Sync to trunk.

This commit is contained in:
John Schember 2011-09-04 11:18:13 -04:00
commit c3a100fc59
2 changed files with 9 additions and 37 deletions

View File

@ -1,40 +1,10 @@
import re
from lxml.html import parse
from calibre.web.feeds.news import BasicNewsRecipe
class Counterpunch(BasicNewsRecipe):
'''
Parses counterpunch.com for articles
'''
title = 'Counterpunch'
description = 'Daily political opinion from www.Counterpunch.com'
language = 'en'
__author__ = 'O. Emmerson'
keep_only_tags = [dict(name='td', attrs={'width': '522'})]
max_articles_per_feed = 10
title = u'Counterpunch'
oldest_article = 7
max_articles_per_feed = 100
auto_cleanup = True
def parse_index(self):
feeds = []
title, url = 'Counterpunch', 'http://www.counterpunch.com'
articles = self.parse_page(url)
if articles:
feeds.append((title, articles))
return feeds
def parse_page(self, url):
parsed_page = parse(url).getroot()
articles = []
unwanted_text = re.compile('Website\ of\ the|I\ urge\ you|Subscribe\ now|DONATE|\@asis\.com|donation\ button|click\ over\ to\ our')
parsed_articles = [a for a in parsed_page.cssselect("html>body>table tr>td>p[class='style2']") if not unwanted_text.search(a.text_content())]
for art in parsed_articles:
try:
author = art.text
title = art.cssselect("a")[0].text + ' by {0}'.format(author)
art_url = 'http://www.counterpunch.com/' + art.cssselect("a")[0].attrib['href']
articles.append({'title': title, 'url': art_url})
except Exception as e:
e
#print('Handler Error: ', e, 'title :', a.text_content())
pass
return articles
feeds = [(u'Counterpunch', u'http://www.counterpunch.org/category/article/feed/')]

View File

@ -1227,7 +1227,9 @@ class IdentifiersEdit(QLineEdit): # {{{
val[k] = v
ids = sorted(val.iteritems(), key=keygen)
txt = ', '.join(['%s:%s'%(k.lower(), v) for k, v in ids])
self.setText(txt.strip())
# Use clear + insert instead of setText so that undo works
self.clear()
self.insert(txt.strip())
self.setCursorPosition(0)
return property(fget=fget, fset=fset)
@ -1319,7 +1321,7 @@ class ISBNDialog(QDialog) : # {{{
self.line_edit.setStyleSheet('QLineEdit { background-color: %s }'%col)
def text(self):
return unicode(self.line_edit.text())
return check_isbn(unicode(self.line_edit.text()))
# }}}