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

...

Browse Source
This commit is contained in:
Kovid Goyal 2010-04-07 08:00:52 +05:30
parent 2ee2c64e9e
commit 47cee548dc
1 changed files with 5 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
resources/recipes/smh.recipe
View File

@ -5,6 +5,7 @@ smh.com.au
''' '''
from calibre import strftime from calibre import strftime
from calibre.web.feeds.news import BasicNewsRecipe from calibre.web.feeds.news import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import BeautifulSoup
class Smh_au(BasicNewsRecipe): class Smh_au(BasicNewsRecipe):
title = 'The Sydney Morning Herald - Printed edition' title = 'The Sydney Morning Herald - Printed edition'
@ -20,7 +21,7 @@ class Smh_au(BasicNewsRecipe):
language = 'en_AU' language = 'en_AU'
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{font-size: x-small} '
conversion_options = { conversion_options = {
@ -41,11 +42,12 @@ class Smh_au(BasicNewsRecipe):
def parse_index(self): def parse_index(self):
articles = [] articles = []
soup = self.index_to_soup('http://www.smh.com.au/todays-paper') rawc = self.index_to_soup('http://www.smh.com.au/todays-paper',True)
soup = BeautifulSoup(rawc,fromEncoding=self.encoding)
for itimg in soup.findAll('img',src=True): for itimg in soup.findAll('img',src=True):
if itimg['src'].endswith('frontpage.jpg'): if itimg['src'].endswith('frontpage.jpg'):
self.cover_url = itimg['src'] self.cover_url = itimg['src']
for item in soup.findAll(attrs={'class':'cN-storyHeadlineLead cfix'}): for item in soup.findAll(attrs={'class':'cN-storyHeadlineLead cfix'}):
description = '' description = ''
title_prefix = '' title_prefix = ''