mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-09-29 15:31:08 -04:00
6f93d75c06
(2) Added new recipe which scrapes Bloomberg for columnists' blog postings (3) Revisions to existing economist recipe, added description metadata via populate_article_metadata.
279 lines
11 KiB
Python
279 lines
11 KiB
Python
#!/usr/bin/env python2
|
|
|
|
__license__ = 'GPL v3'
|
|
__copyright__ = '2008, Kovid Goyal <kovid at kovidgoyal.net>'
|
|
'''
|
|
economist.com
|
|
'''
|
|
import cookielib
|
|
from collections import OrderedDict
|
|
|
|
from calibre.ebooks.BeautifulSoup import NavigableString, Tag
|
|
from calibre.web.feeds.news import BasicNewsRecipe
|
|
from calibre.utils.cleantext import clean_ascii_chars
|
|
|
|
|
|
def classes(classes):
|
|
q = frozenset(classes.split(' '))
|
|
return dict(attrs={
|
|
'class': lambda x: x and frozenset(x.split()).intersection(q)})
|
|
|
|
|
|
class NoArticles(Exception):
|
|
pass
|
|
|
|
|
|
def process_url(url):
|
|
if url.startswith('/'):
|
|
url = 'https://www.economist.com' + url
|
|
return url
|
|
|
|
|
|
class Economist(BasicNewsRecipe):
|
|
|
|
title = 'The Economist'
|
|
language = 'en'
|
|
|
|
__author__ = "Kovid Goyal"
|
|
INDEX = 'https://www.economist.com/printedition'
|
|
description = (
|
|
'Global news and current affairs from a European'
|
|
' perspective. Best downloaded on Friday mornings (GMT)'
|
|
)
|
|
extra_css = '''
|
|
.headline {font-size: x-large;}
|
|
h2 { font-size: small; }
|
|
h1 { font-size: medium; }
|
|
em.Bold {font-weight:bold;font-style:normal;}
|
|
em.Italic {font-style:italic;}
|
|
p.xhead {font-weight:bold;}
|
|
.pullquote {
|
|
float: right;
|
|
font-size: larger;
|
|
font-weight: bold;
|
|
font-style: italic;
|
|
page-break-inside:avoid;
|
|
border-bottom: 3px solid black;
|
|
border-top: 3px solid black;
|
|
width: 228px;
|
|
margin: 0px 0px 10px 15px;
|
|
padding: 7px 0px 9px;
|
|
}
|
|
.flytitle-and-title__flytitle {
|
|
display: block;
|
|
font-size: smaller;
|
|
color: red;
|
|
}
|
|
'''
|
|
oldest_article = 7.0
|
|
resolve_internal_links = True
|
|
remove_tags = [
|
|
dict(name=['script', 'noscript', 'title', 'iframe', 'cf_floatingcontent', 'aside', 'footer']),
|
|
dict(attrs={
|
|
'class': [
|
|
'dblClkTrk', 'ec-article-info', 'share_inline_header',
|
|
'related-items', 'main-content-container', 'ec-topic-widget',
|
|
'teaser', 'blog-post__bottom-panel-bottom', 'blog-post__comments-label',
|
|
'blog-post__foot-note', 'blog-post__sharebar', 'blog-post__bottom-panel',
|
|
'newsletter-form','share-links-header','teaser--wrapped', 'latest-updates-panel__container',
|
|
'latest-updates-panel__article-link','blog-post__section'
|
|
]
|
|
}
|
|
),
|
|
dict(attrs={
|
|
'class': lambda x: x and 'blog-post__siblings-list-aside' in x.split()}),
|
|
classes('share-links-header teaser--wrapped latest-updates-panel__container latest-updates-panel__article-link blog-post__section newsletter-form')
|
|
]
|
|
keep_only_tags = [dict(name='article', id=lambda x: not x)]
|
|
no_stylesheets = True
|
|
remove_attributes = ['data-reactid']
|
|
# economist.com has started throttling after about 60% of the total has
|
|
# downloaded with connection reset by peer (104) errors.
|
|
delay = 1
|
|
|
|
needs_subscription = False
|
|
|
|
def __init__(self, *args, **kwargs):
|
|
BasicNewsRecipe.__init__(self, *args, **kwargs)
|
|
if self.output_profile.short_name.startswith('kindle'):
|
|
# Reduce image sizes to get file size below amazon's email
|
|
# sending threshold
|
|
self.web2disk_options.compress_news_images = True
|
|
self.web2disk_options.compress_news_images_auto_size = 5
|
|
self.log.warn('Kindle Output profile being used, reducing image quality to keep file size below amazon email threshold')
|
|
|
|
def get_browser(self):
|
|
br = BasicNewsRecipe.get_browser(self)
|
|
# Add a cookie indicating we have accepted Economist's cookie
|
|
# policy (needed when running from some European countries)
|
|
ck = cookielib.Cookie(
|
|
version=0,
|
|
name='notice_preferences',
|
|
value='2:',
|
|
port=None,
|
|
port_specified=False,
|
|
domain='.economist.com',
|
|
domain_specified=False,
|
|
domain_initial_dot=True,
|
|
path='/',
|
|
path_specified=False,
|
|
secure=False,
|
|
expires=None,
|
|
discard=False,
|
|
comment=None,
|
|
comment_url=None,
|
|
rest={'HttpOnly': None},
|
|
rfc2109=False
|
|
)
|
|
br.cookiejar.set_cookie(ck)
|
|
br.set_handle_gzip(True)
|
|
return br
|
|
|
|
def preprocess_raw_html(self, raw, url):
|
|
import html5lib
|
|
root = html5lib.parse(raw, namespaceHTMLElements=False, treebuilder='lxml')
|
|
from lxml import etree
|
|
for div in root.xpath('//div[@class="lazy-image"]'):
|
|
noscript = list(div.iter('noscript'))
|
|
if noscript and noscript[0].text:
|
|
img = list(html5lib.parse(noscript[0].text, namespaceHTMLElements=False, treebuilder='lxml').iter('img'))
|
|
if img:
|
|
p = noscript[0].getparent()
|
|
idx = p.index(noscript[0])
|
|
p.insert(idx, p.makeelement('img', src=img[0].get('src')))
|
|
p.remove(noscript[0])
|
|
for x in root.xpath('//*[name()="script" or name()="style" or name()="source" or name()="meta"]'):
|
|
x.getparent().remove(x)
|
|
raw = etree.tostring(root, encoding=unicode)
|
|
return raw
|
|
|
|
def populate_article_metadata(self, article, soup, first):
|
|
els = soup.findAll(name=['span', 'p'],
|
|
attrs={'class': ['flytitle-and-title__title', 'blog-post__rubric']})
|
|
result = []
|
|
for el in els[0:2]:
|
|
if el is not None:
|
|
for descendant in el.contents:
|
|
if isinstance(descendant, NavigableString):
|
|
result.append(unicode(descendant))
|
|
article.summary = u'. '.join(result).encode('utf-8') + '.'
|
|
article.text_summary = clean_ascii_chars(article.summary)
|
|
|
|
def parse_index(self):
|
|
# return [('Articles', [{'title':'test',
|
|
# 'url':'http://www.economist.com/news/business/21718916-worlds-biggest-software-firm-has-transformed-its-culture-better-getting-cloud'
|
|
# }])]
|
|
raw = self.index_to_soup(self.INDEX, raw=True)
|
|
# with open('/t/raw.html', 'wb') as f:
|
|
# f.write(raw)
|
|
soup = self.index_to_soup(raw)
|
|
ans = self.economist_parse_index(soup)
|
|
if not ans:
|
|
raise NoArticles(
|
|
'Could not find any articles, either the '
|
|
'economist.com server is having trouble and you should '
|
|
'try later or the website format has changed and the '
|
|
'recipe needs to be updated.'
|
|
)
|
|
return ans
|
|
|
|
def economist_parse_index(self, soup):
|
|
img = soup.find(attrs={'srcset': True, 'class': lambda x: x and 'print-edition__cover-widget__image' in x.split()})
|
|
if img is not None:
|
|
for part in img['srcset'].split():
|
|
if part.startswith('/'):
|
|
part = part.replace('200-width', '640-width')
|
|
self.cover_url = 'https://www.economist.com' + part
|
|
self.log('Got cover:', self.cover_url)
|
|
break
|
|
|
|
sections = soup.findAll('div', attrs={'class': 'list__title'})
|
|
if sections:
|
|
feeds = []
|
|
for section in sections:
|
|
articles = []
|
|
secname = self.tag_to_string(section)
|
|
self.log(secname)
|
|
for a in section.findNextSiblings('a', href=True):
|
|
spans = a.findAll('span')
|
|
if len(spans) == 2:
|
|
title = u'{}: {}'.format(*map(self.tag_to_string, spans))
|
|
else:
|
|
title = self.tag_to_string(a)
|
|
articles.append({'title': title, 'url': process_url(a['href'])})
|
|
self.log(' ', title, articles[-1]['url'])
|
|
if articles:
|
|
feeds.append((secname, articles))
|
|
return feeds
|
|
return self.economist_parse_old_index(soup)
|
|
|
|
def economist_parse_old_index(self, soup):
|
|
feeds = OrderedDict()
|
|
for section in soup.findAll(attrs={'class': lambda x: x and 'section' in x}):
|
|
h4 = section.find('h4')
|
|
if h4 is None:
|
|
continue
|
|
section_title = self.tag_to_string(h4).strip()
|
|
if not section_title:
|
|
continue
|
|
self.log('Found section: %s' % section_title)
|
|
articles = []
|
|
subsection = ''
|
|
for node in section.findAll(attrs={'class': 'article'}):
|
|
subsec = node.findPreviousSibling('h5')
|
|
if subsec is not None:
|
|
subsection = self.tag_to_string(subsec)
|
|
prefix = (subsection + ': ') if subsection else ''
|
|
a = node.find('a', href=True)
|
|
if a is not None:
|
|
url = a['href']
|
|
if url.startswith('/'):
|
|
url = 'https://www.economist.com' + url
|
|
url += '/print'
|
|
title = self.tag_to_string(a)
|
|
if title:
|
|
title = prefix + title
|
|
self.log('\tFound article:', title)
|
|
articles.append({
|
|
'title': title,
|
|
'url': url,
|
|
'description': '',
|
|
'date': ''
|
|
})
|
|
|
|
if articles:
|
|
if section_title not in feeds:
|
|
feeds[section_title] = []
|
|
feeds[section_title] += articles
|
|
|
|
ans = [(key, val) for key, val in feeds.iteritems()]
|
|
return ans
|
|
|
|
def eco_find_image_tables(self, soup):
|
|
for x in soup.findAll('table', align=['right', 'center']):
|
|
if len(x.findAll('font')) in (1, 2) and len(x.findAll('img')) == 1:
|
|
yield x
|
|
|
|
def postprocess_html(self, soup, first):
|
|
for img in soup.findAll('img', srcset=True):
|
|
del img['srcset']
|
|
for table in list(self.eco_find_image_tables(soup)):
|
|
caption = table.find('font')
|
|
img = table.find('img')
|
|
div = Tag(soup, 'div')
|
|
div['style'] = 'text-align:left;font-size:70%'
|
|
ns = NavigableString(self.tag_to_string(caption))
|
|
div.insert(0, ns)
|
|
div.insert(1, Tag(soup, 'br'))
|
|
del img['width']
|
|
del img['height']
|
|
img.extract()
|
|
div.insert(2, img)
|
|
table.replaceWith(div)
|
|
return soup
|
|
|
|
def canonicalize_internal_url(self, url, is_link=True):
|
|
if url.endswith('/print'):
|
|
url = url.rpartition('/')[0]
|
|
return BasicNewsRecipe.canonicalize_internal_url(self, url, is_link=is_link)
|