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

Fix #8984 (Updated recipe for Sudney Morning herald)

Browse Source
This commit is contained in:
Kovid Goyal 2011-02-14 11:44:23 -07:00
parent a7b652e8ff
commit bbb6698e6b
1 changed files with 22 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

25
resources/recipes/smh.recipe
View File

@ -1,5 +1,5 @@
__license__ = 'GPL v3' __license__ = 'GPL v3'
__copyright__ = '2010, Darko Miletic <darko.miletic at gmail.com>' __copyright__ = '2010-2011, Darko Miletic <darko.miletic at gmail.com>'
''' '''
smh.com.au smh.com.au
''' '''
@ -22,7 +22,11 @@ class Smh_au(BasicNewsRecipe):
remove_empty_feeds = True remove_empty_feeds = True
masthead_url = 'http://images.smh.com.au/2010/02/02/1087188/smh-620.jpg' masthead_url = 'http://images.smh.com.au/2010/02/02/1087188/smh-620.jpg'
publication_type = 'newspaper' publication_type = 'newspaper'
extra_css = ' h1{font-family: Georgia,"Times New Roman",Times,serif } body{font-family: Arial,Helvetica,sans-serif} .cT-imageLandscape{font-size: x-small} ' extra_css = """
h1{font-family: Georgia,"Times New Roman",Times,serif }
body{font-family: Arial,Helvetica,sans-serif}
.cT-imageLandscape,.cT-imagePortrait{font-size: x-small}
"""
conversion_options = { conversion_options = {
'comment' : description 'comment' : description
@ -38,7 +42,11 @@ class Smh_au(BasicNewsRecipe):
] ]
remove_tags_after = [dict(name='div',attrs={'class':'articleBody'})] remove_tags_after = [dict(name='div',attrs={'class':'articleBody'})]
keep_only_tags = [dict(name='div',attrs={'id':'content'})] keep_only_tags = [dict(name='div',attrs={'id':'content'})]
remove_attributes = ['width','height'] remove_tags = [
dict(attrs={'class':'hidden'}),
dict(name=['link','meta','base','embed','object','iframe'])
]
remove_attributes = ['width','height','lang']
def parse_index(self): def parse_index(self):
articles = [] articles = []
@ -66,3 +74,14 @@ class Smh_au(BasicNewsRecipe):
,'description':description ,'description':description
}) })
return [(self.tag_to_string(soup.find('title')), articles)] return [(self.tag_to_string(soup.find('title')), articles)]
def preprocess_html(self, soup):
for item in soup.findAll(style=True):
del item['style']
for item in soup.findAll('bod'):
item.name = 'div'
for item in soup.findAll('img'):
if not item.has_key('alt'):
item['alt'] = 'image'
return soup