Debian Codegraph @

recipes/heise_ix.recipe

raw
1 from calibre.web.feeds.recipes import BasicNewsRecipe
2
3 __license__ = 'GPL v3'
4 __copyright__ = 'Ralf Hein - ralfhein at GMX dot DE'
5 '''
6 Heise Select Magazine - iX
7 '''
8
9
10 class heise_select(BasicNewsRecipe):
11 issue = None
12 # overwrite this for easy download of previous issues
13 # issue = '/select/ix/2020/3'
14
15 title = 'iX'
16 timefmt = ''
17 __author__ = 'Ralf Hein'
18 needs_subscription = True
19 description = 'Das iX Magazin als ePaper. Benötigt Heise Plus Digitalabo (siehe https://www.heise.de/plus/)'
20 publisher = 'Heise Verlag'
21 authors = 'Heise Verlag'
22 category = 'it'
23 tags = 'Magazin, IT, computer, ix'
24 publication_type = 'magazine'
25 no_stylesheets = True
26 use_embedded_content = False
27 compress_news_images = True
28 encoding = 'utf-8'
29 language = 'de'
30
31 conversion_options = {'base_font_size': 10, 'no_inline_navbars': True, 'language': language, 'publisher': publisher, 'authors': publisher}
32
33 # some code cleanup
34 remove_tags = [
35 dict(name='meta'),
36 dict(name='link', attrs={'rel': 'icon'}),
37 dict(name='link', attrs={'rel': 'dns-prefetch'}),
38 dict(name='link', attrs={'rel': 'preconnect'}),
39 dict(name='div', attrs={'class': 'meta__group--issue'}),
40 dict(name='p', attrs={'class': 'comment'}),
41 dict(name='div', attrs={'class': 'pswp'}),
42 dict(name='div', attrs={'class': 'bottom-links'}),
43 ]
44
45 # content is neatly within <main> element
46 remove_tags_before = [dict(name='main')]
47 remove_tags_after = [dict(name='main')]
48
49 def parse_index(self):
50 baseref = 'https://www.heise.de'
51 # find current issue if not defined
52 if self.issue is None:
53 soup = self.index_to_soup(baseref + '/select')
54 sec = soup.find('section', attrs={'class': 'magazine--ix'})
55 self.issue = sec.find('a', attrs={'class': 'magazine__link--issue'}, href=True)['href']
56
57 issue_num = self.issue.replace('/select/ix/', '')
58 # fix title with issue number to keep them neatly organised
59 self.title += ' ' + issue_num.replace('/', '-')
60 self.cover_url = 'https://www.heise.de/select/thumbnail/ix/' + issue_num
61
62 soup = self.index_to_soup(baseref + self.issue)
63 toc = []
64
65 for h3 in soup.findAll('h3', attrs={'class': 'xp__inhalt__title'}):
66 section_title = h3.text
67 articles = []
68 ul = h3.find_next('ul')
69
70 for li in ul.findAll('li', attrs={'class': 'xp__toc__item'}):
71 article_uri = li.find('a', attrs={'class': 'xp__link'})['href']
72 article_title = li.find('span', attrs={'class': 'xp__toc__item-subtitle'}).text
73 article = {'title': article_title, 'url': baseref + article_uri}
74 articles.append(article)
75 toc.append((section_title, articles))
76
77 return toc
78
79 def get_browser(self):
80 br = BasicNewsRecipe.get_browser(self)
81 if self.username is not None and self.password is not None:
82 loginURL = 'https://www.heise.de/sso/login?forward=%2Fselect'
83 br.open(loginURL)
84 br.select_form(action='/sso/login/login')
85 br['username'] = self.username
86 br['password'] = self.password
87 br.submit()
88
89 return br
90
91 def preprocess_html(self, soup):
92 # images are dynamically sized via js + a-img tag, epub can not work with this
93 # construct ordinary img from it
94 for aimg in soup.findAll('a', attrs={'class': 'js-pswp-image'}):
95 if aimg['href'] is not None and aimg['data-pswp-bu'] is not None:
96 img = soup.new_tag('img', src=aimg['href'], alt=aimg['data-pswp-bu'], style='display: block;')
97 if img is not None:
98 aimg.replaceWith(img)
99
100 return soup