У меня есть код, который очищает футбольные матчи.
Как я могу улучшить свой код?
import os
import pandas as pd
from selenium import webdriver
from tabulate import tabulate
from datetime import datetime
import time
from bs4 import BeautifulSoup as bs
start = datetime.now()
browser = webdriver.Chrome()
class GameData:
def __init__(self):
self.date = []
self.time = []
self.game = []
self.score = []
self.home_odds = []
self.draw_odds = []
self.away_odds = []
def parse_data(url):
browser.get(url)
df = pd.read_html(browser.page_source, header=0)[0]
time.sleep(3)
html = browser.page_source
soup = bs(html,"lxml")
cont = soup.find('div', {'id':'wrap'})
conti = cont.find('div', {'id':'col-content'})
content = conti.find('table', {'class':'table-main'}, {'id':'tournamentTable'})
main = content.find('th', {'class':'first2 tl'})
count = main.findAll('a')
country = count[1].text
league = count[2].text
game_data = GameData()
game_date = None
for row in df.itertuples():
if not isinstance(row[1], str):
continue
elif ':' not in row[1]:
game_date = row[1].split('-')[0]
continue
game_data.date.append(game_date)
game_data.time.append(row[1])
game_data.game.append(row[2])
game_data.score.append(row[3])
game_data.home_odds.append(row[4])
game_data.draw_odds.append(row[5])
game_data.away_odds.append(row[6])
browser.quit()
return game_data, country, league
# You can input as many URLs you want
urls = {
"https://www.oddsportal.com/soccer/europe/champions-league/results/"
}
if __name__ == '__main__':
results = None
for url in urls:
game_data, country, competition = parse_data(url)
result = pd.DataFrame(game_data.__dict__)
if results is None:
results = result
else:
results = results.append(result, ignore_index=True)
print('The country is',country)
print('The competition is', competition)
print(tabulate(results.head(), headers="keys", tablefmt="github"))
end = datetime.now()
time_taken = end - start
print('Time taken to complete: ', time_taken)
результат:
The country is Europe
The competition is Champions League
| | date | time | game | score | home_odds | draw_odds | away_odds |
|----|-------------|--------|--------------------------------------|---------|-------------|-------------|-------------|
| 0 | 17 Mar 2021 | 20:00 | Bayern Munich - Lazio | 2:1 | 1.35 | 6.02 | 7.53 |
| 1 | 17 Mar 2021 | 20:00 | Chelsea - Atl. Madrid | 2:0 | 2.33 | 3.1 | 3.49 |
| 2 | 16 Mar 2021 | 20:00 | Manchester City - B. Monchengladbach | 2:0 | 1.28 | 6.25 | 10.17 |
| 3 | 16 Mar 2021 | 20:00 | Real Madrid - Atalanta | 3:1 | 2.26 | 3.6 | 3.16 |
| 4 | 10 Mar 2021 | 20:00 | Liverpool - RB Leipzig | 2:0 | 2.37 | 3.8 | 2.85 |
Time taken to complete: 0:00:15.875088
ожидаемый результат:
date time country competition game score home_odds draw_odds away_odds
--- -------- -------- -------- -------- -------------------------------------- --------- ------------ ------------- ------------
0 17/03/2021 20:00 Europe Champions League Bayern Munich - Lazio 2:01 1.35 6.02 7.53
1 17/03/2021 20:00 Europe Champions League Chelsea - Atl. Madrid 2:00 2.33 3.1 3.49
2 16/03/2021 20:00 Europe Champions League Manchester City - B. Monchengladbach 2:00 1.28 6.25 10.17
3 16/03/2021 20:00 Europe Champions League Real Madrid - Atalanta 3:01 2.26 3.6 3.16
4 10/03/2021 20:00 Europe Champions League Liverpool - RB Leipzig 2:00 2.37 3.8 2.85
** требуется объяснение с помощью codereview .. Я добавил обновленный фрейм данных, чтобы отразить ожидаемые результаты.