calibre/recipes/science_news.recipe

100 lines
2.7 KiB
Python

#!/usr/bin/env python
__license__ = 'GPL v3'
'''
sciencenews.org
'''
from calibre.web.feeds.news import BasicNewsRecipe
import datetime
import re
class ScienceNewsIssue(BasicNewsRecipe):
title = u'Science News'
description = ("Science News is an award-winning bi-weekly newsmagazine covering the most important research"
" in all fields of science. This recipe downloads all the articles from the latest issue.")
category = u'Science, Technology, News'
publisher = u'Society for Science & the Public'
oldest_article = 14
language = 'en'
max_articles_per_feed = 50
no_stylesheets = True
use_embedded_content = False
timefmt = ' [%A, %d %B, %Y]'
auto_cleanup = False
keep_only_tags = [
dict(
attrs={
'class':
lambda x: x and (
'single__content___' in x or 'header-default__title___' in x or
'header-default__deck___' in x or 'header-default__figure___' in
x
)
}
)
]
remove_tags = [
dict(
attrs={'class': lambda x: x and ('newsletter-signup__wrapper___' in x)}
)
]
def parse_index(self):
# Get URL of latest mag page
ld = self._get_mag_date()
url = f"https://www.sciencenews.org/sn-magazine/{ld:%B}-{ld.day}-{ld.year}"
url = url.lower()
# Get articles
soup = self.index_to_soup(url)
re_article = re.compile("https://www.sciencenews.org/article/")
stories = []
past_urls = set()
for sec in soup.find_all(href=re_article):
article_url = sec["href"]
article_title = sec.text.strip()
# Ignore image URLs which do not have text title
if article_title == "":
continue
# Ignore if link is a duplicate
if article_url in past_urls:
continue
past_urls.add(article_url)
article_info = {
"url": article_url,
"title": article_title,
}
stories.append(article_info)
index = [
("Articles", stories),
]
return index
def _get_mag_date(self):
"""Return date of latest magazine issue.
It is published every 2 weeks."""
d = datetime.date(2022, 6, 18)
t = datetime.date.today()
ld = None
while d <= t:
ld = d
d += datetime.timedelta(days=14)
return ld
def get_cover_url(self):
ld = self._get_mag_date()
url = ld.strftime(
"https://www.sciencenews.org/wp-content/uploads/%Y/%m/%m%d%y_cover.jpg"
)
return url