mirror of
https://github.com/kovidgoyal/calibre.git
synced 2025-07-07 10:14:46 -04:00
rewritte my_parse_index so it handles more sections
This commit is contained in:
Loic Houpert
parent
34da60c843
commit
c7174bc8d6
@ -3,15 +3,17 @@
|
||||
from __future__ import unicode_literals
|
||||
|
||||
__license__ = 'GPL v3'
|
||||
__copyright__ = '2016, Daniel Bonnery ? (contact: DanielBonnery sur mobileread.com) 2009, Mathieu Godlewski <mathieu at godlewski.fr>; 2010-2012, Louis Gesbert <meta at antislash dot info>' # noqa
|
||||
__copyright__ = '2021, Loïc Houpert <houpertloic at gmail .com>. Adapted from: 2016, Daniel Bonnery; 2009, Mathieu Godlewski; 2010-2012, Louis Gesbert' # noqa
|
||||
'''
|
||||
Mediapart
|
||||
'''
|
||||
|
||||
import sys
|
||||
|
||||
import re
|
||||
from calibre.web.feeds.news import BasicNewsRecipe
|
||||
from calibre.web.feeds import feeds_from_index
|
||||
from datetime import date, timedelta
|
||||
from datetime import datetime, timedelta
|
||||
|
||||
|
||||
def classes(classes):
|
||||
@ -22,7 +24,7 @@ def classes(classes):
|
||||
|
||||
class Mediapart(BasicNewsRecipe):
|
||||
title = 'Mediapart'
|
||||
__author__ = 'Daniel Bonnery from a version by Mathieu Godlewski, Louis Gesbert'
|
||||
__author__ = 'Loïc Houpert (adapted from a version by Daniel Bonnery, Mathieu Godlewski and Louis Gesbert)'
|
||||
description = 'Global news in French from news site Mediapart'
|
||||
publication_type = 'newspaper'
|
||||
language = 'fr'
|
||||
@ -45,89 +47,157 @@ class Mediapart(BasicNewsRecipe):
|
||||
|
||||
# --
|
||||
|
||||
oldest_article_date = date.today() - timedelta(days=oldest_article)
|
||||
oldest_article_date = datetime.today() - timedelta(days=oldest_article)
|
||||
|
||||
# -- get the index (the feed at 'http://www.mediapart.fr/articles/feed' only has
|
||||
# the 10 last elements :/)
|
||||
|
||||
feeds = [
|
||||
('La Une', 'http://www.mediapart.fr/articles/feed'),
|
||||
]
|
||||
|
||||
# The feed at 'http://www.mediapart.fr/articles/feed' only displayed the 10
|
||||
# last elements so the articles are indexed on specific pages
|
||||
# in the function my_parse_index. In this function the article are parsed
|
||||
# using the funtion get_articles and the dict values dict_article_sources
|
||||
|
||||
|
||||
|
||||
def parse_feeds(self):
|
||||
feeds = super(Mediapart, self).parse_feeds()
|
||||
feeds += feeds_from_index(self.my_parse_index(feeds))
|
||||
print("\n======================================================" +
|
||||
"======================================================\n")
|
||||
print("======================================================" +
|
||||
"======================================================\n")
|
||||
print(f" List of feeds: {feeds}")
|
||||
#sys.exit("sys.exit for debug")
|
||||
return feeds
|
||||
|
||||
def my_parse_index(self, la_une):
|
||||
|
||||
dict_article_sources = [
|
||||
{'type':'Brèves',
|
||||
'webpage':'https://www.mediapart.fr/journal/fil-dactualites',
|
||||
'separador':{'page':'ul','thread':'li'}
|
||||
},
|
||||
{'type':'International',
|
||||
'webpage':'https://www.mediapart.fr/journal/international',
|
||||
'separador':{'page':'div','thread':'div'}
|
||||
},
|
||||
{'type':'France',
|
||||
'webpage':'https://www.mediapart.fr/journal/france',
|
||||
'separador':{'page':'div','thread':'div'}
|
||||
},
|
||||
{'type':'Économie',
|
||||
'webpage':'https://www.mediapart.fr/journal/economie',
|
||||
'separador':{'page':'div','thread':'div'}
|
||||
},
|
||||
{'type':'Culture',
|
||||
'webpage':'https://www.mediapart.fr/journal/culture-idees',
|
||||
'separador':{'page':'div','thread':'div'}
|
||||
},
|
||||
]
|
||||
|
||||
def get_articles(type_of_article,webpage,
|
||||
separador_page='ul', separador_thread='li'):
|
||||
|
||||
print("\n======================================================" +
|
||||
"======================================================")
|
||||
print(f"[Type of Article]:{type_of_article}")
|
||||
print(f"[Webpage]:{webpage}")
|
||||
print("\n======================================================" +
|
||||
"======================================================\n")
|
||||
|
||||
specific_articles = []
|
||||
|
||||
webpage_article = []
|
||||
soup = self.index_to_soup(webpage)
|
||||
page = soup.find('main', {'class': 'global-wrapper'})
|
||||
fils = page.find(separador_page, {'class': 'post-list universe-journal'})
|
||||
|
||||
# print(f"Print value of fils.findAll('li'):\n {fils.findAll('li')} ")
|
||||
all_articles = fils.findAll(separador_thread)
|
||||
# print(soup.prettify())
|
||||
for article in all_articles:
|
||||
try:
|
||||
title = article.find('h3', recursive=False)
|
||||
if title is None or ''.join(title['class']) == 'title-specific':
|
||||
# print(f"[BAD title entry] Print value of title:\n {title}")
|
||||
continue
|
||||
# print(f"\n[OK title entry] Print value of title:\n {title}\n")
|
||||
|
||||
try:
|
||||
article_mot_cle = article.find('a', {'href': re.compile(
|
||||
r'.*\/mot-cle\/.*')}).renderContents().decode('utf-8')
|
||||
except:
|
||||
article_mot_cle = ''
|
||||
|
||||
try:
|
||||
article_type = article.find('a', {'href': re.compile(
|
||||
r'.*\/type-darticles\/.*')}).renderContents().decode('utf-8')
|
||||
except:
|
||||
article_type = ''
|
||||
|
||||
# print(f"Article Type:\n {article_type}\n")
|
||||
|
||||
for s in title('span'):
|
||||
s.replaceWith(s.renderContents().decode('utf-8') + "\n")
|
||||
url = title.find('a', href=True)['href']
|
||||
|
||||
date = article.find('time', datetime=True)['datetime']
|
||||
article_date = datetime.strptime(date,'%Y-%m-%d')
|
||||
if article_date < self.oldest_article_date:
|
||||
print("article_date < self.oldest_article_date\n")
|
||||
continue
|
||||
|
||||
# print("-------- Recent article added to the list ------- \n")
|
||||
all_authors = article.findAll(
|
||||
'a', {'class': re.compile(r'\bjournalist\b')})
|
||||
authors = [self.tag_to_string(a) for a in all_authors]
|
||||
# print(f"Authors in tag <a>: {authors}")
|
||||
|
||||
# If not link to the author profile is available the
|
||||
# html separador is a span tag
|
||||
if not all_authors:
|
||||
try:
|
||||
all_authors = article.findAll(
|
||||
'span', {'class': re.compile(r'\bjournalist\b')})
|
||||
authors = [self.tag_to_string(a) for a in all_authors]
|
||||
# print(f"Authors in tag <span>: {authors}")
|
||||
except:
|
||||
authors = 'unknown'
|
||||
|
||||
description = article.find('p').renderContents().decode('utf-8')
|
||||
# print(f" <p> in article : {self.tag_to_string(description).strip()} ")
|
||||
|
||||
summary = {
|
||||
'title': self.tag_to_string(title).strip(),
|
||||
'description': description,
|
||||
'date': article_date.strftime("%a, %d %b, %Y %H:%M"),
|
||||
'author': ', '.join(authors),
|
||||
'article_type': article_type,
|
||||
'mot_cle': article_mot_cle.capitalize(),
|
||||
'url': 'https://www.mediapart.fr' + url,
|
||||
}
|
||||
|
||||
# print(f"\nSummary: {summary}")
|
||||
|
||||
webpage_article.append(summary)
|
||||
except:
|
||||
pass
|
||||
|
||||
specific_articles += [(type_of_article, webpage_article)] if webpage_article else []
|
||||
return specific_articles
|
||||
|
||||
articles = []
|
||||
|
||||
breves = []
|
||||
liens = []
|
||||
confidentiels = []
|
||||
for category in dict_article_sources:
|
||||
articles += get_articles(category['type'],category['webpage'],
|
||||
category['separador']['page'],
|
||||
category['separador']['thread']
|
||||
)
|
||||
|
||||
soup = self.index_to_soup(
|
||||
'https://www.mediapart.fr/journal/fil-dactualites')
|
||||
page = soup.find('main', {'class': 'global-wrapper'})
|
||||
fils = page.find('ul', {'class': 'post-list universe-journal'})
|
||||
|
||||
for article in fils.findAll('li'):
|
||||
try:
|
||||
title = article.find('h3', recursive=False)
|
||||
|
||||
if title is None or ''.join(title['class']) == 'title-specific':
|
||||
continue
|
||||
|
||||
# print "found fil ",title
|
||||
article_type = article.find('a', {'href': re.compile(
|
||||
r'.*\/type-darticles\/.*')}).renderContents().decode('utf-8')
|
||||
# print "kind: ",article_type
|
||||
|
||||
for s in title('span'):
|
||||
s.replaceWith(s.renderContents().decode('utf-8') + "\n")
|
||||
url = title.find('a', href=True)['href']
|
||||
|
||||
# article_date = self.parse_french_date(article.find("span", "article-date").renderContents().decode('utf-8'))
|
||||
# print("################################# 9")
|
||||
# print(article_date)
|
||||
|
||||
# if article_date < self.oldest_article_date:
|
||||
# print "too old"
|
||||
# continue
|
||||
|
||||
authors = article.findAll(
|
||||
'a', {'class': re.compile(r'\bjournalist\b')})
|
||||
authors = [self.tag_to_string(a) for a in authors]
|
||||
|
||||
# description = article.find('div', {'class': lambda c: c != 'taxonomy-teaser'}, recursive=False).findAll('p')
|
||||
|
||||
# print "fil ",title," by ",authors," : ",description
|
||||
|
||||
summary = {
|
||||
'title': self.tag_to_string(title).strip(),
|
||||
'author': ', '.join(authors),
|
||||
'url': 'https://www.mediapart.fr' + url
|
||||
}
|
||||
if article_type == 'Lien':
|
||||
liens.append(summary)
|
||||
if article_type == 'Confidentiel':
|
||||
confidentiels.append(summary)
|
||||
if article_type not in ['Lien', 'Confidentiel']:
|
||||
breves.append(summary)
|
||||
except:
|
||||
pass
|
||||
|
||||
# print 'La Une: ', len(la_une), ' articles'
|
||||
# for a in la_une: print a["title"]
|
||||
# print 'Brèves: ', len(breves), ' articles'
|
||||
# print 'Revue web: ', len(liens), ' articles'
|
||||
# print 'Confidentiel: ', len(confidentiels), ' articles'
|
||||
|
||||
articles += [('Brèves', breves)] if breves else []
|
||||
articles += [('Revue du Web', liens)] if liens else []
|
||||
articles += [('Confidentiel', confidentiels)] if confidentiels else []
|
||||
print(articles)
|
||||
return articles
|
||||
# -- print-version
|
||||
|
||||
conversion_options = {'smarten_punctuation': True}
|
||||
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user