mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-09 03:04:10 -04:00
Frieze Magazine by Kabonix
This commit is contained in:
Kovid Goyal
parent
6155d07b1d
commit
27c5a12ed4
234
recipes/frieze.recipe
Normal file
234
recipes/frieze.recipe
Normal file
@ -0,0 +1,234 @@
|
|||||||
|
#!/usr/bin/env python
|
||||||
|
'''
|
||||||
|
frieze.com - Magazine d'art contemporain
|
||||||
|
'''
|
||||||
|
import re
|
||||||
|
from datetime import datetime
|
||||||
|
from zoneinfo import ZoneInfo
|
||||||
|
|
||||||
|
from calibre.web.feeds.news import BasicNewsRecipe
|
||||||
|
|
||||||
|
|
||||||
|
class FriezeMagazineRecipe(BasicNewsRecipe):
|
||||||
|
title = 'Frieze Magazine'
|
||||||
|
__author__ = 'Kabonix'
|
||||||
|
description = "Magazine international d'art contemporain"
|
||||||
|
language = 'en'
|
||||||
|
oldest_article = 60
|
||||||
|
max_articles_per_feed = 50
|
||||||
|
auto_cleanup = False
|
||||||
|
encoding = 'utf-8'
|
||||||
|
no_stylesheets = True
|
||||||
|
remove_javascript = True
|
||||||
|
scale_news_images_to_device = True
|
||||||
|
|
||||||
|
base_url = 'https://www.frieze.com'
|
||||||
|
|
||||||
|
keep_only_tags = [
|
||||||
|
{'class': 'article-header-container'},
|
||||||
|
{'class': 'article-header-title'},
|
||||||
|
{'class': 'article-header-headline'},
|
||||||
|
{'class': 'article-header-author'},
|
||||||
|
{'class': 'body-text'},
|
||||||
|
{'class': 'body-field'}
|
||||||
|
]
|
||||||
|
|
||||||
|
remove_tags = [
|
||||||
|
{'class': 'social-share-container'},
|
||||||
|
{'class': 'share-buttons'},
|
||||||
|
{'class': 'article-header-social-responsive'},
|
||||||
|
{'class': 'newsletter-subscribe-container'},
|
||||||
|
{'class': 'suggested-articles-container'},
|
||||||
|
{'class': 'article-footer-container'},
|
||||||
|
{'class': 'ad-surround'},
|
||||||
|
{'id': re.compile(r'ad-.*')},
|
||||||
|
{'class': 'hidden'}
|
||||||
|
]
|
||||||
|
|
||||||
|
extra_css = '''
|
||||||
|
img { max-width: 100%; height: auto; display: block; margin: 1em auto; }
|
||||||
|
h1 { font-size: 2em; margin: 1em 0; font-weight: bold; }
|
||||||
|
h2 { font-size: 1.5em; margin: 1em 0; font-weight: bold; }
|
||||||
|
p { font-size: 1.1em; line-height: 1.6; margin-bottom: 1em; }
|
||||||
|
figcaption { font-style: italic; font-size: 0.9em; color: #555; text-align: center; }
|
||||||
|
em { font-style: italic; }
|
||||||
|
.article-header-title { margin-bottom: 0.5em; }
|
||||||
|
.article-header-headline { margin-bottom: 1.5em; }
|
||||||
|
.article-header-author { font-size: 0.9em; margin-bottom: 2em; color: #555; }
|
||||||
|
'''
|
||||||
|
|
||||||
|
def parse_index(self):
|
||||||
|
magazine_url = self.base_url + '/magazines/frieze-magazine'
|
||||||
|
soup = self.index_to_soup(magazine_url)
|
||||||
|
|
||||||
|
# Recherche du dernier numéro de façon plus robuste
|
||||||
|
# Essayer plusieurs sélecteurs possibles pour trouver la liste des numéros
|
||||||
|
issue_containers = soup.find_all('div', {'class': re.compile(r'teaser-search-col')})
|
||||||
|
|
||||||
|
# S'il n'y a pas de conteneurs trouvés avec la classe spécifique, cherchons plus largement
|
||||||
|
if not issue_containers:
|
||||||
|
self.log.warning('Recherche élargie des numéros du magazine')
|
||||||
|
issue_containers = soup.find_all('div', {'class': re.compile(r'teaser')})
|
||||||
|
|
||||||
|
# Trouvons le premier numéro de magazine (le plus récent) parmi les conteneurs
|
||||||
|
latest_issue = None
|
||||||
|
for container in issue_containers:
|
||||||
|
# Vérifie si c'est un conteneur de magazine (pas un article)
|
||||||
|
if container.find('a', href=re.compile(r'/magazines/|/issues/')):
|
||||||
|
latest_issue = container
|
||||||
|
break
|
||||||
|
|
||||||
|
if not latest_issue:
|
||||||
|
self.log.warning('Aucun numéro spécifique trouvé, utilisation de la page principale')
|
||||||
|
issue_url = magazine_url
|
||||||
|
else:
|
||||||
|
issue_link = latest_issue.find('a')
|
||||||
|
if issue_link and issue_link.get('href'):
|
||||||
|
issue_url = self.base_url + issue_link['href'] if issue_link['href'].startswith('/') else issue_link['href']
|
||||||
|
self.log.info(f'Dernier numéro trouvé: {issue_url}')
|
||||||
|
else:
|
||||||
|
issue_url = magazine_url
|
||||||
|
self.log.warning('Lien vers le numéro non trouvé')
|
||||||
|
|
||||||
|
self.log.info(f'Accès au numéro: {issue_url}')
|
||||||
|
issue_soup = self.index_to_soup(issue_url)
|
||||||
|
|
||||||
|
# Extraction plus robuste de l'image de couverture
|
||||||
|
# Essayons plusieurs sélecteurs possibles
|
||||||
|
cover_image = None
|
||||||
|
|
||||||
|
# Méthode 1: Chercher dans la section d'en-tête du magazine
|
||||||
|
cover_div = issue_soup.find('div', class_=re.compile(r'magazine-header-image|issue-header-image'))
|
||||||
|
if cover_div:
|
||||||
|
img_tag = cover_div.find('img')
|
||||||
|
if img_tag and img_tag.get('src'):
|
||||||
|
cover_image = img_tag['src']
|
||||||
|
|
||||||
|
# Méthode 2: Chercher dans la section principale
|
||||||
|
if not cover_image:
|
||||||
|
main_section = issue_soup.find('section', class_=re.compile(r'main|content'))
|
||||||
|
if main_section:
|
||||||
|
img_tag = main_section.find('img')
|
||||||
|
if img_tag and img_tag.get('src'):
|
||||||
|
cover_image = img_tag['src']
|
||||||
|
|
||||||
|
# Méthode 3: Recherche générale d'une grande image en haut de la page
|
||||||
|
if not cover_image:
|
||||||
|
for img in issue_soup.find_all('img', src=True)[:5]: # Limiter aux 5 premières images
|
||||||
|
if 'cover' in img.get('src', '').lower() or 'header' in img.get('src', '').lower():
|
||||||
|
cover_image = img['src']
|
||||||
|
break
|
||||||
|
|
||||||
|
if cover_image:
|
||||||
|
self.cover_url = cover_image if cover_image.startswith('http') else self.base_url + cover_image
|
||||||
|
self.log.info(f'Couverture trouvée: {self.cover_url}')
|
||||||
|
else:
|
||||||
|
self.log.warning("Pas d'image de couverture trouvée")
|
||||||
|
|
||||||
|
# Extraction du titre du numéro
|
||||||
|
issue_title = None
|
||||||
|
for heading in issue_soup.find_all(['h1', 'h2']):
|
||||||
|
if 'issue' in heading.text.lower() or 'magazine' in heading.text.lower():
|
||||||
|
issue_title = heading.text.strip()
|
||||||
|
break
|
||||||
|
|
||||||
|
if not issue_title:
|
||||||
|
# Recherche plus générale
|
||||||
|
for heading in issue_soup.find_all(['h1', 'h2'])[:3]: # Limiter aux 3 premiers titres
|
||||||
|
issue_title = heading.text.strip()
|
||||||
|
if issue_title:
|
||||||
|
break
|
||||||
|
|
||||||
|
if not issue_title:
|
||||||
|
issue_title = 'Frieze Magazine - Dernier numéro'
|
||||||
|
|
||||||
|
# Extraction des articles
|
||||||
|
articles = []
|
||||||
|
|
||||||
|
# Recherche différents modèles de teasers d'articles
|
||||||
|
article_containers = issue_soup.find_all('div', {'class': re.compile(r'teaser-content|article-teaser|article-item')})
|
||||||
|
|
||||||
|
if not article_containers:
|
||||||
|
# Recherche plus large si les conteneurs spécifiques ne sont pas trouvés
|
||||||
|
article_containers = issue_soup.find_all('div', {'class': re.compile(r'teaser|article')})
|
||||||
|
|
||||||
|
for container in article_containers:
|
||||||
|
article_link = container.find('a')
|
||||||
|
if not article_link or not article_link.get('href'):
|
||||||
|
continue
|
||||||
|
|
||||||
|
article_url = article_link['href']
|
||||||
|
if article_url.startswith('/'):
|
||||||
|
article_url = self.base_url + article_url
|
||||||
|
|
||||||
|
# Extraction du titre avec différentes classes possibles
|
||||||
|
title_element = container.find(['div', 'h2', 'h3', 'h4'], {'class': re.compile(r'title|heading')})
|
||||||
|
title = title_element.text.strip() if title_element else 'Sans titre'
|
||||||
|
|
||||||
|
# Extraction de la description
|
||||||
|
desc_element = container.find(['div', 'p'], {'class': re.compile(r'deck|description|summary|excerpt')})
|
||||||
|
description = desc_element.text.strip() if desc_element else ''
|
||||||
|
|
||||||
|
# Extraction de l'auteur
|
||||||
|
author_element = container.find(['div', 'span'], {'class': re.compile(r'author|byline')})
|
||||||
|
if author_element:
|
||||||
|
author_links = author_element.find_all('a')
|
||||||
|
if author_links:
|
||||||
|
author = ', '.join([a.text.strip() for a in author_links])
|
||||||
|
else:
|
||||||
|
author = author_element.text.strip()
|
||||||
|
else:
|
||||||
|
author = ''
|
||||||
|
|
||||||
|
# Vérification que c'est bien un article et pas un lien interne
|
||||||
|
if '/article/' in article_url or '/feature/' in article_url or '/review/' in article_url:
|
||||||
|
articles.append({
|
||||||
|
'title': title,
|
||||||
|
'url': article_url,
|
||||||
|
'description': description,
|
||||||
|
'author': author,
|
||||||
|
'date': datetime.now(ZoneInfo('Europe/Paris')).strftime('%Y-%m-%d')
|
||||||
|
})
|
||||||
|
self.log.debug(f'Article trouvé: {title} - {article_url}')
|
||||||
|
|
||||||
|
self.log.info(f"Nombre d'articles trouvés: {len(articles)}")
|
||||||
|
return [(issue_title, articles)]
|
||||||
|
|
||||||
|
def get_cover_url(self):
|
||||||
|
return getattr(self, 'cover_url', None)
|
||||||
|
|
||||||
|
def preprocess_html(self, soup):
|
||||||
|
# Nettoyage des éléments superflus
|
||||||
|
for element in soup.find_all(class_=lambda c: c and ('share' in c or 'ad-' in c)):
|
||||||
|
element.decompose()
|
||||||
|
|
||||||
|
for tag in soup.find_all(['script', 'style']):
|
||||||
|
tag.decompose()
|
||||||
|
|
||||||
|
# Optimisation des images
|
||||||
|
for img in soup.find_all('img'):
|
||||||
|
# Sauvegarde des attributs importants seulement
|
||||||
|
src = img.get('src') or img.get('data-src') or img.get('data-lazy-src')
|
||||||
|
alt = img.get('alt', '')
|
||||||
|
|
||||||
|
# Réinitialiser tous les attributs
|
||||||
|
img.attrs = {}
|
||||||
|
|
||||||
|
# Réappliquer seulement src et alt
|
||||||
|
if src:
|
||||||
|
img['src'] = src
|
||||||
|
if alt:
|
||||||
|
img['alt'] = alt
|
||||||
|
|
||||||
|
# Formatage des métadonnées
|
||||||
|
for class_name, style in [
|
||||||
|
('article-header-title', 'font-size: 2em; font-weight: bold;'),
|
||||||
|
('article-header-headline', 'font-style: italic;'),
|
||||||
|
('article-header-author', 'color: #555;')
|
||||||
|
]:
|
||||||
|
element = soup.find('div', class_=class_name)
|
||||||
|
if element:
|
||||||
|
element.attrs.clear()
|
||||||
|
element['style'] = style
|
||||||
|
|
||||||
|
return soup
|
||||||
Loading…
x
Reference in New Issue
Block a user