noball/http_root/magazine.json

#!/usr/bin/env python3

import datetime
import json
import re

import bs4
import requests

print('Content-Type: application/json')
print()
cricinfo = []

soup = bs4.BeautifulSoup(requests.get('http://www.espncricinfo.com/ci/content/story/features.html').text, 'html.parser')

for article in map(lambda title: title.parent, soup.find_all("h2", class_="story-title")):
    cricinfo.append({
        'title': article.find('h2', class_='story-title').get_text(),
        'url': 'http://www.espncricinfo.com' + article.find('h2', class_='story-title').a['href'],
        'description': article.find('p', class_='story-brief').contents[1].strip(),
        'date': (datetime.datetime.strptime(article.find('strong', class_='story-date').get_text(), "%b %d, %Y") - datetime.datetime.fromtimestamp(0)).total_seconds() * 1000,
        'author': article.find('div', class_='author').get_text().strip()
    })
    imageParentTag = article.find('figure', class_='story-img')
    if imageParentTag:
        cricinfo[-1]['image'] = imageParentTag.img['src']
    
print(json.dumps(cricinfo))