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 Nikkei News

Browse Source
This commit is contained in:
Kovid Goyal 2014-11-18 14:28:54 +05:30
parent 10773ceb24
commit f28a295362
1 changed files with 22 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
recipes/nikkei_news.recipe
View File

@ -26,6 +26,8 @@ class NikkeiNet_paper_subscription(BasicNewsRecipe):
# {'class':"cmnc-zoom"}, # {'class':"cmnc-zoom"},
{'class':"cmn-hide"}, {'class':"cmn-hide"},
{'name':'form'}, {'name':'form'},
{'class':'cmn-print_headline cmn-clearfix'},
{'id':'ABOUT_NIKKEI'},
] ]
remove_tags_after = {'class':"cmn-indent"} remove_tags_after = {'class':"cmn-indent"}
@ -85,7 +87,7 @@ class NikkeiNet_paper_subscription(BasicNewsRecipe):
sect_title = sect.find('h3', 'cmnc-title').string sect_title = sect.find('h3', 'cmnc-title').string
sect_result = [] sect_result = []
for elem in sect.findAll(attrs={'class':['cmn-article_title']}): for elem in sect.findAll(attrs={'class':['cmn-article_title']}):
if elem.span.a == None or elem.span.a['href'].startswith('javascript') : if elem.span.a is None or elem.span.a['href'].startswith('javascript') :
continue continue
url = 'http://www.nikkei.com' + elem.span.a['href'] url = 'http://www.nikkei.com' + elem.span.a['href']
url = re.sub("/article/", "/print-article/", url) # print version. url = re.sub("/article/", "/print-article/", url) # print version.
@ -97,3 +99,8 @@ class NikkeiNet_paper_subscription(BasicNewsRecipe):
result.append([sect_title, sect_result]) result.append([sect_title, sect_result])
return result return result
def populate_article_metadata(self, article, soup, first):
elm = soup.find('div', {"class":"cmn-article_text JSID_key_fonttxt"})
elm_text = ''.join([s.string for s in elm])
article.summary = elm_text
article.text_summary = elm_text