225 lines (195 loc) · 6.63 KB
/
Tomtom.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# vim: filetype=python
import pandas as pd
import datetime as dt
import time
import numpy as np
import requests
import os
os.chdir('k:/')
#
def scrape(city):
prefix = 'https://api.midway.tomtom.com/ranking/liveHourly/'
session = requests.Session()
response = session.get(prefix+city)
return response
#
def etl(rawdata,target,city,historic_avg):
# json keys
cols = rawdata['data'][0].keys()
# add some missing column
for i in range(len(rawdata['data'])):
for j in cols:
if j not in rawdata['data'][i].keys():
rawdata['data'][i][j] = np.nan
df = pd.DataFrame()
# fill in data
for col in cols:
df[col] = [i[col] for i in rawdata['data']]
# there is only system time
t0 = dt.datetime(1970, 1, 1, 1, 44)
# convert system time to real time
df['datetime'] = [
t0 + dt.timedelta(minutes=i / 60000) for i in df['UpdateTime'].tolist()
]
# change column name
df.columns = df.columns.str.replace('TrafficIndexLive', 'LiveCongestion')
df.columns = df.columns.str.replace(
'TrafficIndexHistoric', 'LastYearAverageCongestion'
)
# get daily average
df['datetime'] = pd.to_datetime(df['datetime'])
datelist = set(df['datetime'].dt.date)
df.set_index('datetime', inplace=True)
# create cols
df['LiveCongestionDaily'] = np.nan
df['LastYearAverageCongestionDaily'] = np.nan
df['location'] = target[city]['location']
df['country'] = target[city]['country']
# create daily average
for i in datelist:
df['LiveCongestionDaily'][
i.strftime('%Y-%m-%d') : i.strftime('%Y-%m-%d')
] = df['LiveCongestion'][
i.strftime('%Y-%m-%d') : i.strftime('%Y-%m-%d')
].mean()
# there used to be last year avg
# if it reappears, take daily average instead of 15 min interval by default
if "LastYearAverageCongestion" in df.columns:
df['LastYearAverageCongestionDaily'][
i.strftime('%Y-%m-%d') : i.strftime('%Y-%m-%d')
] = df['LastYearAverageCongestion'][
i.strftime('%Y-%m-%d') : i.strftime('%Y-%m-%d')
].mean()
# if no historic, use historic avg
else:
df['LastYearAverageCongestionDaily'][
i.strftime('%Y-%m-%d') : i.strftime('%Y-%m-%d')
] = historic_avg[target[city]['location']][dt.datetime.weekday(i)]
# create output
df.reset_index(inplace=True)
df.to_csv(f'{target[city]["location"]}.csv')
#
def main():
# target to be scraped
target = {
'FRA%2FCircle%2Fparis': {'country': 'France', 'location': 'Paris'},
'ITA%2FCircle%2Fmilan': {'country': 'Italy', 'location': 'Milan'},
'DEU%2FCircle%2Ffrankfurt-am-main': {
'country': 'Germany',
'location': 'Frankfurt',
},
'GBR%2FCircle%2Flondon': {'country': 'United Kingdom', 'location': 'London'},
'USA%2FCircle%2Fnew-york': {'country': 'United States', 'location': 'New York'},
'JPN%2FCircle%2Ftokyo': {'country': 'Japan', 'location': 'Tokyo'},
'AUS%2FCircle%2Fsydney': {'country': 'Australia', 'location': 'Sydney'},
'ESP%2FCircle%2Fmadrid': {'country': 'Spain', 'location': 'Madrid'},
'USA%2FCircle%2Flos-angeles': {
'country': 'United States',
'location': 'Los Angeles',
},
'USA%2FCircle%2Fseattle': {'country': 'United States', 'location': 'Seattle'},
}
# tomtom used to offer historical data in api
# now we have to hardcode the number
historic_avg = {
'Frankfurt': {
0: 14.828168159761104,
1: 18.556550951847704,
2: 18.764821684086105,
3: 20.81831114679017,
4: 15.212893625192013,
5: 9.440824468085108,
6: 5.451007326007326,
},
'London': {
0: 21.20389254385965,
1: 25.025871360582304,
2: 25.890477245862883,
3: 27.638587079798576,
4: 24.563016917293233,
5: 18.371318922305765,
6: 14.034886809414841,
},
'Los Angeles': {
0: 18.081597222222225,
1: 24.602430555555554,
2: 27.961805555555557,
3: 29.181798245614033,
4: 27.713230861965037,
5: 23.43154761904762,
6: 13.758666928309788,
},
'Madrid': {
0: 12.490570175438597,
1: 13.618031189083823,
2: 14.09101382667662,
3: 14.179331140350877,
4: 12.251941150954309,
5: 4.372204447288434,
6: 2.938329142699487,
},
'Milan': {
0: 16.595997807017547,
1: 19.678281697150677,
2: 20.116642559412714,
3: 21.798127320117878,
4: 20.93199688049912,
5: 11.182520463392523,
6: 7.4401126039613885,
},
'New York': {
0: 16.63888888888889,
1: 20.151041666666668,
2: 20.938764732923374,
3: 22.204457295793247,
4: 21.864376130198917,
5: 14.28361528822055,
6: 10.565672422815279,
},
'Paris': {
0: 22.678165437974368,
1: 27.70737293144208,
2: 27.65354658845982,
3: 29.25440264472295,
4: 28.879417293233082,
5: 15.110823934837091,
6: 11.962517707311758,
},
'Seattle': {
0: 12.897569444444445,
1: 19.878472222222225,
2: 21.339887521222412,
3: 22.007419590643277,
4: 19.98502486437613,
5: 15.204010025062656,
6: 8.828100470957613,
},
'Sydney': {
0: 16.886235062293416,
1: 19.2371895783413,
2: 20.033814183747694,
3: 20.24800293601769,
4: 17.27066753884507,
5: 13.693233082706767,
6: 12.606057987711214,
},
'Tokyo': {
0: 22.880642162471393,
1: 24.30436652357845,
2: 23.2880849082068,
3: 25.036028679855665,
4: 26.748143194524776,
5: 22.496804511278196,
6: 15.289682095309194,
}}
for city in target:
time.sleep(5)
print(city)
response = scrape(city)
rawdata = response.json()
etl(rawdata,target,city,historic_avg)
return
if __name__ == "__main__":
main()