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 #3655 (Fixed recipe for Honolulu Advertiser)

Browse Source
This commit is contained in:
Kovid Goyal 2009-10-01 11:02:37 -06:00
parent 407a64c7e9
commit e9bb0ee6af
1 changed files with 55 additions and 60 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

115
src/calibre/web/feeds/recipes/recipe_honoluluadvertiser.py
View File

@ -1,60 +1,55 @@
#!/usr/bin/env python #!/usr/bin/env python
__license__ = 'GPL v3' __license__ = 'GPL v3'
__copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>' __copyright__ = '2009, Darko Miletic <darko.miletic at gmail.com>'
''' '''
honoluluadvertiser.com honoluluadvertiser.com
''' '''
from calibre.web.feeds.news import BasicNewsRecipe from calibre.web.feeds.news import BasicNewsRecipe
class Honoluluadvertiser(BasicNewsRecipe): class Honoluluadvertiser(BasicNewsRecipe):
title = 'Honolulu Advertiser' title = 'Honolulu Advertiser'
__author__ = 'Darko Miletic' __author__ = 'Darko Miletic'
description = "Latest national and local Hawaii sports news from The Honolulu Advertiser." description = "Latest national and local Hawaii sports news from The Honolulu Advertiser."
publisher = 'Honolulu Advertiser' publisher = 'Honolulu Advertiser'
category = 'news, Honolulu, Hawaii' category = 'news, Honolulu, Hawaii'
oldest_article = 2 oldest_article = 2
language = 'en' language = 'en'
max_articles_per_feed = 100
max_articles_per_feed = 100 no_stylesheets = True
no_stylesheets = True use_embedded_content = False
use_embedded_content = False encoding = 'cp1252'
encoding = 'cp1252'
remove_javascript = True conversion_options = {
cover_url = 'http://www.honoluluadvertiser.com/graphics/branding.gif' 'comments' : description
,'tags' : category
html2lrf_options = [ ,'language' : language
'--comment' , description ,'publisher' : publisher
, '--category' , category }
, '--publisher' , publisher
] keep_only_tags = [dict(name='td')]
html2epub_options = 'publisher="' + publisher + '"\ncomments="' + description + '"\ntags="' + category + '"' remove_tags = [dict(name=['object','link'])]
remove_attributes = ['style']
keep_only_tags = [dict(name='td')]
feeds = [
remove_tags = [dict(name=['object','link'])] (u'Breaking news', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS01&MIME=XML' )
,(u'Local news', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS02&MIME=XML' )
,(u'Sports', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS03&MIME=XML' )
feeds = [ ,(u'Island life', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS05&MIME=XML' )
(u'Breaking news', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS01&MIME=XML' ) ,(u'Entertainment', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS06&MIME=XML' )
,(u'Local news', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS02&MIME=XML' ) ,(u'Business', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS04&MIME=XML' )
,(u'Sports', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS03&MIME=XML' ) ]
,(u'Island life', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS05&MIME=XML' )
,(u'Entertainment', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS06&MIME=XML' ) def preprocess_html(self, soup):
,(u'Business', u'http://www.honoluluadvertiser.com/apps/pbcs.dll/section?Category=RSS04&MIME=XML' ) st = soup.find('td')
] if st:
st.name = 'div'
def preprocess_html(self, soup): return soup
for item in soup.findAll(style=True):
del item['style'] def print_version(self, url):
mtag = '\n<meta http-equiv="Content-Language" content="en"/>\n' ubody, sep, rest = url.rpartition('?source')
soup.head.insert(0,mtag) root, sep2, article_id = ubody.partition('/article/')
return soup return u'http://www.honoluluadvertiser.com/apps/pbcs.dll/article?AID=/' + article_id + '&template=printart'
def print_version(self, url):
ubody, sep, rest = url.rpartition('/-1/')
root, sep2, article_id = ubody.partition('/article/')
return u'http://www.honoluluadvertiser.com/apps/pbcs.dll/article?AID=/' + article_id + '&template=printart'