mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-09 03:04:10 -04:00
...
This commit is contained in:
parent
4be922103a
commit
4fdb7163b6
@ -94,14 +94,10 @@ class Arcamax(BasicNewsRecipe):
|
|||||||
page_soup = self.index_to_soup(url)
|
page_soup = self.index_to_soup(url)
|
||||||
if page_soup:
|
if page_soup:
|
||||||
title = page_soup.find(name='div', attrs={'class':'comics-header'}).h1.contents[0]
|
title = page_soup.find(name='div', attrs={'class':'comics-header'}).h1.contents[0]
|
||||||
print 'title is: ', title
|
|
||||||
page_url = url
|
page_url = url
|
||||||
print 'url is: ', url
|
|
||||||
# orig prev_page_url = 'http://www.arcamax.com' + page_soup.find('a', attrs={'class':'prev'}, text='Previous').parent['href']
|
# orig prev_page_url = 'http://www.arcamax.com' + page_soup.find('a', attrs={'class':'prev'}, text='Previous').parent['href']
|
||||||
prev_page_url = 'http://www.arcamax.com' + page_soup.find('span', text='Previous').parent.parent['href']
|
prev_page_url = 'http://www.arcamax.com' + page_soup.find('span', text='Previous').parent.parent['href']
|
||||||
print 'prev_page_url is: ', prev_page_url
|
|
||||||
date = self.tag_to_string(page_soup.find(name='b', attrs={'class':['current']}))
|
date = self.tag_to_string(page_soup.find(name='b', attrs={'class':['current']}))
|
||||||
print 'date is: ', date
|
|
||||||
current_articles.append({'title': title, 'url': page_url, 'description':'', 'date': date})
|
current_articles.append({'title': title, 'url': page_url, 'description':'', 'date': date})
|
||||||
url = prev_page_url
|
url = prev_page_url
|
||||||
current_articles.reverse()
|
current_articles.reverse()
|
||||||
|
Loading…
x
Reference in New Issue
Block a user