calibre/recipes/mediapart.recipe
Kovid Goyal c577025037
pep8
2021-01-17 19:19:05 +05:30

317 lines
12 KiB
Python

#!/usr/bin/env python
# vim:fileencoding=utf-8
#
# 11 Jan 2021 - L. Houpert - Major changes in the Mediapart recipe:
# 1) Summary of the article are noow available
# 2) Additional sections International, France, Economie and Culture have
# been added through custom entries in the function my_parse_index.
# 3) Fix the cover image so it doesnt disappear from the Kindle menu
# ( cover image format is changed to .jpeg)
# 14 Jan 2021 - Add Mediapart Logo url as masthead_url and change cover
# by overlaying the date on top of the Mediapart cover
from __future__ import unicode_literals
__license__ = 'GPL v3'
__copyright__ = '2021, Loïc Houpert <houpertloic at gmail .com>. Adapted from: 2016, Daniel Bonnery; 2009, Mathieu Godlewski; 2010-2012, Louis Gesbert' # noqa
'''
Mediapart
'''
import re
from datetime import date, datetime, timezone, timedelta
from calibre.web.feeds import feeds_from_index
from calibre.web.feeds.news import BasicNewsRecipe
def classes(classes):
q = frozenset(classes.split(' '))
return dict(
attrs={'class': lambda x: x and frozenset(x.split()).intersection(q)}
)
class Mediapart(BasicNewsRecipe):
title = 'Mediapart'
__author__ = 'Loïc Houpert'
description = 'Global news in French from news site Mediapart'
publication_type = 'newspaper'
language = 'fr'
needs_subscription = True
oldest_article = 2
use_embedded_content = False
no_stylesheets = True
keep_only_tags = [
dict(name='h1'),
dict(name='div', **classes('author')),
classes('introduction content-article')
]
remove_tags = [classes('login-subscribe print-source_url')]
conversion_options = {'smarten_punctuation': True}
masthead_url = "https://raw.githubusercontent.com/lhoupert/calibre_contrib/main/mediapart_masthead.png"
# cover_url = 'https://raw.githubusercontent.com/lhoupert/calibre_contrib/main/mediapart.jpeg'
# --
# Get date in french time zone format
today = datetime.now(timezone.utc) + timedelta(hours=1)
oldest_article_date = today - timedelta(days=oldest_article)
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))
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'
):
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'})
all_articles = fils.findAll(separador_thread)
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 Exception:
article_mot_cle = ''
try:
article_type = article.find(
'a', {
'href': re.compile(r'.*\/type-darticles\/.*')
}
).renderContents().decode('utf-8')
except Exception:
article_type = ''
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')
# Add French timezone to date of the article for date check
article_date = article_date.replace(tzinfo=timezone.utc) + timedelta(hours=1)
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,
}
webpage_article.append(summary)
except Exception:
pass
specific_articles += [(type_of_article,
webpage_article)] if webpage_article else []
return specific_articles
articles = []
for category in dict_article_sources:
articles += get_articles(
category['type'], category['webpage'], category['separador']['page'],
category['separador']['thread']
)
return articles
# non-locale specific date parse (strptime("%d %b %Y",s) would work with
# french locale)
def parse_french_date(self, date_str):
date_arr = date_str.lower().split()
return date(
day=int(date_arr[0]),
year=int(date_arr[2]),
month=[
None, 'janvier', 'février', 'mars', 'avril', 'mai', 'juin',
'juillet', 'août', 'septembre', 'octobre', 'novembre', 'décembre'
].index(date_arr[1])
)
def get_browser(self):
# -- Handle login
def is_form_login(form):
return "id" in form.attrs and form.attrs['id'] == "logFormEl"
br = BasicNewsRecipe.get_browser(self)
if self.username is not None and self.password is not None:
br.open('https://www.mediapart.fr/login')
br.select_form(predicate=is_form_login)
br['name'] = self.username
br['password'] = self.password
br.submit()
return br
def default_cover(self, cover_file):
'''
Create a generic cover for recipes that don't have a cover
'''
from PyQt5.Qt import QImage, QPainter, QPen, Qt, QFont, QRect
from calibre.gui2 import ensure_app, load_builtin_fonts, pixmap_to_data
def init_environment():
ensure_app()
load_builtin_fonts()
def create_cover_mediapart(date):
' Create a cover for mediapart adding the date on Mediapart Cover'
init_environment()
# Get data
image_url = 'https://raw.githubusercontent.com/lhoupert/calibre_contrib/main/mediapart.jpeg'
data = self.index_to_soup(image_url, raw=True)
# Get date and hour corresponding to french time zone
today = datetime.now(timezone.utc) + timedelta(hours=1)
wkd = today.weekday()
french_weekday={0:'Mon',1:'Mar',2:'Mer',3:'Jeu',4:'Ven',5:'Sam',6:'Dim'}
day = french_weekday[wkd]+'.'
date = day + ' ' + today.strftime('%d %b. %Y')
edition = today.strftime('Édition de %Hh')
# Get Cover data
img = QImage()
img.loadFromData(data)
# Overlay date on cover
p = QPainter(img)
pen = QPen(Qt.black)
pen.setWidth(6)
p.setPen(pen)
font = QFont()
font.setFamily('Times')
font.setPointSize(78)
p.setFont(font)
r = QRect(0, 600, 744,100)
p.drawText(r, Qt.AlignmentFlag.AlignJustify | Qt.AlignmentFlag.AlignVCenter | Qt.AlignmentFlag.AlignCenter, date)
p.end()
# Overlay edition information on cover
p = QPainter(img)
pen = QPen(Qt.black)
pen.setWidth(4)
p.setPen(pen)
font = QFont()
font.setFamily('Times')
font.setItalic(True)
font.setPointSize(66)
p.setFont(font)
# Add date
r = QRect(0, 720, 744,100)
p.drawText(r, Qt.AlignmentFlag.AlignJustify | Qt.AlignmentFlag.AlignVCenter | Qt.AlignmentFlag.AlignCenter, edition)
p.end()
return pixmap_to_data(img)
try:
today=datetime.today()
date = today.strftime('%d %b %Y')
img_data = create_cover_mediapart(date)
cover_file.write(img_data)
cover_file.flush()
except Exception:
self.log.exception('Failed to generate default cover')
return False
return True