Дивиденды, выплаченные компаниями в 2018-2019-2020, взято с https://investmint.ru/

moex_dividend_yield_2018_2019_2020.csv

moex_dividend_yield_2018_2019_2020.py

import re
import requests

from bs4 import BeautifulSoup
from datetime import datetime


def parse_value(value):
    """парсит числовое значение в html таблице, возвращет float или None"""
    value = re.sub(r'\s+', '', value) \
        .split('(')[0] \
        .replace('$', '') \
        .replace('₽', '') \
        .replace('%', '') \
        .replace(',', '.')

    return float(value) if value else None


MONTHS = ['янв', 'фев', 'мар', 'апр', 'мая', 'июн', 'июл', 'авг', 'сен', 'окт', 'ноя', 'дек']
dividends = []


for year in [2018, 2019]:
    r = requests.get('https://investmint.ru/dividendnyj-kalendar-{}/'.format(year))
    soup = BeautifulSoup(r.text, 'html.parser')

    for row in soup.select('table tr'):
        if not row.select_one('.smallbadge'):
            continue

        ticker = row.select_one('.smallbadge').text

        close = parse_value(row.select_one('td:nth-of-type(3)').text)
        if not close:
            print('no price for {}: {}'.format(ticker, row.select_one('td:nth-of-type(3)').text))
            continue

        percent = parse_value(row.select_one('td:nth-of-type(5)').text)
        if not percent:
            print('no percent for {}: {}'.format(ticker, row.select_one('td:nth-of-type(5)').text))
            continue

        dividend_string = row.select_one('td:nth-of-type(4)').text
        if '$' in dividend_string:
            # для $ высчитываем дивиденд от цены закрытия и процента доходности
            dividend = close * percent * 0.01
        else:
            dividend = parse_value(dividend_string)
        if not dividend:
            print('no dividend for {}: {}'.format(ticker, row.select_one('td:nth-of-type(4)').text))
            continue

        dt = re.sub(r'\s+', '', row.select_one('td:nth-of-type(2)').text)
        for idx, month_name in enumerate(MONTHS):
            dt = dt.replace(month_name, '.{}.'.format(idx + 1))
        dt = datetime.strptime(dt, '%d.%m.%Y').strftime('%Y-%m-%d')

        dividends.append((ticker, dt, close, dividend, percent))
        # print(ticker, dt, close, dividend, percent)


for year in [2020]:
    # для 2020 отдельный парсер таблицы
    r = requests.get('https://investmint.ru/')
    soup = BeautifulSoup(r.text, 'html.parser')

    for row in soup.select('table tr[data-category*="all past ru 2020"]'):
        if not row.select_one('.smallbadge'):
            continue

        ticker = row.select_one('.smallbadge').text

        percent = parse_value(row.select_one('td:nth-of-type(6)').text)
        if not percent:
            print('no percent for {}: {}'.format(ticker, row.select_one('td:nth-of-type(6)').text))
            continue

        dividend_string = row.select_one('td:nth-of-type(5)').text
        if '$' in dividend_string:
            # пока цены нет, считаем цену рублях ¯\_(ツ)_/¯
            dividend = parse_value(dividend_string) * 63.5
        else:
            dividend = parse_value(dividend_string)
        if not dividend:
            print('no dividend for {}: {}'.format(ticker, row.select_one('td:nth-of-type(5)').text))
            continue

        # вычисляем цену закрытия
        close = dividend / (percent * 0.01)

        dt = re.sub(r'\s+', '', row.select_one('td:nth-of-type(3)').text)
        for idx, month_name in enumerate(MONTHS):
            dt = dt.replace(month_name, '.{}.'.format(idx + 1))
        dt = datetime.strptime(dt, '%d.%m.%Y').strftime('%Y-%m-%d')

        dividends.append((ticker, dt, close, dividend, percent))
        # print(ticker, dt, close, dividend, percent)


with open('moex_dividend_yield_2018_2019_2020.csv', 'w') as f:
    f.write('ticker,datetime,close,dividend,percent\n')
    for item in dividends:
        f.write('{},{},{},{},{}\n'.format(*item))