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

Update Business Week Magazine

Browse Source
This commit is contained in:
Kovid Goyal 2012-09-02 09:12:20 +05:30
parent def42a1da3
commit 0e6cce50e6
1 changed files with 17 additions and 19 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
recipes/bwmagazine2.recipe
View File

@ -33,38 +33,36 @@ class BusinessWeekMagazine(BasicNewsRecipe):
div0 = soup.find ('div', attrs={'class':'column left'}) div0 = soup.find ('div', attrs={'class':'column left'})
section_title = '' section_title = ''
feeds = OrderedDict() feeds = OrderedDict()
articles = [] for div in div0.findAll('h4'):
for div in div0.findAll('a'): articles = []
section_title = self.tag_to_string(div.findPrevious('h3')).strip() section_title = self.tag_to_string(div.findPrevious('h3')).strip()
self.log('Processing section:', section_title) title=self.tag_to_string(div.a).strip()
title=self.tag_to_string(div).strip() url=div.a['href']
url=div['href']
soup0 = self.index_to_soup(url) soup0 = self.index_to_soup(url)
urlprint=soup0.find('li', attrs={'class':'print'}).a['href'] urlprint=soup0.find('li', attrs={'class':'print'}).a['href']
articles.append({'title':title, 'url':urlprint, 'description':'', 'date':''}) articles.append({'title':title, 'url':urlprint, 'description':'', 'date':''})
if articles: if articles:
if section_title not in feeds: if section_title not in feeds:
feeds[section_title] = [] feeds[section_title] = []
feeds[section_title] += articles feeds[section_title] += articles
div1 = soup.find ('div', attrs={'class':'column center'}) div1 = soup.find ('div', attrs={'class':'column center'})
section_title = '' section_title = ''
articles = [] for div in div1.findAll('h5'):
for div in div1.findAll('a'): articles = []
desc=self.tag_to_string(div.findNext('p')).strip() desc=self.tag_to_string(div.findNext('p')).strip()
section_title = self.tag_to_string(div.findPrevious('h3')).strip() section_title = self.tag_to_string(div.findPrevious('h3')).strip()
self.log('Processing section:', section_title) title=self.tag_to_string(div.a).strip()
title=self.tag_to_string(div).strip() url=div.a['href']
url=div['href']
soup0 = self.index_to_soup(url) soup0 = self.index_to_soup(url)
urlprint=soup0.find('li', attrs={'class':'print'}).a['href'] urlprint=soup0.find('li', attrs={'class':'print'}).a['href']
articles.append({'title':title, 'url':urlprint, 'description':desc, 'date':''}) articles.append({'title':title, 'url':urlprint, 'description':desc, 'date':''})
if articles: if articles:
if section_title not in feeds: if section_title not in feeds:
feeds[section_title] = [] feeds[section_title] = []
feeds[section_title] += articles feeds[section_title] += articles
ans = [(key, val) for key, val in feeds.iteritems()] ans = [(key, val) for key, val in feeds.iteritems()]
return ans return ans