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 Le Monde Diplomatique - cono sur

Browse Source 
Fixes #1946593 [Updated recipe for El diplo](https://bugs.launchpad.net/calibre/+bug/1946593)
This commit is contained in:
Kovid Goyal 2021-10-10 19:14:28 +05:30
parent 2447aa0745
commit 65b0bada6f
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 9 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
recipes/el_diplo.recipe
View File

@ -1,9 +1,8 @@
# -*- mode: python; coding: utf-8; -*-
# vim: set syntax=python fileencoding=utf-8
from __future__ import absolute_import, division, print_function, unicode_literals
__license__ = 'GPL v3'
__copyright__ = '2020, Darko Miletic <darko.miletic at gmail.com>'
__copyright__ = '2021, Darko Miletic <darko.miletic at gmail.com>'
'''
www.eldiplo.org
@ -27,6 +26,7 @@ class ElDiplo2020(BasicNewsRecipe):
remove_empty_feeds = True
publication_type = 'magazine'
auto_cleanup = True
auto_cleanup_keep = '//div[contains(@class, "autor")] | //div[@class="edicion"]'
delay = 1
simultaneous_downloads = 1
timeout = 8
@ -38,12 +38,13 @@ class ElDiplo2020(BasicNewsRecipe):
handle_gzip = True
compress_news_images = True
scale_news_images_to_device = True
masthead_url = 'https://www.eldiplo.org/wp-content/themes/eldiplo/img/logo.png'
masthead_url = 'https://www.eldiplo.org/wp-content/themes/_polenta_/assets/diplo.png'
INDEX = 'https://www.eldiplo.org/'
extra_css = """
body{font-family: Roboto, sans-serif}
.entry-title{font-family: Spectral, serif}
body{font-family: "GT Super", serif}
.autor{font-family: Inter, serif}
"""
conversion_options = {
@ -63,11 +64,10 @@ class ElDiplo2020(BasicNewsRecipe):
def parse_index(self):
articles = []
soup = self.index_to_soup(self.INDEX)
mylink = soup.find('a', {'class':'edicion_cover'})
mylink = soup.find('span', text='Sumario')
if mylink is None:
return None
self.cover_url = mylink.img['src']
indexurl = mylink['href']
return None
indexurl = "https://www.eldiplo.org" + mylink.parent['href']
self.log(indexurl)
parts = indexurl.split('www.eldiplo.org/', 1)
series = parts[1].split('-', 1)[0]