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 #995076 (recipe: mainichi.recipe don't show its article)

Browse Source
This commit is contained in:
Kovid Goyal 2012-05-05 19:20:33 +05:30
parent f01c91b68f
commit d54e468053
1 changed files with 25 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

29
recipes/mainichi.recipe
View File

@ -16,12 +16,12 @@ class MainichiDailyNews(BasicNewsRecipe):
publisher = 'Mainichi Daily News' publisher = 'Mainichi Daily News'
category = 'news, japan' category = 'news, japan'
language = 'ja' language = 'ja'
index = 'http://mainichi.jp/select/'
feeds = [(u'daily news', u'http://mainichi.jp/rss/etc/flash.rss')] remove_javascript = True
masthead_title = u'MAINICHI DAILY NEWS'
remove_tags_before = {'class':"NewsTitle"} remove_tags_before = {'class':"NewsTitle"}
remove_tags = [{'class':"RelatedArticle"}] remove_tags_after = {'class':"NewsBody clr"}
remove_tags_after = {'class':"Credit"}
def parse_feeds(self): def parse_feeds(self):
@ -32,9 +32,30 @@ class MainichiDailyNews(BasicNewsRecipe):
for a,curarticle in enumerate(curfeed.articles): for a,curarticle in enumerate(curfeed.articles):
if re.search(r'pheedo.jp', curarticle.url): if re.search(r'pheedo.jp', curarticle.url):
delList.append(curarticle) delList.append(curarticle)
if re.search(r'rssad.jp', curarticle.url):
delList.append(curarticle)
if len(delList)>0: if len(delList)>0:
for d in delList: for d in delList:
index = curfeed.articles.index(d) index = curfeed.articles.index(d)
curfeed.articles[index:index+1] = [] curfeed.articles[index:index+1] = []
return feeds return feeds
def parse_index(self):
feeds = []
soup = self.index_to_soup(self.index)
topstories = soup.find('ul',attrs={'class':'MaiLink'})
if topstories:
newsarticles = []
for itt in topstories.findAll('li'):
itema = itt.find('a',href=True)
if itema:
newsarticles.append({
'title' :itema.string
,'date' :''
,'url' :itema['href']
,'description':''
})
feeds.append(('latest', newsarticles))
return feeds