Fix #915555 (New recipe for news portal Novi list)

This commit is contained in:
Kovid Goyal 2012-01-13 09:00:45 +05:30
parent 57fe40fe16
commit 05bfde3f25
2 changed files with 49 additions and 0 deletions

Binary file not shown.

After

Width:  |  Height:  |  Size: 944 B

View File

@ -0,0 +1,49 @@
__license__ = 'GPL v3'
__copyright__ = '2012, Darko Miletic <darko.miletic at gmail.com>'
'''
www.novilist.hr
'''
import re
from calibre.web.feeds.news import BasicNewsRecipe
class NoviList_Portal_hr(BasicNewsRecipe):
title = 'Novi List - online portal'
__author__ = 'Darko Miletic'
description = 'Portal Novog Lista'
publisher = 'NOVI LIST d.d.'
category = 'Novi list, politika, hrvatski dnevnik, Novine, Hrvatska, Croatia, News, newspaper, Hrvatski,Primorje, dnevni list, Rijeka'
oldest_article = 2
max_articles_per_feed = 200
no_stylesheets = True
encoding = 'utf8'
use_embedded_content = False
language = 'hr'
publication_type = 'newsportal'
masthead_url = 'http://www.novilist.hr/design/novilist/images/logo-print.gif'
extra_css = """
body{font-family: Geneva,Arial,Helvetica,Swiss,sans-serif }
h1{font-family: Georgia,serif}
img{display:block; margin-bottom: 0.4em; margin-top: 0.4em}
"""
preprocess_regexps = [(re.compile(u'\u0110'), lambda match: u'\u00D0')]
conversion_options = {
'comment' : description
, 'tags' : category
, 'publisher' : publisher
, 'language' : language
, 'linearize_tables' : True
}
keep_only_tags = [dict(name='div', attrs={'id':'content'})]
remove_tags = [dict(name=['meta', 'link', 'iframe', 'embed', 'object'])]
remove_attributes=['border', 'lang']
feeds = [(u'Vijesti', u'http://www.novilist.hr/rss/feed/sve.xml')]
def print_version(self, url):
return url.replace('http://www.novilist.hr/','http://www.novilist.hr/layout/set/print/')