mirror of
https://github.com/kovidgoyal/calibre.git
synced 2026-04-02 23:31:59 -04:00
Telerama by Kabonix
This commit is contained in:
Kovid Goyal
parent
049cdb9c61
commit
1f76bb161c
189
recipes/telerama.recipe
Normal file
189
recipes/telerama.recipe
Normal file
@ -0,0 +1,189 @@
|
||||
#!/usr/bin/env python
|
||||
# vim:fileencoding=utf-8
|
||||
from __future__ import absolute_import, division, print_function, unicode_literals
|
||||
|
||||
__license__ = 'GPL v3'
|
||||
__copyright__ = '2026, Kabonix'
|
||||
|
||||
import json
|
||||
import re
|
||||
from urllib.parse import urlparse
|
||||
|
||||
from calibre.web.feeds.news import BasicNewsRecipe
|
||||
|
||||
|
||||
class TeleramaPremium(BasicNewsRecipe):
|
||||
title = 'Télérama'
|
||||
__author__ = 'Kabonix'
|
||||
description = 'Édition complète (API Bypass) - Cover HD & Lecture Pure'
|
||||
publisher = 'Télérama'
|
||||
language = 'fr'
|
||||
encoding = 'utf-8'
|
||||
|
||||
oldest_article = 7
|
||||
max_articles_per_feed = 50
|
||||
no_stylesheets = True
|
||||
ignore_duplicate_articles = {'title', 'url'}
|
||||
|
||||
# On laisse les images des articles tranquilles
|
||||
scale_news_images = None
|
||||
|
||||
# --- API ---
|
||||
headers = {
|
||||
'User-Agent': 'Telerama/4.3.5 (Android; 14)',
|
||||
'X-Lmd-Token': 'TWPLMOLMO',
|
||||
'Accept': 'application/json'
|
||||
}
|
||||
|
||||
def get_browser(self, *args, **kwargs):
|
||||
br = BasicNewsRecipe.get_browser(self, *args, **kwargs)
|
||||
for name, val in self.headers.items():
|
||||
br.addheaders.append((name, val))
|
||||
return br
|
||||
|
||||
# --- COUVERTURE DYNAMIQUE (Ta demande) ---
|
||||
def get_cover_url(self):
|
||||
cover_url = None
|
||||
try:
|
||||
self.log('🔍 Recherche de la dernière couverture...')
|
||||
# On va sur la page kiosque
|
||||
soup = self.index_to_soup('https://www.telerama.fr/kiosque/telerama')
|
||||
|
||||
# On cherche le premier élément "popin-link" avec une data-cover-url
|
||||
# Le premier de la liste est toujours le dernier numéro paru
|
||||
link = soup.find('a', attrs={'class': 'popin-link', 'data-cover-url': True})
|
||||
|
||||
if link:
|
||||
url = link['data-cover-url']
|
||||
# L'URL contient /180/ (basse déf). On passe en HD /1200/
|
||||
# Ex: .../0/0/180/0/... -> .../0/0/1200/0/...
|
||||
cover_url = url.replace('/180/', '/1200/')
|
||||
self.log(f'✅ Couverture trouvée : {cover_url}')
|
||||
else:
|
||||
self.log('⚠️ Aucune couverture trouvée dans le kiosque.')
|
||||
except Exception as e:
|
||||
self.log(f'❌ Erreur récupération couverture : {e}')
|
||||
|
||||
return cover_url
|
||||
|
||||
# --- BYPASS API ---
|
||||
def get_article_url(self, article):
|
||||
url = article.get('link', article.get('url', ''))
|
||||
path = urlparse(url).path
|
||||
return 'https://apps.telerama.fr/tlr/v1/premium-android-phone/element?id={}'.format(path)
|
||||
|
||||
# --- JSON TO HTML ---
|
||||
def preprocess_raw_html(self, raw_html, url):
|
||||
if '/tlr/v1/' in url:
|
||||
try:
|
||||
data = json.loads(raw_html)
|
||||
content = ''
|
||||
title = 'Télérama'
|
||||
|
||||
if 'templates' in data and 'raw_content' in data['templates']:
|
||||
content = data['templates']['raw_content']['content']
|
||||
elif 'body' in data:
|
||||
content = data['body']
|
||||
|
||||
if 'template_vars' in data:
|
||||
title = data['template_vars'].get('share_title', 'Article Télérama')
|
||||
|
||||
if not content:
|
||||
return '<html><body><h2>Contenu vide</h2></body></html>'
|
||||
|
||||
# Nettoyage préventif
|
||||
content = content.replace('{{{ scripts_bottom }}}', '')
|
||||
content = re.sub(r'>\s*[•·]\s*<', '><', content)
|
||||
|
||||
# Fix images
|
||||
content = content.replace('{{width}}', '1200').replace('{{height}}', '')
|
||||
content = content.replace('%7B%7Bwidth%7D%7D', '1200')
|
||||
|
||||
html = f'<html><head><title>{title}</title></head><body><h1 id="main-title">{title}</h1>{content}</body></html>'
|
||||
return html
|
||||
|
||||
except Exception as e:
|
||||
self.log(f'Erreur JSON : {e}')
|
||||
return raw_html
|
||||
return raw_html
|
||||
|
||||
# --- NETTOYAGE ---
|
||||
def preprocess_html(self, soup):
|
||||
# 1. Suppression doublons structurels
|
||||
for header in soup.find_all(attrs={'class': re.compile(r'article__page-header|header__article', re.I)}):
|
||||
header.decompose()
|
||||
for ns in soup.find_all('noscript'):
|
||||
ns.decompose()
|
||||
|
||||
# 2. Suppression "À lire aussi"
|
||||
for p in soup.find_all(['p', 'h3', 'h4', 'div', 'aside']):
|
||||
text = p.get_text().strip()
|
||||
if re.search(r'^(À|A) lire aussi', text, re.IGNORECASE):
|
||||
p.decompose()
|
||||
|
||||
# 3. Nettoyage Méta TV et Puces
|
||||
for tag in soup.find_all(['p', 'div', 'span', 'li', 'ul']):
|
||||
text = tag.get_text().strip()
|
||||
normalized_text = re.sub(r'\s+', ' ', text)
|
||||
|
||||
# Puces seules
|
||||
if re.match(r'^[\s\n\r•·|\-.]+$', text):
|
||||
tag.decompose()
|
||||
continue
|
||||
# Mots clés TV seuls
|
||||
if re.match(r'^(Direct|Inédit|Replay|\s)+$', normalized_text, re.IGNORECASE):
|
||||
tag.decompose()
|
||||
continue
|
||||
# Mots clés TV avec séparateurs
|
||||
if re.search(r'(Direct|Inédit|Replay)\s*[•·-]', text, re.IGNORECASE):
|
||||
tag.decompose()
|
||||
|
||||
# 4. SUPPRESSION DES LIENS (Lecture Pure)
|
||||
for a in soup.find_all('a'):
|
||||
a.unwrap()
|
||||
|
||||
return soup
|
||||
|
||||
keep_only_tags = [
|
||||
dict(name='h1', attrs={'id': 'main-title'}),
|
||||
dict(attrs={'class': ['article__page-content', 'article-body']}),
|
||||
]
|
||||
|
||||
remove_tags = [
|
||||
dict(attrs={'class': re.compile(r'paywall|premium-banner|banner|pubstack|marketing', re.I)}),
|
||||
dict(attrs={'class': re.compile(r'sharing|social|bookmark|button|btn|openapp|listBtns', re.I)}),
|
||||
dict(attrs={'class': re.compile(r'OUTBRAIN|forecast|overlay', re.I)}),
|
||||
dict(name=['script', 'style', 'nav', 'footer', 'button', 'iframe'])
|
||||
]
|
||||
|
||||
extra_css = '''
|
||||
h1 {
|
||||
font-family: "Georgia", serif;
|
||||
font-size: 1.5em;
|
||||
font-weight: bold;
|
||||
text-align: center;
|
||||
margin-bottom: 0.5em;
|
||||
color: #111;
|
||||
}
|
||||
.article__label-subscriber {
|
||||
display: block; background-color: #ffe600; color: #000; font-weight: bold;
|
||||
font-size: 0.8em; text-transform: uppercase; padding: 4px 8px;
|
||||
margin: 0 auto 1em auto; width: fit-content; border-radius: 4px;
|
||||
}
|
||||
.article__chapeau { font-weight: bold; font-style: italic; margin: 1.5em 0; font-size: 1.1em; color: #444; }
|
||||
p { text-align: justify; line-height: 1.5; margin-bottom: 1em; }
|
||||
figure { margin: 1.5em 0; }
|
||||
img { display: block; margin: 0 auto; max-width: 100%; height: auto; }
|
||||
figcaption, .media__caption, .media__legend { font-size: 0.75em; color: #666; text-align: center; font-style: italic; margin-top: 0.5em; }
|
||||
.author { font-weight: bold; margin-top: 2em; border-top: 1px solid #eee; padding-top: 1em; color: #333; }
|
||||
a { color: inherit; text-decoration: none; pointer-events: none; }
|
||||
'''
|
||||
|
||||
feeds = [
|
||||
('À la une', 'https://www.telerama.fr/rss/une.xml'),
|
||||
('Cinéma', 'https://www.telerama.fr/rss/cinema.xml'),
|
||||
('Séries', 'https://www.telerama.fr/rss/series.xml'),
|
||||
('Télévision', 'https://www.telerama.fr/rss/television.xml'),
|
||||
('Musique', 'https://www.telerama.fr/rss/musique.xml'),
|
||||
('Livres', 'https://www.telerama.fr/rss/livres.xml'),
|
||||
]
|
||||
Loading…
x
Reference in New Issue
Block a user