将 Highcharts 数据抓取到 CSV 文件

Question

我想使用 Python 从CoinMarketCap website 中抓取特定硬币的历史数据。我想要尽可能细化的数据。我认为数据来自以下字符串：

我想使用 BeautifulSoup 来获取这些数据并将其保存到 CSV 文件，但我不知道如何。

Answer 1

它似乎使用了使用 url 的 JavaScript

https://graphs.coinmarketcap.com/currencies/verge/1513624444000/1514229244000/

以 JOSN 格式获取数据，以便您可以轻松地将其作为 python 字典获取

编辑： 它在 url 中使用 timestamp*1000 来获取历史数据。它写在csv

import requests
import datetime
import csv

date1 = '2016.01.01'
date2 = '2017.01.01'

date1 = datetime.datetime.strptime(date1, '%Y.%m.%d')
date2 = datetime.datetime.strptime(date2, '%Y.%m.%d')

date1 = int(date1.timestamp() * 1000)
date2 = int(date2.timestamp() * 1000)

print('dates:', date1, date2)
print('-----')

url = 'https://graphs.coinmarketcap.com/currencies/verge/{}/{}/'.format(date1, date2)

response = requests.get(url)
data = response.json()

for key in data.keys():
    print('key:', key)
print('-----')

f = open('output.csv', 'w')
csv_writer = csv.writer(f)

row = ('date', 'price_btc', 'price_usd', 'volume_usd')
csv_writer.writerow(row)

for item1, item2, item3 in zip(data['price_btc'], data['price_usd'], data['volume_usd']): #[:10]:

    date = datetime.datetime.fromtimestamp(item1[0]//1000)
    date = date.strftime('%Y.%m.%d %H:%M:%S')

    print('date:', date)
    print(' btc:', item1[1])
    print(' usd:', item2[1])
    print(' vol:', item3[1])
    print('-----')

    row = (date, item1[1], item3[1], item3[1])
    csv_writer.writerow(row)

f.close()    

部分结果：

dates: 1451602800000 1483225200000
-----
key: market_cap_by_available_supply
key: price_btc
key: price_usd
key: volume_usd
-----
date: 2016.01.01 00:04:19
 btc: 3.00032e-08
 usd: 1.29262e-05
 vol: 548
-----
date: 2016.01.02 00:04:19
 btc: 2.1964e-08
 usd: 9.52678e-06
 vol: 246
-----
date: 2016.01.03 00:04:19
 btc: 2.67174e-08
 usd: 1.15926e-05
 vol: 2805
-----

---

编辑：循环读取

import requests
import datetime
import csv
import webbrowser


def get_data(name, timestamp1, timestamp2, csv_writer):

    url = 'https://graphs.coinmarketcap.com/currencies/{}/{}/{}/'.format(name, timestamp1, timestamp2)

    response = requests.get(url)

    try:
        data = response.json()
    except Exception:
        with open('output.html', 'w') as f:
            f.write(response.text)
        webbrowser.open('output.html')
        exit()

    for item1, item2, item3 in zip(data['price_btc'], data['price_usd'], data['volume_usd']): #[:10]:

        date = datetime.datetime.fromtimestamp(item1[0]//1000)
        date = date.strftime('%Y.%m.%d %H:%M:%S')

        row = (date, item1[1], item3[1], item3[1])
        csv_writer.writerow(row)


def scrape(name, start_date, finish_date):

    f = open(name + '.csv', 'w')
    csv_writer = csv.writer(f)

    row = ('date', 'price_btc', 'price_usd', 'volume_usd')
    csv_writer.writerow(row)

    one_day = datetime.timedelta(days=1)

    start_date = datetime.datetime.strptime(start_date, '%Y.%m.%d')
    finish_date = datetime.datetime.strptime(finish_date, '%Y.%m.%d')

    date1 = start_date
    date2 = start_date + one_day

    while date1 < finish_date:
        print(name, date1, date2)

        date1_timestamp = int(date1.timestamp() * 1000)
        date2_timestamp = int(date2.timestamp() * 1000)

        get_data(name, date1_timestamp, date2_timestamp, csv_writer)

        date1 = date2
        date2 += one_day

    f.close()    

# --- main ---

scrape('verge', '2016.01.01', '2017.01.01')
scrape('bitcoin', '2016.01.01', '2017.01.01')
scrape('ethereum', '2016.01.01', '2017.01.01')