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

Merge branch 'hindu-cover' of https://github.com/shivaprsd/calibre

Browse Source
This commit is contained in:
Kovid Goyal 2021-06-13 21:05:56 +05:30
commit 4df402cacf
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
recipes/hindu.recipe
View File

@ -15,12 +15,15 @@ def classes(classes):
class TheHindu(BasicNewsRecipe): class TheHindu(BasicNewsRecipe):
title = u'The Hindu' title = u'The Hindu'
language = 'en_IN' language = 'en_IN'
epaper_url = 'https://epaper.thehindu.com'
oldest_article = 1 oldest_article = 1
__author__ = 'Kovid Goyal' __author__ = 'Kovid Goyal'
max_articles_per_feed = 100 max_articles_per_feed = 100
no_stylesheets = True no_stylesheets = True
remove_attributes = ['style'] remove_attributes = ['style']
extra_css = '.lead-img-cont { text-align: center; } ' \
'.lead-img-caption { font-size: small; font-style: italic; }'
ignore_duplicate_articles = {'title', 'url'} ignore_duplicate_articles = {'title', 'url'}
keep_only_tags = [ keep_only_tags = [
@ -29,6 +32,15 @@ class TheHindu(BasicNewsRecipe):
dict(id=lambda x: x and x.startswith('content-body-')), dict(id=lambda x: x and x.startswith('content-body-')),
] ]
def get_browser(self):
br = BasicNewsRecipe.get_browser(self)
br.addheaders += [('Referer', self.epaper_url)] # needed for fetching cover
return br
def get_cover_url(self):
url = self.index_to_soup(self.epaper_url + '/Login/DefaultImage', raw=True)
return url.replace(br'\\', b'/').decode('utf-8')[1:-1]
def preprocess_html(self, soup): def preprocess_html(self, soup):
img = soup.find('img', attrs={'class': 'lead-img'}) img = soup.find('img', attrs={'class': 'lead-img'})
try: try: