Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Update 1843

Browse Source
This commit is contained in:
Kovid Goyal 2019-02-10 09:30:51 +05:30
parent f9ab883383
commit 2b0811b31a
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 6 additions and 38 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

44
recipes/1843.recipe
View File

@ -20,43 +20,11 @@ class E1843(BasicNewsRecipe):
no_stylesheets = True
remove_javascript = True
encoding = 'utf-8'
keep_only_tags = [
dict(name='h1', attrs={'class': 'title'}),
classes('field-name-field-rubric-summary article-header__overlay-main-image meta-info__author article__body'),
feeds = [
'https://www.1843magazine.com/rss/content',
]
def parse_index(self):
soup = self.index_to_soup('https://www.1843magazine.com')
a = soup.find(text='Print edition').parent
soup = self.index_to_soup(a['href'])
h1 = soup.find(**classes('cover-image__main'))
self.timefmt = ' [%s]' % self.tag_to_string(h1)
img = soup.find(**classes('cover-image__image')).find('img')
self.cover_url = img['src']
ans = []
current_section = articles = None
for div in soup.findAll(**classes('field-name-field-header node-article')):
if 'field-header' in div['class']:
if current_section and articles:
ans.append((current_section, articles))
current_section = self.tag_to_string(div)
self.log(current_section)
articles = []
else:
a = div.find('a', href=True)
title = self.tag_to_string(a)
url = a['href']
self.log('\t', title, ' at ', url)
desc = ''
r = div.find(**classes('article-rubric'))
if r is not None:
desc = self.tag_to_string(r)
articles.append(
{'title': title, 'url': url, 'description': desc})
if current_section and articles:
ans.append((current_section, articles))
return ans
keep_only_tags = [
dict(name='h1', attrs={'class': lambda x: x and 'title' in x.split()}),
classes('field-name-field-rubric-summary article-header__overlay-main-image meta-info__author article__body'),
]