calibre/recipes/independent.recipe
2011-12-17 08:57:22 +05:30

514 lines
20 KiB
Plaintext

# adapted from old recipe by Darko Miletic <darko.miletic at gmail.com>
import re
from calibre.web.feeds.recipes import BasicNewsRecipe
from calibre.ebooks.BeautifulSoup import Tag, NavigableString
class TheIndependentNew(BasicNewsRecipe):
# flag to enable/disable article graphics on business pages/some others
# eg http://www.independent.co.uk/news/world/europe/berlusconi-departure-fails-to-calm-the-markets-6259682.html
# -max dimensions can be altered using the .pictureContainer img selector in the css
_FETCH_ARTICLE_GRAPHICS = True
#Flag to enable/disable image fetching (not business)
_FETCH_IMAGES = True
#used for converting rating to stars
_STAR_URL = 'http://www.independent.co.uk/skins/ind/images/rating_star.png'
_NO_STAR_URL = 'http://www.independent.co.uk/skins/ind/images/rating_star_grey.png'
title = u'The Independent'
__author__ = 'Will'
description = 'The latest in UK News and World News from The \
Independent. Wide range of international and local news, sports \
news, commentary and opinion pieces.Independent News - Breaking news \
that matters. Your daily comprehensive news source - The \
Independent Newspaper'
publisher = 'The Independent'
category = 'news, UK'
no_stylesheets = True
use_embedded_content = False
remove_empty_feeds = True
language = 'en_GB'
publication_type = 'newspaper'
masthead_url = 'http://www.independent.co.uk/independent.co.uk/editorial/logo/independent_Masthead.png'
encoding = 'utf-8'
remove_tags =[
dict(attrs={'id' : ['RelatedArtTag','renderBiography']}),
dict(attrs={'class' : ['autoplay','openBiogPopup']}),
dict(name='img',attrs={'alt' : ['Get Adobe Flash player']}),
dict(attrs={'style' : re.compile('.*')}),
]
keep_only_tags =[dict(attrs={'id':'main'})]
recursions = 0
# fixes non compliant html nesting and 'marks' article graphics links
preprocess_regexps = [
(re.compile('<span class="storyTop ">(?P<nested>.*?)</span>', re.DOTALL),
lambda match: '<div class="storyTop">' + match.group('nested') + '</div>'),
(re.compile('(<strong>.*?[Cc]lick.*?<a.*?((HERE)|([Hh]ere)).*?</strong>)', re.DOTALL),
lambda match: '<div class="article-graphic">' + match.group(0) + '</div>'),
]
conversion_options = {
'comment' : description
, 'tags' : category
, 'publisher' : publisher
, 'language' : language
}
extra_css = """
h1{font-family: Georgia,serif }
body{font-family: Verdana,Arial,Helvetica,sans-serif}
img{margin-bottom: 0.4em; display:block}
.starRating img {float: left}
.starRating {margin-top:0.4em; display: block}
.image {clear:left; font-size: x-small; color:#888888;}
.articleByTimeLocation {font-size: x-small; color:#888888;
margin-bottom:0.2em ; margin-top:0.2em ; display:block}
.subtitle {clear:left}
.column-1 h1 { color: #191919}
.column-1 h2 { color: #333333}
.column-1 h3 { color: #444444}
.column-1 p { color: #777777}
.column-1 p,a,h1,h2,h3 { margin: 0; }
.column-1 div{color:#888888; margin: 0;}
.articleContent {display: block; clear:left;}
.storyTop{}
.pictureContainer img { max-width: 400px; max-height: 400px;}
"""
oldest_article = 1
max_articles_per_feed = 100
_processed_urls = []
def get_article_url(self, article):
url = super(self.__class__,self).get_article_url(article)
title = article.get('title', None)
if title and re.search("^Video:",title):
return None
#remove duplicates
if not (url in self._processed_urls):
self._processed_urls.append(url)
else:
url = None
return url
def populate_article_metadata(self, article, soup, first):
if first and hasattr(self, 'add_toc_thumbnail'):
picdiv = soup.find('img')
if picdiv is not None:
self.add_toc_thumbnail(article,picdiv['src'])
def preprocess_html(self, soup):
#remove 'advertorial articles'
strapline = soup.find('div',attrs={'class' : re.compile('.*strapLine.*')})
if strapline:
for para in strapline.findAll('p'):
if len(para.contents) and isinstance(para.contents[0],NavigableString) \
and para.contents[0] == 'ADVERTORIAL FEATURE':
return None
items_to_extract = []
slideshow_elements = []
for item in soup.findAll(attrs={'class' : re.compile("widget.*")}):
remove = True
pattern = re.compile('((articleContent)|(title))$')
if (pattern.search(item['class'])) is not None:
remove = False
# corrections
# story content always good
pattern = re.compile('storyContent')
if (pattern.search(item['class'])) is not None:
remove = False
#images
pattern = re.compile('slideshow')
if (pattern.search(item['class'])) is not None:
if self._FETCH_IMAGES:
remove = False
slideshow_elements.append(item)
else:
remove = True
#social widgets always bad
pattern = re.compile('socialwidget')
if (pattern.search(item['class'])) is not None:
remove = True
if remove:
items_to_extract.append(item)
for item in items_to_extract:
item.extract()
items_to_extract = []
if self._FETCH_IMAGES:
for element in slideshow_elements:
for item in element.findAll('a',attrs={'href' : re.compile('.*')}):
if item.img is not None:
#use full size image
img = item.findNext('img')
img['src'] = item['href']
#insert caption if available
if img.get('title') and (len(img['title']) > 1):
tag = Tag(soup,'h3')
text = NavigableString(img['title'])
tag.insert(0,text)
#picture before text
img.extract()
item.insert(0,img)
item.insert(1,tag)
# remove link
item.name = "div"
item["class"]='image'
del item["href"]
#remove empty subtitles
"""
currently the subtitle is located in first paragraph after
sibling <h3 class="subtitle"> tag. This may be 'fixed' at
some point.
"""
subtitle = soup.find('h3',attrs={'class' : 'subtitle'})
if subtitle is not None:
subtitleText = subtitle.findNext('p')
if subtitleText is not None:
if len(subtitleText.contents[0]) <= 1 :
subtitleText.extract()
subtitle.extract()
#replace rating numbers with stars
for item in soup.findAll('div',attrs={ 'class' : 'starRating'}):
if item is not None:
soup2 = self._insertRatingStars(soup,item)
if soup2 is not None:
soup = soup2
#remove empty paragraph tags in storyTop which can leave a space
#between first paragraph and rest of story
nested_content = False
storyTop = soup.find('div',attrs={ 'class' : ['storyTop']})
for item in storyTop.findAll('p'):
for nested in item:
if isinstance(nested, Tag):
nested_content = True
break
if not nested_content and item.contents is not None and len(item.contents[0]) <= 1 :
items_to_extract.append(item)
for item in items_to_extract:
item.extract()
items_to_extract = []
#remove line breaks immediately next to tags with default margins
#to prevent double line spacing and narrow columns of text
storyTop = soup.find('div',attrs={ 'class' : ['storyTop']})
self._remove_undesired_line_breaks_from_tag(storyTop,soup)
#replace article graphics link with the graphics themselves
if self._FETCH_ARTICLE_GRAPHICS:
items_to_insert = []
for item in soup.findAll('div', attrs={'class' : ['article-graphic']}):
strong = item.find('strong')
if not strong:
continue
for child in strong:
if isinstance(child,Tag):
if str(child.name) == 'a':
items_to_insert.extend(self._get_article_graphic(strong,child['href'],soup))
for item in items_to_insert:
item[0].replaceWith(item[1])
for item in items_to_extract:
item.extract()
return soup
def _get_article_graphic(self,old_item,url,soup):
items_to_insert = []
if re.search('\.jpg$',str(url)):
div = Tag(soup,'div')
div['class'] = 'pictureContainer'
img = Tag(soup,'img')
img['src'] = url
img['alt'] = 'article graphic'
div.insert(0,img)
items_to_insert.append((old_item,div,))
return items_to_insert
soup2 = self.index_to_soup(url)
for item in soup2.findAll('div',attrs={'class' : re.compile("widget picture article.*")}):
items_to_insert.append((old_item,item),)
return items_to_insert
def _insertRatingStars(self,soup,item):
if item.contents is None or len(item.contents) < 1:
return
rating = item.contents[0]
try:
rating = float(item.contents[0])
except:
print 'Could not convert decimal rating to star: malformatted float.'
return
for i in range(1,6):
star = Tag(soup,'img')
if i <= rating:
star['src'] = self._STAR_URL
else:
star['src'] = self._NO_STAR_URL
star['alt'] = 'star number ' + str(i)
item.insert(i,star)
#item.contents[0] = NavigableString('(' + str(rating) + ')')
item.contents[0] = ''
def postprocess_html(self,soup, first_fetch):
#find broken images and remove captions
items_to_extract = []
for item in soup.findAll('div', attrs={'class' : 'image'}):
img = item.findNext('img')
if img and img.get('src'):
# broken images still point to remote url
pattern = re.compile('http://www.independent.co.uk.*')
if pattern.match(img["src"]) is not None:
caption = img.findNextSibling('h3')
if caption is not None:
items_to_extract.append(caption)
items_to_extract.append(img)
for item in items_to_extract:
item.extract()
return soup
def _recurisvely_linearise_tag_tree(
self,
item,
linearised= None,
count=0,
limit = 100
):
linearised = linearised or []
count = count + 1
if count > limit:
return linearised
if not (isinstance(item,Tag)):
return linearised
for nested in item:
linearised.append(nested)
linearised = self._recurisvely_linearise_tag_tree(nested,linearised, count)
return linearised
def _get_previous_tag(self,current_index, tag_tree):
if current_index == 0:
return None
else:
return tag_tree[current_index - 1]
def _get_next_tag(self,current_index, tag_tree):
if current_index < len(tag_tree) - 1:
return tag_tree[current_index + 1]
else:
return None
def _list_match(self,test_str, list_regex):
for regex in list_regex:
match = re.match(regex, test_str)
if match is not None:
return True
return False
def _remove_undesired_line_breaks_from_tag(self,parent,soup):
if parent is None:
return
tag_tree = self._recurisvely_linearise_tag_tree(parent)
items_to_remove = []
for item in tag_tree:
if item == u'\n':
items_to_remove.append(item)
continue;
for item in items_to_remove:
tag_tree.remove(item)
spaced_tags = [r'p', r'h\d', r'blockquote']
tags_to_extract = []
tags_to_replace = []
for (i, tag) in enumerate(tag_tree):
if isinstance(tag, Tag):
if str(tag) == '<br />':
previous_tag = self._get_previous_tag(i, tag_tree)
if isinstance(previous_tag, Tag):
previous_tag_is_spaced = previous_tag is not None\
and self._list_match(str(previous_tag.name),
spaced_tags)
else:
previous_tag_is_spaced = False
next_tag = self._get_next_tag(i, tag_tree)
if isinstance(next_tag, Tag):
next_tag_is_spaced = next_tag is not None\
and self._list_match(str(next_tag.name), spaced_tags)
else:
next_tag_is_spaced = False
if previous_tag_is_spaced or next_tag_is_spaced or i == 0\
or i == len(tag_tree) - 1:
tags_to_extract.append(tag)
else:
tags_to_replace.append((tag,NavigableString(' '),))
for pair in tags_to_replace:
pair[0].replaceWith(pair[1])
for tag in tags_to_extract:
tag.extract()
feeds = [
(u'News - UK',
u'http://www.independent.co.uk/news/uk/?service=rss'),
(u'News - World',
u'http://www.independent.co.uk/news/world/?service=rss'),
(u'News - Business',
u'http://www.independent.co.uk/news/business/?service=rss'),
(u'News - People',
u'http://www.independent.co.uk/news/people/?service=rss'),
(u'News - Science',
u'http://www.independent.co.uk/news/science/?service=rss'),
(u'News - Media',
u'http://www.independent.co.uk/news/media/?service=rss'),
(u'News - Education',
u'http://www.independent.co.uk/news/education/?service=rss'),
(u'News - Obituaries',
u'http://www.independent.co.uk/news/obituaries/?service=rss'),
(u'News - Corrections',
u'http://www.independent.co.uk/news/corrections/?service=rss'
),
(u'Opinion',
u'http://www.independent.co.uk/opinion/?service=rss'),
(u'Environment',
u'http://www.independent.co.uk/environment/?service=rss'),
(u'Sport - Athletics',
u'http://www.independent.co.uk/sport/general/athletics/?service=rss'
),
(u'Sport - Cricket',
u'http://www.independent.co.uk/sport/cricket/?service=rss'),
(u'Sport - Football',
u'http://www.independent.co.uk/sport/football/?service=rss'),
(u'Sport - Golf',
u'http://www.independent.co.uk/sport/golf/?service=rss'),
(u'Sport - Motor racing',
u'http://www.independent.co.uk/sport/motor-racing/?service=rss'
),
(u'Sport - Olympics',
u'http://www.independent.co.uk/sport/olympics/?service=rss'),
(u'Sport - Racing',
u'http://www.independent.co.uk/sport/racing/?service=rss'),
(u'Sport - Rugby League',
u'http://www.independent.co.uk/sport/general/rugby-league/?service=rss'),
(u'Sport - Rugby Union',
u'http://www.independent.co.uk/sport/rugby/rugby-union/?service=rss'
),
(u'Sport - Sailing',
u'http://www.independent.co.uk/sport/general/sailing/?service=rss'
),
(u'Sport - Tennis',
u'http://www.independent.co.uk/sport/tennis/?service=rss'),
(u'Sport - Others',
u'http://www.independent.co.uk/sport/general/others/?service=rss'
),
(u'Life & Style - Fashion',
u'http://www.independent.co.uk/life-style/fashion/?service=rss'
),
(u'Life & Style -Food & Drink',
u'http://www.independent.co.uk/life-style/food-and-drink/?service=rss'
),
(u'Life & Style - Health and Families',
u'http://www.independent.co.uk/life-style/health-and-families/?service=rss'
),
(u'Life & Style - House & Home',
u'http://www.independent.co.uk/life-style/house-and-home/'),
(u'Life & Style - History',
u'http://www.independent.co.uk/life-style/history/?service=rss'
),
(u'Life & Style - Gadgets & Tech',
u'http://www.independent.co.uk/life-style/gadgets-and-tech/?service=rss'
),
(u'Life & Style - Motoring',
u'http://www.independent.co.uk/life-style/motoring/?service=rss'
),
(u'Arts & Ents - Art',
u'http://www.independent.co.uk/arts-entertainment/art/?service=rss'
),
(u'Arts & Ents - Architecture',
u'http://www.independent.co.uk/arts-entertainment/architecture/?service=rss'
),
(u'Arts & Ents - Music',
u'http://www.independent.co.uk/arts-entertainment/music/?service=rss'
),
(u'Arts & Ents - Classical',
u'http://www.independent.co.uk/arts-entertainment/classical/?service=rss'
),
(u'Arts & Ents - Films',
u'http://www.independent.co.uk/arts-entertainment/films/?service=rss'
),
(u'Arts & Ents - TV',
u'http://www.independent.co.uk/arts-entertainment/tv/?service=rss'
),
(u'Arts & Ents - Theatre and Dance',
u'http://www.independent.co.uk/arts-entertainment/theatre-dance/?service=rss'
),
(u'Arts & Ents - Comedy',
u'http://www.independent.co.uk/arts-entertainment/comedy/?service=rss'
),
(u'Arts & Ents - Books',
u'http://www.independent.co.uk/arts-entertainment/books/?service=rss'
),
(u'Travel', u'http://www.independent.co.uk/travel/?service=rss'
),
(u'Money', u'http://www.independent.co.uk/money/?service=rss'),
(u'IndyBest',
u'http://www.independent.co.uk/extras/indybest/?service=rss'),
]