calibre/recipes/wired.recipe
Kovid Goyal c370f9e8d1 ...
2016-10-09 18:20:33 +05:30

110 lines
4.2 KiB
Plaintext

__license__ = 'GPL v3'
__copyright__ = '2014, Darko Miletic <darko.miletic at gmail.com>'
'''
www.wired.com
'''
from calibre.web.feeds.news import BasicNewsRecipe
from datetime import date
import urllib2
class WiredDailyNews(BasicNewsRecipe):
title = 'Wired Magazine, Monthly Edition'
__author__ = 'Darko Miletic, update by Zach Lapidus, Michael Marotta'
description = ('Wired is a full-color monthly American magazine, '
'published in both print and online editions, that '
'reports on how emerging technologies affect culture, '
'the economy and politics. '
'Monthly edition, best run at the start of every month.')
publisher = 'Conde Nast'
category = 'news, IT, computers, technology'
oldest_article = 45
max_articles_per_feed = 200
no_stylesheets = True
encoding = 'utf-8'
use_embedded_content = False
language = 'en'
ignore_duplicate_articles = {'url'}
remove_empty_feeds = True
publication_type = 'newsportal'
extra_css = """
.entry-header{
text-transform: uppercase;
vertical-align: baseline;
display: inline;
}
"""
remove_tags = [
dict(name=['meta', 'link']),
dict(name='div', attrs={'class': 'podcast_storyboard'}),
dict(name='figure', attrs={'data-js': 'slide'}),
dict(name='div', attrs={'class': 'no-underline fix-height'}),
dict(name='div',
attrs={'class': 'no-underline marg-t-med big-marg-b-med fix-height'}),
dict(id=['sharing', 'social', 'article-tags', 'sidebar', 'related']),
]
keep_only_tags = [
dict(attrs={'data-js': ['post', 'postHeader']}),
dict(attrs={'class': 'exchange fsb-content relative'}),
]
def get_magazine_year_month(self, seperator):
monthurl = str('{:02d}'.format(date.today().month))
yearurl = str(date.today().year - 1992)
return yearurl + seperator + monthurl
def get_date_url(self):
'''
get month and year, add year modifier, append to wired magazine url,
:return: url
'''
baseurl = 'https://www.wired.com/tag/magazine-'
magazine_year_month = self.get_magazine_year_month('-')
# monthurl = str('{:02d}'.format(date.today().month))
# yearurl = str(date.today().year - 1992)
dateurl = baseurl + magazine_year_month + '/page/'
return dateurl
def parse_wired_index_page(self, currenturl, seen):
soup = self.index_to_soup(currenturl)
for a in soup.find('main').findAll('a', href=True):
url = a['href']
if url.startswith('https://www.wired.com/') and url.endswith('/'):
title = self.tag_to_string(a.parent.find('h2'))
dateloc = a.parent.find('time')
date = self.tag_to_string(dateloc)
if title.lower() != 'read more' and title and url not in seen:
seen.add(url)
self.log('Found article:', title)
yield {
'title': title,
'date': date,
'url': url,
'description': ''
}
def parse_index(self):
'''
get the current month's url, index first page to soup,
find number of pages, just keep adding to page num until
soup is not none instead of scraping page for :return:
'''
baseurl = self.get_date_url()
pagenum = 1
articles = []
seen = set()
morepages = True
while morepages:
try:
urllib2.urlopen(baseurl + str(pagenum))
currenturl = baseurl + str(pagenum)
articles.extend(self.parse_wired_index_page(currenturl, seen))
pagenum += 1
except urllib2.HTTPError:
morepages = False
magazine_year_month = self.get_magazine_year_month('.')
return [('Magazine-' + magazine_year_month, articles)]