Debian Codegraph @

recipes/wired.recipe

raw
1 __license__ = 'GPL v3'
2 __copyright__ = '2014, Darko Miletic <darko.miletic at gmail.com>'
3 '''
4 www.wired.com
5 '''
6
7 from calibre import browser
8 from calibre.web.feeds.news import BasicNewsRecipe
9
10
11 def classes(classes):
12 q = frozenset(classes.split(' '))
13 return dict(attrs={'class': lambda x: x and frozenset(x.split()).intersection(q)})
14
15
16 class WiredDailyNews(BasicNewsRecipe):
17 title = 'Wired Magazine, Monthly Edition'
18 __author__ = 'Darko Miletic, update by Howard Cornett, Zach Lapidus, Michael Marotta'
19 description = (
20 'Wired is a full-color monthly American magazine, '
21 'published in both print and online editions, that '
22 'reports on how emerging technologies affect culture, '
23 'the economy and politics. '
24 'Monthly edition, best run at the start of every month.'
25 )
26 publisher = 'Conde Nast'
27 category = 'news, IT, computers, technology'
28 oldest_article = 45
29 max_articles_per_feed = 200
30 no_stylesheets = True
31 encoding = 'utf-8'
32 use_embedded_content = False
33 language = 'en'
34 ignore_duplicate_articles = {'url'}
35 remove_empty_feeds = True
36 extra_css = '''
37 .entry-header{
38 text-transform: uppercase;
39 vertical-align: baseline;
40 display: inline;
41 }
42 ul li{display: inline}
43 '''
44
45 keep_only_tags = [
46 classes('content-header lead-asset article__body'),
47 ]
48 remove_tags = [
49 classes(
50 'related-cne-video-component tags-component callout--related-list iframe-embed podcast_storyboard'
51 ' inset-left-component ad consumer-marketing-component social-icons'
52 ),
53 dict(name=['meta', 'link']),
54 dict(id=['sharing', 'social', 'article-tags', 'sidebar']),
55 ]
56 remove_attributes = ['srcset']
57
58 def parse_wired_index_page(self, currenturl, seen):
59 self.log('Parsing index page', currenturl)
60 soup = self.index_to_soup(currenturl)
61 baseurl = 'https://www.wired.com'
62 for a in soup.find('div', {'class': 'multi-packages'}).findAll('a', href=True):
63 url = a['href']
64 if url.startswith('/story') and url.endswith('/'):
65 title = self.tag_to_string(a.parent.find('h3'))
66 dateloc = a.parent.find('time')
67 date = self.tag_to_string(dateloc)
68 if title.lower() != 'read more' and title and url not in seen:
69 seen.add(url)
70 self.log('Found article:', title)
71 yield {
72 'title': title,
73 'date': date,
74 'url': baseurl + url,
75 }
76
77 def parse_index(self):
78 baseurl = 'https://www.wired.com/magazine/?page={}/'
79 pagenum = 1
80 articles = []
81 seen = set()
82 for pagenum in range(1, 4):
83 articles.extend(self.parse_wired_index_page(baseurl.format(pagenum), seen))
84
85 return [('Magazine Articles', articles)]
86
87 # Wired changes the content it delivers based on cookies, so the
88 # following ensures that we send no cookies
89 def get_browser(self, *args, **kwargs):
90 return self
91
92 def clone_browser(self, *args, **kwargs):
93 return self.get_browser()
94
95 def open_novisit(self, *args, **kwargs):
96 br = browser()
97 return br.open_novisit(*args, **kwargs)
98
99 open = open_novisit