| 1 |
#!/usr/bin/env python |
| 2 |
# vim:fileencoding=UTF-8:ts=4:sw=4:sta:et:sts=4:fdm=marker:ai |
| 3 |
from __future__ import absolute_import, division, print_function, unicode_literals |
| 4 |
|
| 5 |
__author__ = 'S. Durand <[email protected]>' |
| 6 |
__license__ = 'GPL v3' |
| 7 |
|
| 8 |
''' |
| 9 |
lemonde.fr |
| 10 |
''' |
| 11 |
|
| 12 |
import re |
| 13 |
from datetime import date |
| 14 |
|
| 15 |
from calibre.web.feeds.news import BasicNewsRecipe, classes |
| 16 |
|
| 17 |
|
| 18 |
class LeMondeNumerique(BasicNewsRecipe): |
| 19 |
title = 'Le Monde: Édition abonnés' |
| 20 |
__author__ = 'Sylvain Durand' |
| 21 |
description = 'La version numérique du quotidien Le Monde' |
| 22 |
publisher = 'Société Editrice du Monde' |
| 23 |
publication_type = 'newspaper' |
| 24 |
needs_subscription = True |
| 25 |
language = 'fr' |
| 26 |
|
| 27 |
no_stylesheets = True |
| 28 |
ignore_duplicate_articles = {'title', 'url'} |
| 29 |
|
| 30 |
conversion_options = {'publisher': publisher} |
| 31 |
|
| 32 |
masthead_url = 'http://upload.wikimedia.org/wikipedia/commons/thumb/5/54/Le_monde_logo.svg/800px-Le_monde_logo.svg.png' |
| 33 |
|
| 34 |
lm_sections = [ |
| 35 |
'international:International', |
| 36 |
'politique:Politique', |
| 37 |
'societe:Société', |
| 38 |
'economie:Éco', |
| 39 |
'culture:Culture', |
| 40 |
'idees:Idées', |
| 41 |
'planete:Planète', |
| 42 |
'sport:Sport', |
| 43 |
'sciences:Sciences', |
| 44 |
'pixels:Pixels', |
| 45 |
'campus:Campus', |
| 46 |
] |
| 47 |
|
| 48 |
keep_only_tags = [classes('article__header'), dict(name='section', attrs={'class': ['article__content', 'article__heading', 'article__wrapper']})] |
| 49 |
|
| 50 |
remove_tags = [ |
| 51 |
classes('article__status meta__date meta__reading-time meta__social multimedia-embed'), |
| 52 |
dict(name=['footer', 'link']), |
| 53 |
dict(name='img', attrs={'class': ['article__author-picture']}), |
| 54 |
dict(name='section', attrs={'class': ['article__reactions', 'author', 'catcher', 'portfolio', 'services-inread']}), |
| 55 |
] |
| 56 |
|
| 57 |
remove_attributes = ['data-sizes', 'height', 'sizes', 'width'] |
| 58 |
|
| 59 |
preprocess_regexps = [ |
| 60 |
# insert space between author name and description |
| 61 |
(re.compile(r'(<span class="[^"]*author__desc[^>]*>)([^<]*</span>)', re.IGNORECASE), lambda match: match.group(1) + ' ' + match.group(2)), |
| 62 |
# insert " | " between article type and description |
| 63 |
(re.compile(r'(<span class="[^"]*article__kicker[^>]*>[^<]*)(</span>)', re.IGNORECASE), lambda match: match.group(1) + ' | ' + match.group(2)), |
| 64 |
] |
| 65 |
|
| 66 |
extra_css = ''' |
| 67 |
h2 { font-size: 1em; } |
| 68 |
h3 { font-size: 1em; } |
| 69 |
.article__desc { font-weight: bold; } |
| 70 |
.article__fact { font-weight: bold; text-transform: uppercase; } |
| 71 |
.article__kicker { text-transform: uppercase; } |
| 72 |
.article__legend { font-size: 0.6em; margin-bottom: 1em; } |
| 73 |
.article__title { margin-top: 0em; } |
| 74 |
''' |
| 75 |
|
| 76 |
def get_browser(self): |
| 77 |
br = BasicNewsRecipe.get_browser(self) |
| 78 |
if self.username is not None and self.password is not None: |
| 79 |
try: |
| 80 |
br.open('https://secure.lemonde.fr/sfuser/connexion') |
| 81 |
br.select_form(nr=0) |
| 82 |
br['email'] = self.username |
| 83 |
br['password'] = self.password |
| 84 |
br.submit() |
| 85 |
except Exception as e: |
| 86 |
self.log('Login failed with error:', str(e)) |
| 87 |
return br |
| 88 |
|
| 89 |
def get_cover_url(self): |
| 90 |
# today's date is a reasonable guess for the ID of the cover |
| 91 |
cover_id = date.today().strftime('%Y%m%d') |
| 92 |
soup = self.index_to_soup('https://www.lemonde.fr/') |
| 93 |
a = soup.find('a', {'id': 'jelec_link', 'style': True}) |
| 94 |
if a and a['style']: |
| 95 |
url = a['style'].split('/') |
| 96 |
if len(url) > 5 and url[3].isdigit(): |
| 97 |
# overwrite guess if actual cover ID was found |
| 98 |
cover_id = url[3] |
| 99 |
return 'https://www.lemonde.fr/thumbnail/journal/' + cover_id + '/1000/1490' |
| 100 |
|
| 101 |
def parse_index(self): |
| 102 |
ans = [] |
| 103 |
for x in self.lm_sections: |
| 104 |
s, section_title = x.partition(':')[::2] |
| 105 |
self.log('Processing section', section_title, '...') |
| 106 |
articles = list(self.parse_section('https://www.lemonde.fr/%s/' % s)) |
| 107 |
if articles: |
| 108 |
ans.append((section_title, articles)) |
| 109 |
return ans |
| 110 |
|
| 111 |
def parse_section(self, url): |
| 112 |
soup = self.index_to_soup(url) |
| 113 |
for article in soup.find_all('section', {'class': 'teaser'}): |
| 114 |
# extract URL |
| 115 |
a = article.find('a', {'class': 'teaser__link', 'href': True}) |
| 116 |
if a is None: |
| 117 |
continue |
| 118 |
url = a['href'] |
| 119 |
# skip articles without relevant content (e.g., videos) |
| 120 |
for el in 'blog chat live newsletters podcasts portfolio video visuel'.split(): |
| 121 |
if '/' + el + '/' in url: |
| 122 |
url = None |
| 123 |
break |
| 124 |
if url is None: |
| 125 |
continue |
| 126 |
# extract title |
| 127 |
h3 = article.find('h3', {'class': 'teaser__title'}) |
| 128 |
if h3 is None: |
| 129 |
continue |
| 130 |
title = self.tag_to_string(h3) |
| 131 |
# extract description |
| 132 |
desc = '' |
| 133 |
p = article.find('p', {'class': 'teaser__desc'}) |
| 134 |
if p is not None: |
| 135 |
desc = self.tag_to_string(p) |
| 136 |
self.log('\tFound article', title, 'at', url) |
| 137 |
yield {'title': title, 'url': url, 'description': desc} |
| 138 |
|
| 139 |
def preprocess_html(self, soup): |
| 140 |
# when an image is available in multiple sizes, select the smallest one |
| 141 |
for img in soup.find_all('img', {'data-srcset': True}): |
| 142 |
data_srcset = img['data-srcset'].split() |
| 143 |
if len(data_srcset) > 1: |
| 144 |
img['src'] = data_srcset[-2] |
| 145 |
del img['data-srcset'] |
| 146 |
return soup |
| 147 |
|
| 148 |
def postprocess_html(self, soup, first_fetch): |
| 149 |
# remove local hyperlinks |
| 150 |
for a in soup.find_all('a', {'href': True}): |
| 151 |
if '.lemonde.fr/' in a['href']: |
| 152 |
a.replace_with(self.tag_to_string(a)) |
| 153 |
# clean up header |
| 154 |
for ul in soup.find_all('ul', {'class': 'breadcrumb'}): |
| 155 |
div = soup.new_tag('div') |
| 156 |
category = '' |
| 157 |
for li in ul.find_all('li', {'class': True}): |
| 158 |
category += self.tag_to_string(li).strip().upper() + ' - ' |
| 159 |
div.string = category[:-3] |
| 160 |
ul.replace_with(div) |
| 161 |
return soup |