Update b92. Fixes #961211 (Updated recipe for B92)

This commit is contained in:
Kovid Goyal 2012-03-22 09:27:27 +05:30
parent 72134db772
commit ce5375db54
2 changed files with 13 additions and 7 deletions

View File

@ -1,6 +1,6 @@
__license__ = 'GPL v3'
__copyright__ = '2008-2011, Darko Miletic <darko.miletic at gmail.com>'
__copyright__ = '2008-2012, Darko Miletic <darko.miletic at gmail.com>'
'''
b92.net
'''
@ -20,13 +20,13 @@ class B92(BasicNewsRecipe):
encoding = 'cp1250'
language = 'sr'
publication_type = 'newsportal'
masthead_url = 'http://www.b92.net/images/fp/logo.gif'
masthead_url = 'http://b92s.net/v4/img/new-logo.png'
extra_css = """
@font-face {font-family: "serif1";src:url(res:///opt/sony/ebook/FONT/tt0011m_.ttf)}
@font-face {font-family: "sans1";src:url(res:///opt/sony/ebook/FONT/tt0003m_.ttf)}
body{font-family: Arial,Helvetica,sans1,sans-serif}
.articledescription{font-family: serif1, serif}
.article-info2,.article-info1{text-transform: uppercase; font-size: small}
img{display: block}
.sms{font-weight: bold}
"""
conversion_options = {
@ -37,11 +37,17 @@ class B92(BasicNewsRecipe):
, 'linearize_tables' : True
}
preprocess_regexps = [(re.compile(u'\u0110'), lambda match: u'\u00D0')]
preprocess_regexps = [
(re.compile(u'\u0110'), lambda match: u'\u00D0'),
(re.compile(r'<html.*?<body>', re.DOTALL|re.IGNORECASE), lambda match: '<html><head><title>something</title></head><body>')
]
keep_only_tags = [dict(attrs={'class':['article-info1','article-text']})]
remove_attributes = ['width','height','align','hspace','vspace','border']
remove_tags = [dict(name=['embed','link','base','meta'])]
remove_attributes = ['width','height','align','hspace','vspace','border','lang','xmlns:fb']
remove_tags = [
dict(name=['embed','link','base','meta','iframe'])
,dict(attrs={'id':'social'})
]
feeds = [
(u'Vesti' , u'http://www.b92.net/info/rss/vesti.xml' )

Binary file not shown.

Before

Width:  |  Height:  |  Size: 413 B

After

Width:  |  Height:  |  Size: 1.5 KiB