extract_universities.py 2.04 KB
Newer Older
Florent Chehab's avatar
Florent Chehab committed
1
2
3
4
5
6
7
#!/usr/bin/env python3

"""
Script to insert the country data in the database

IT HAS TO BE RUN INSIDE ./manage.py shell
"""
Florent Chehab's avatar
linting    
Florent Chehab committed
8
import csv
Florent Chehab's avatar
Florent Chehab committed
9
10
11
import os
import time
from geopy.geocoders import Nominatim
Florent Chehab's avatar
Florent Chehab committed
12
import reverse_geocoder as rg
Florent Chehab's avatar
Florent Chehab committed
13

14
tmp = os.path.join(os.path.realpath(__file__), "../../assets/destinations.csv")
Florent Chehab's avatar
Florent Chehab committed
15
16
destinations_path = os.path.abspath(tmp)

17
18
19
tmp = os.path.join(
    os.path.realpath(__file__), "../../assets/destinations_extracted.csv"
)
Florent Chehab's avatar
Florent Chehab committed
20
21
22
23
24
25
destinations_extracted_path = os.path.abspath(tmp)

if not os.path.isfile(destinations_path):
    print(destinations_path)
    raise Exception("Missing file containing country data")

26
27
with open(destinations_path, "rt") as input:
    with open(destinations_extracted_path, "w") as output:
Florent Chehab's avatar
Florent Chehab committed
28
29
30
31
32
33
34
35
36
37
        print("ini")
        reader = csv.reader(input)
        spamwriter = csv.writer(output, quoting=csv.QUOTE_NONNUMERIC)
        geolocator = Nominatim()

        failed = []
        i = 0
        for row in reader:
            # handle the header
            if i == 0:
38
                header = ["university", "city", "country", "lat", "lon"]
Florent Chehab's avatar
Florent Chehab committed
39
40
41
                spamwriter.writerow(header)
                i += 1
            else:
Florent Chehab's avatar
linting    
Florent Chehab committed
42
43
                query = row[2]  # + ', ' + row[1] + ', ' + row[0]

Florent Chehab's avatar
Florent Chehab committed
44
45
46
47
                while True:
                    try:
                        location = geolocator.geocode(query)
                        break
Florent Chehab's avatar
linting    
Florent Chehab committed
48
                    except:  # noqa: E722
Florent Chehab's avatar
Florent Chehab committed
49
                        print("error during query, retrying")
Florent Chehab's avatar
Florent Chehab committed
50
                        time.sleep(0.5)
Florent Chehab's avatar
Florent Chehab committed
51
                if location is not None:
Florent Chehab's avatar
Florent Chehab committed
52
53
                    coord = (location.latitude, location.longitude)
                    res = rg.search(coord, verbose=False)
54
55
56
57
58
59
60
                    line = [
                        row[2],
                        row[1],
                        res[0]["cc"],
                        location.latitude,
                        location.longitude,
                    ]
Florent Chehab's avatar
Florent Chehab committed
61
62
                    print(line)
                    spamwriter.writerow(line)
Florent Chehab's avatar
Florent Chehab committed
63
64
65
                else:
                    failed.append(query)

Florent Chehab's avatar
linting    
Florent Chehab committed
66
        print(failed)