Thanks truth1ness! What a great update. I just ran the script and had to make a slight tweak to get it to fetch the "long read" article about the silk road -- had to add something to the keep_only_tags to get the article to actually download, as it is (sigh) in a different format:
Code:
__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 = 2
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(id=['sharing', 'social', 'article-tags', 'sidebar']),
]
keep_only_tags=[
dict(attrs={'data-js':['post', 'postHeader']}),
dict(attrs={'class':'exchange fsb-content relative'})
]
def get_date_url(self):
'''
get month and year, add year modifier, append to wired magazine url,
:return: url
'''
baseurl = 'http://www.wired.com/tag/magazine-'
monthurl = str('{:02d}'.format(date.today().month))
yearurl = str(date.today().year - 1992)
dateurl = baseurl + yearurl + '-' + monthurl + '/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('http://www.wired.com/') and url.endswith('/'):
title = self.tag_to_string(a.find('h2'))
dateloc = a.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
return [('Articles', articles)]
Only one problem -- the svg images used in the article appear as black rectangles in both Calibre's viewer, and on the kindle. I tried fetching the news source as an ePub instead, thinking that that format did not rasterize svg, but got the black squares again. Any ideas?