Last active
February 19, 2020 21:53
-
-
Save olarclara/2179e88f022b7875e86b73405538ea50 to your computer and use it in GitHub Desktop.
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 requests | |
import math | |
import json | |
import pandas as pd | |
# IronMan WC 2019 id | |
eventId = 'AFAAC1DC-73BA-E811-A967-000D3A37468C' | |
limitPerPage = 100 | |
def requestData(page): | |
skip = page * limitPerPage | |
url = f'https://data.competitor.com/result/subevent/{eventId}?%24limit={limitPerPage}&%24skip={skip}&%24sort%5BFinishRankOverall%5D=1' | |
response = requests.get(url) | |
content = json.loads(response.content) | |
return content | |
def parseData(data): | |
results = data | |
for result in results: | |
copy = result | |
copy['FullName'] = result['Contact']['FullName'] | |
copy['Gender'] = result['Contact']['Gender'] | |
copy['Country'] = result['Country']['ISO2'] | |
del copy['Contact'] | |
return results | |
initialRequest = requestData(0) | |
totalPages = initialRequest['total'] | |
data = initialRequest['data'] | |
results = parseData(data) | |
for page in range(1, math.ceil(totalPages/limitPerPage)): | |
requestResponse = requestData(page) | |
results.extend(parseData(requestResponse['data'])) | |
df = pd.DataFrame(results)[['AgeGroup', 'BikeTimeConverted', 'FullName', 'Gender', 'Country', 'EventStatus', 'FinishRankGender', 'FinishRankGroup', | |
'FinishRankOverall', 'FinishTimeConverted', 'RunTimeConverted', 'SwimTimeConverted', 'Transition1TimeConverted', 'Transition2TimeConverted']] | |
df.to_csv('results.csv') |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment