Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-08 10:44:09 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix recipe for The Atlantic

Browse Source
This commit is contained in:
Kovid Goyal 2009-10-19 10:31:45 -06:00
parent 1bf25f3235
commit eddf7201af
1 changed files with 8 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
resources/recipes/atlantic.recipe
View File

@ -21,9 +21,12 @@ class TheAtlantic(BasicNewsRecipe):
dict(name='div', id=['seealso','storybottom', 'footer', 'ad_banner_top', 'sidebar','articletoolstop','subcontent',]), dict(name='div', id=['seealso','storybottom', 'footer', 'ad_banner_top', 'sidebar','articletoolstop','subcontent',]),
dict(name='p', attrs={'id':["pagination"]}), dict(name='p', attrs={'id':["pagination"]}),
dict(name='table',attrs={'class':"tools"}), dict(name='table',attrs={'class':"tools"}),
dict(name='style'),
dict(name='a', href='/a/newsletters.mhtml') dict(name='a', href='/a/newsletters.mhtml')
] ]
remove_attributes = ['icap', 'callout', 'style']
no_stylesheets = True no_stylesheets = True
conversion_options = { 'linearize_tables':True }
extra_css = ''' extra_css = '''
#timestamp{font-family:Arial,Helvetica,sans-serif; color:#666666 ;font-size:x-small} #timestamp{font-family:Arial,Helvetica,sans-serif; color:#666666 ;font-size:x-small}
@ -50,10 +53,14 @@ class TheAtlantic(BasicNewsRecipe):
for item in soup.findAll('div', attrs={'class':'item'}): for item in soup.findAll('div', attrs={'class':'item'}):
a = item.find('a') a = item.find('a')
if a and a.has_key('href'): if a and a.has_key('href'):
url = a['href']#.replace('/doc', 'doc/print') url = a['href']
if not url.startswith('http://'): if not url.startswith('http://'):
url = 'http://www.theatlantic.com/'+url url = 'http://www.theatlantic.com/'+url
url = url.replace('/doc/', '/doc/print/')
title = self.tag_to_string(a) title = self.tag_to_string(a)
if title in ('VIDEO', 'AUDIO', 'INTERACTIVE MAP', 'SIDEBAR', 'RECIPES'):
continue
title = title.replace('&', '&')
byline = item.find(attrs={'class':'byline'}) byline = item.find(attrs={'class':'byline'})
date = self.tag_to_string(byline) if byline else '' date = self.tag_to_string(byline) if byline else ''
description = '' description = ''