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 recipe for Le Monde

Browse Source
This commit is contained in:
Kovid Goyal 2009-09-14 13:34:05 -06:00
parent 00601f38f0
commit 257c92fa14
1 changed files with 43 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

59
src/calibre/web/feeds/recipes/recipe_le_monde.py
View File

@ -7,13 +7,12 @@ lemonde.fr
''' '''
import re import re
#from datetime import date
from calibre.web.feeds.news import BasicNewsRecipe from calibre.web.feeds.news import BasicNewsRecipe
class LeMonde(BasicNewsRecipe): class LeMonde(BasicNewsRecipe):
title = 'LeMonde.fr' title = 'LeMonde.fr'
__author__ = 'Mathieu Godlewski <mathieu at godlewski.fr>' __author__ = 'Mathieu Godlewski and Sujata Raman'
description = 'Global news in french' description = 'Global news in french'
oldest_article = 3 oldest_article = 3
language = 'fr' language = 'fr'
@ -23,24 +22,27 @@ class LeMonde(BasicNewsRecipe):
remove_javascript = True remove_javascript = True
#cover_url='http://abonnes.lemonde.fr/titresdumonde/'+date.today().strftime("%y%m%d")+'/1.jpg' # cover_url='http://abonnes.lemonde.fr/titresdumonde/'+date.today().strftime("%y%m%d")+'/1.jpg'
extra_css = ''' extra_css = '''
.dateline{color:#666666;font-family:verdana,sans-serif;font-size:xx-small;} .dateline{color:#666666;font-family:verdana,sans-serif;font-size:x-small;}
.author{font-family:verdana,sans-serif;font-size:x-small;color:#222222;}
.articleImage{color:#666666;font-family:verdana,sans-serif;font-size:x-small;}
.mainText{font-family:Georgia,serif;color:#222222;} .mainText{font-family:Georgia,serif;color:#222222;}
.LM_articleText{font-family:Georgia,serif;} .LM_articleText{font-family:Arial,Helvetica,sans-serif;}
.LM_titleZone{font-family:Arial,Helvetica,sans-serif;}
.mainContent{font-family:Georgia,serif;} .mainContent{font-family:Georgia,serif;}
.mainTitle{font-family:Georgia,serif;}
.LM_content{font-family:Georgia,serif;} .LM_content{font-family:Georgia,serif;}
.LM_caption{font-family:Georgia,serif;font-size:-small;}
.LM_imageSource{font-family:Arial,Helvetica,sans-serif;font-size:x-small;color:#666666;}
h1{font-family:Arial,Helvetica,sans-serif;font-size:medium;color:#000000;}
.post{font-family:Arial,Helvetica,sans-serif;}
.mainTitle{font-family:Georgia,serif;}
.content{font-family:Georgia,serif;} .content{font-family:Georgia,serif;}
.LM_caption{font-family:Georgia,serif;font-size:xx-small;} .entry{font-family:Georgia,serif;}
.LM_imageSource{font-family:Arial,Helvetica,sans-serif;font-size:xx-small;color:#666666;} h2{font-family:Arial,Helvetica,sans-serif;font-size:large;}
h1{font-family:Georgia,serif;font-size:medium;color:#000000;} small{font-family:Arial,Helvetica,sans-serif; color:#ED1B23;}
.entry{font-family:Georgia,Times New Roman,Times,serif;}
.mainTitle{font-family:Georgia,Times New Roman,Times,serif;}
h2{font-family:Georgia,Times New Roman,Times,serif;font-size:large;}
small{{font-family:Arial,Helvetica,sans-serif;font-size:xx-small;}
''' '''
feeds = [ feeds = [
@ -71,11 +73,10 @@ class LeMonde(BasicNewsRecipe):
dict(name='iframe', attrs={}), dict(name='iframe', attrs={}),
dict(name='table', attrs={'id':["toolBox"]}), dict(name='table', attrs={'id':["toolBox"]}),
dict(name='table', attrs={'class':["bottomToolBox"]}), dict(name='table', attrs={'class':["bottomToolBox"]}),
dict(name='div', attrs={'class':["pageNavigation","fenetreBoxesContainer","breakingNews","LM_toolsBottom","LM_comments","LM_tools","pave_meme_sujet_hidden","boxMemeSujet"]}), dict(name='div', attrs={'class':["pageNavigation","LM_pagination","fenetreBoxesContainer","breakingNews","LM_toolsBottom","LM_comments","LM_tools","pave_meme_sujet_hidden","boxMemeSujet"]}),
dict(name='div', attrs={'id':["miniUne","LM_sideBar"]}), dict(name='div', attrs={'id':["miniUne","LM_sideBar"]}),
] ]
preprocess_regexps = [ (re.compile(i[0], re.IGNORECASE|re.DOTALL), i[1]) for i in preprocess_regexps = [ (re.compile(i[0], re.IGNORECASE|re.DOTALL), i[1]) for i in
[ [
(r'<html.*(<div class="post".*?>.*?</div>.*?<div class="entry">.*?</div>).*You can start editing here.*</html>', lambda match : '<html><body>'+match.group(1)+'</body></html>'), (r'<html.*(<div class="post".*?>.*?</div>.*?<div class="entry">.*?</div>).*You can start editing here.*</html>', lambda match : '<html><body>'+match.group(1)+'</body></html>'),
@ -101,6 +102,16 @@ class LeMonde(BasicNewsRecipe):
# Used to filter duplicated articles # Used to filter duplicated articles
articles_list = [] articles_list = []
def get_cover_url(self):
cover_url = None
soup = self.index_to_soup('http://www.lemonde.fr/web/monde_pdf/0,33-0,1-0,0.html')
link_item = soup.find('div',attrs={'class':'pg-gch'})
if link_item and link_item.img:
cover_url = link_item.img['src']
return cover_url
def get_article_url(self, article): def get_article_url(self, article):
url=article.get('link', None) url=article.get('link', None)
url=url[0:url.find("#")] url=url[0:url.find("#")]
@ -109,8 +120,13 @@ class LeMonde(BasicNewsRecipe):
return False return False
if self.is_article_wanted(url): if self.is_article_wanted(url):
self.articles_list.append(url) self.articles_list.append(url)
return url if '/portfolio/' in url or '/video/' in url:
url = None
return url
self.log_debug(_('Skipping filtered article: %s')%url) self.log_debug(_('Skipping filtered article: %s')%url)
url = article.get('guid', None)
return False return False
@ -122,4 +138,15 @@ class LeMonde(BasicNewsRecipe):
return False return False
return False return False
def preprocess_html(self, soup):
for item in soup.findAll(style=True):
del item['style']
for item in soup.findAll(face=True):
del item['face']
for tag in soup.findAll(name=['ul','li']):
tag.name = 'div'
return soup