calibre/recipes/foreignaffairs.recipe
Kovid Goyal 553912d235 ...
2016-11-15 20:33:57 +05:30

123 lines
4.4 KiB
Python

#!/usr/bin/env python2
from calibre.web.feeds.news import BasicNewsRecipe
import re
import html5lib
from lxml import html
def select_form(form):
return form.attrs.get('id', None) == 'user-login'
class ForeignAffairsRecipe(BasicNewsRecipe):
''' there are three modifications:
1) fetch issue cover
2) toggle ignore premium articles
3) extract proper section names, ie. "Comments", "Essay"
by Chen Wei, 2012-02-05
Additional modifications to support rebranded website
by anisotrope, 27 June 2015
'''
__license__ = 'GPL v3'
__author__ = 'Rick Shang, kwetal, anisotrope'
language = 'en'
version = 1.02
title = u'Foreign Affairs (Subcription)'
publisher = u'Council on Foreign Relations'
category = u'USA, Foreign Affairs'
description = u'The leading forum for serious discussion of American foreign policy and international affairs.'
no_stylesheets = True
remove_javascript = True
needs_subscription = True
INDEX = 'http://www.foreignaffairs.com'
FRONTPAGE = INDEX + '/magazine'
keep_only_tags = [
dict(attrs={'class': lambda x: x and set(x.split()).intersection(
set('article-header l-article-column'.split()))}),
]
conversion_options = {'comments': description, 'tags': category, 'language': 'en',
'publisher': publisher}
def parse_index(self):
answer = []
soup = self.index_to_soup(html.tostring(self.clean_fa_html(
self.index_to_soup(self.FRONTPAGE, as_tree=True))))
div = soup.find(
'div', attrs={'class': 'magazine-hero__image image_auto_width'})
self.cover_url = div.find('img')['src']
# get dates
date = re.split('\s\|\s', self.tag_to_string(
soup.head.title.string))[0]
self.title = "Foreign Affairs ({})".format(date)
self.timefmt = u' [%s]' % date
for section in soup.findAll(attrs={'class':lambda x: x and 'magazine-list' in x.split()}):
articles = []
section_title = self.tag_to_string(section.find('h1'))
for h2 in section.findAll('h2'):
a = h2.parent
if a.get('href'):
title = self.tag_to_string(h2)
url = a['href']
atr = a.findNextSibling(attrs={'class':'author'})
author = self.tag_to_string(atr) if atr else ''
desc = a.findNextSibling(attrs={'class': 'deck'})
if desc is not None:
description = self.tag_to_string(desc)
else:
description = ''
articles.append({'title': title, 'url': url,
'description': description, 'author': author})
self.log(title)
self.log('\t' + url)
if articles:
answer.append((section_title, articles))
return answer
def clean_fa_html(self, root):
for svg in tuple(root.iter('{*}svg')):
svg.getparent().remove(svg)
for meta in tuple(root.iter('{*}meta')):
meta.getparent().remove(meta)
return root
def preprocess_raw_html(self, raw_html, url):
root = html5lib.parse(raw_html, treebuilder='lxml',
namespaceHTMLElements=False).getroot()
self.clean_fa_html(root)
return html.tostring(root)
def preprocess_html(self, soup):
for img in soup.findAll('img', attrs={'src': True}):
if not img['src'].startswith('http'):
img['src'] = self.INDEX + img['src']
return soup
def get_browser(self):
br = BasicNewsRecipe.get_browser(self)
if self.username is not None and self.password is not None:
# mechanize fails to parse the html correctly, so use html5lib to
# sanitize the html first
response = br.open(
'https://www.foreignaffairs.com/user?destination=user%3Fop%3Dlo')
root = html5lib.parse(
response.get_data(), treebuilder='lxml', namespaceHTMLElements=False)
response.set_data(html.tostring(root))
br.set_response(response)
br.select_form(predicate=select_form)
br.form['name'] = self.username
br.form['pass'] = self.password
br.submit()
return br