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 Haaretz

Browse Source 
Fixes #1412859 [Updated recipe for Haaretz in English](https://bugs.launchpad.net/calibre/+bug/1412859)
This commit is contained in:
Kovid Goyal 2015-01-21 08:54:21 +05:30
parent c31284d713
commit c7b434dd07
2 changed files with 17 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

41
recipes/haaretz_en.recipe
View File

@ -1,5 +1,5 @@
__license__ = 'GPL v3' __license__ = 'GPL v3'
__copyright__ = '2010-2012, Darko Miletic <darko.miletic at gmail.com>' __copyright__ = '2010-2015, Darko Miletic <darko.miletic at gmail.com>'
''' '''
www.haaretz.com www.haaretz.com
''' '''
@ -31,8 +31,6 @@ class Haaretz_en(BasicNewsRecipe):
.authorBar {font-size: small} .authorBar {font-size: small}
""" """
preprocess_regexps = [(re.compile(r'</body>.*?</html>', re.DOTALL|re.IGNORECASE),lambda match: '</body></html>')]
conversion_options = { conversion_options = {
'comment' : description 'comment' : description
, 'tags' : category , 'tags' : category
@ -40,28 +38,29 @@ class Haaretz_en(BasicNewsRecipe):
, 'language' : language , 'language' : language
} }
keep_only_tags = [dict(attrs={'id':'threecolumns'})] keep_only_tags = [dict(name='div', attrs={'id':'content'})]
remove_attributes = ['width','height'] remove_attributes = ['width','height']
remove_tags = [ remove_tags = [
dict(name=['iframe','link','object','embed']) dict(name=['iframe','link','object','embed'])
,dict(name='div',attrs={'class':'rightcol'}) ,dict(name='div',attrs={'class':['rightcol', 'fblike']})
,dict(name='div',attrs={'id':'article_sso_form'})
] ]
feeds = [ feeds = [
(u'Headlines' , 'http://feeds.feedburner.com/haaretz/LBao' ) (u'Headlines' , 'http://www.haaretz.com/cmlink/1.263335')
,(u'Opinion' , 'http://feeds.feedburner.com/haaretz/opinions' ) ,(u'Opinion' , 'http://www.haaretz.com/cmlink/1.628752')
,(u'Defence and diplomacy' , 'http://feeds.feedburner.com/DefenseAndDiplomacy' ) ,(u'Defence and diplomacy' , 'http://www.haaretz.com/cmlink/1.628763')
,(u'National' , 'http://feeds.feedburner.com/haaretz/National' ) ,(u'National' , 'http://www.haaretz.com/cmlink/1.628764')
,(u'International' , 'http://feeds.feedburner.com/InternationalRss' ) ,(u'International' , 'http://www.haaretz.com/cmlink/1.628765')
,(u'Jewish World' , 'http://feeds.feedburner.com/JewishWorldRss' ) ,(u'Jewish World' , 'http://www.haaretz.com/cmlink/1.628766')
,(u'Business' , 'http://feeds.feedburner.com/BusinessPrintRss' ) ,(u'Business' , 'http://www.haaretz.com/cmlink/1.628767')
,(u'Real Estate' , 'http://feeds.feedburner.com/RealEstatePrintRss' ) ,(u'Real Estate' , 'http://www.haaretz.com/cmlink/1.628768')
,(u'Features' , 'http://feeds.feedburner.com/FeaturesPrintRss' ) ,(u'Features' , 'http://www.haaretz.com/cmlink/1.628769')
,(u'Arts & Leisure' , 'http://feeds.feedburner.com/ArtsAndLeisureRss' ) ,(u'Arts & Leisure' , 'http://www.haaretz.com/cmlink/1.628771')
,(u'Books' , 'http://www.haaretz.com/cmlink/books-rss-1.264947?localLinksEnabled=false') ,(u'Books' , 'http://www.haaretz.com/cmlink/1.628772')
,(u'Food & Wine' , 'http://feeds.feedburner.com/FoodAndWinePrintRss' ) ,(u'Food & Wine' , 'http://www.haaretz.com/cmlink/1.628773')
,(u'Sports' , 'http://feeds.feedburner.com/haaretz/Sport' ) ,(u'Sports' , 'http://www.haaretz.com/cmlink/1.628774')
] ]
def get_browser(self): def get_browser(self):
@ -78,13 +77,7 @@ class Haaretz_en(BasicNewsRecipe):
br.open('https://sso.haaretz.com/sso/sso/signIn',data) br.open('https://sso.haaretz.com/sso/sso/signIn',data)
return br return br
def get_article_url(self, article):
url = BasicNewsRecipe.get_article_url(self, article)
return self.browser.open_novisit(url).geturl()
def print_version(self, url): def print_version(self, url):
article = url.rpartition('/')[2] article = url.rpartition('/')[2]
return 'http://www.haaretz.com/misc/article-print-page/' + article return 'http://www.haaretz.com/misc/article-print-page/' + article
def preprocess_raw_html(self, raw, url):
return '<html><head>'+raw[raw.find('</head>'):]

BIN
recipes/icons/haaretz_en.png
View File

Binary file not shown.
Side by Side Swipe Overlay

Before

Width:  |  Height:  |  Size: 712 B

After

Width:  |  Height:  |  Size: 277 B