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

Updated Heraldo de Aragon

Browse Source
This commit is contained in:
Kovid Goyal 2010-12-27 10:17:27 -07:00
parent 4e2f5ee60e
commit b87d3eb248
1 changed files with 44 additions and 29 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

67
resources/recipes/heraldo.recipe
View File

@ -1,50 +1,65 @@
#!/usr/bin/env python
__license__ = 'GPL v3'
__author__ = 'Lorenzo Vigentini'
__copyright__ = '2009, Lorenzo Vigentini <l.vigentini at gmail.com>'
__copyright__ = '04 December 2010, desUBIKado'
__author__ = 'desUBIKado'
__description__ = 'Daily newspaper from Aragon'
__version__ = 'v1.01'
__date__ = '30, January 2010'
__version__ = 'v0.03'
__date__ = '11, December 2010'
'''
http://www.heraldo.es/
[url]http://www.heraldo.es/[/url]
'''
import time
from calibre.web.feeds.news import BasicNewsRecipe
class heraldo(BasicNewsRecipe):
author = 'Lorenzo Vigentini'
__author__ = 'desUBIKado'
description = 'Daily newspaper from Aragon'
cover_url = 'http://www.heraldo.es/MODULOS/global/publico/interfaces/img/logo.gif'
title = u'Heraldo de Aragon'
publisher = 'OJD Nielsen'
category = 'News, politics, culture, economy, general interest'
language = 'es'
timefmt = '[%a, %d %b, %Y]'
oldest_article = 1
max_articles_per_feed = 25
max_articles_per_feed = 100
use_embedded_content = False
recursion = 10
remove_javascript = True
no_stylesheets = True
keep_only_tags = [
dict(name='div', attrs={'class':['titularNoticiaNN','textoGrisVerdanaContenidos']})
]
recursion = 10
feeds = [
(u'Portadas ', u'http://www.heraldo.es/index.php/mod.portadas/mem.rss')
(u'Portadas', u'http://www.heraldo.es/index.php/mod.portadas/mem.rss')
]
keep_only_tags = [dict(name='div', attrs={'id':['dts','com']})]
remove_tags = [dict(name='a', attrs={'class':['com flo-r','enl-if','enl-df']}),
dict(name='div', attrs={'class':['brb-b-s con marg-btt','cnt-rel con']}),
dict(name='form', attrs={'class':'form'})]
remove_tags_before = dict(name='div' , attrs={'id':'dts'})
remove_tags_after = dict(name='div' , attrs={'id':'com'})
def get_cover_url(self):
cover = None
st = time.localtime()
year = str(st.tm_year)
month = "%.2d" % st.tm_mon
day = "%.2d" % st.tm_mday
#[url]http://oldorigin-www.heraldo.es/20101211/primeras/portada_aragon.pdf[/url]
cover='http://oldorigin-www.heraldo.es/'+ year + month + day +'/primeras/portada_aragon.pdf'
br = BasicNewsRecipe.get_browser()
try:
br.open(cover)
except:
self.log("\nPortada no disponible")
cover ='http://www.heraldo.es/MODULOS/global/publico/interfaces/img/logo-Heraldo.png'
return cover
extra_css = '''
.articledate {color: gray;font-family: monospace;}
.articledescription {display: block;font-family: sans;font-size: 0.7em; text-indent: 0;}
.firma {color: #666;display: block;font-family: verdana, arial, helvetica;font-size: 1em;margin-bottom: 8px;}
.textoGrisVerdanaContenidos {color: #56595c;display: block;font-family: Verdana;font-size: 1.28571em;padding-bottom: 10px}
.titularNoticiaNN {display: block;padding-bottom: 10px;padding-left: 0;padding-right: 0;padding-top: 4px}
.titulo {color: #003066;font-family: Tahoma;font-size: 1.92857em;font-weight: bold;line-height: 1.2em}
h2{font-family:Arial,Helvetica,sans-serif; font-weight:bold;font-size:xx-large;}
'''