calibre/recipes/lwn_weekly.recipe
Kovid Goyal 9c208906b7 ...
2012-10-19 08:21:22 +05:30

144 lines
4.5 KiB
Python

#!/usr/bin/env python
__license__ = 'GPL v3'
__copyright__ = '2011, Davide Cavalca <davide125 at tiscali.it>'
'''
lwn.net
'''
from calibre.web.feeds.news import BasicNewsRecipe
import re
import sys
class WeeklyLWN(BasicNewsRecipe):
title = 'LWN.net Weekly Edition'
description = 'Weekly summary of what has happened in the free software world.'
__author__ = 'Davide Cavalca'
language = 'en'
site_url = u'http://lwn.net'
extra_css = 'pre,code,samp,kbd,tt { font-size: 80% }\nblockquote {margin-left:0 }\n* { color: black }\n'
no_stylesheets = True
remove_javascript = True
cover_url = site_url + '/images/lcorner.png'
#masthead_url = 'http://lwn.net/images/lcorner.png'
publication_type = 'magazine'
remove_tags_before = dict(attrs={'class':'PageHeadline'})
remove_tags_after = dict(attrs={'class':'ArticleText'})
remove_tags = [dict(name=['h2', 'form'])]
preprocess_regexps = [
# Remove the <hr> and "Log in to post comments"
(re.compile(r'<hr [^>]+>\s*\n\s*.*?comments[)]'), lambda m: ''),
]
conversion_options = {
'linearize_tables' : True,
'no_inline_navbars': True,
}
oldest_article = 7.0
needs_subscription = 'optional'
def get_browser(self):
br = BasicNewsRecipe.get_browser()
if self.username is not None and self.password is not None:
br.open('https://lwn.net/login')
br.select_form(name='loginform')
br['Username'] = self.username
br['Password'] = self.password
br.submit()
return br
def print_version(self, url):
# Strip off anchor
url = url.split('#')[0]
# Prepend site_url
if url[0:len(self.site_url)] != self.site_url:
url = self.site_url + url
# Append printable URL parameter
print_param = '?format=printable'
if url[-len(print_param):] != print_param:
url += print_param
return url
def parse_index(self):
if self.username is not None and self.password is not None:
index_url = self.print_version('/current/bigpage')
else:
index_url = self.print_version('/free/bigpage')
soup = self.index_to_soup(index_url)
curr = soup.body
articles = {}
ans = []
section = soup.title.string
subsection = None
while True:
curr = curr.findNext(attrs = {'class': ['SummaryHL', 'Cat1HL', 'Cat2HL'] })
if curr == None:
break
text = curr.contents[0].string
if 'Cat2HL' in curr.attrMap['class']:
subsection = text
elif 'Cat1HL' in curr.attrMap['class']:
section = text
subsection = None
elif 'SummaryHL' in curr.attrMap['class']:
article_title = text
if subsection:
section_title = "%s: %s" % (section, subsection)
else:
section_title = section
# Most articles have anchors in their titles, *except* the security vulnerabilities
article_anchor = curr.findNext(name = 'a', attrs = { 'href': re.compile('^/Articles/') } )
if article_anchor:
article_url = article_anchor.get('href')
if not article_url:
print >>sys.stderr, 'article_url is None for article_anchor "%s": "%s"' \
% (str(article_anchor), article_title)
continue
else:
print >>sys.stderr, 'article_anchor is None for "%s"; skipping' % article_title
article_url = None
continue
if section_title not in articles:
articles[section_title] = []
if section_title not in ans:
ans.append(section_title)
articles[section_title].append({
'url': article_url,
'title': article_title,
'description': '', 'content': '', 'date': '',
})
else:
print >>sys.stderr, "lwn_weekly.recipe: something bad happened; should not be able to reach this"
ans = [(section, articles[section]) for section in ans if section in articles]
#from pprint import pprint
#pprint(ans)
return ans
# vim: expandtab:ts=4:sw=4