Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-09 03:04:10 -04:00
Code Issues Packages Projects Releases Wiki Activity

Update to follow updated structure of Fokus website

Browse Source
This commit is contained in:
Henrik Holm 2025-05-16 18:33:47 +02:00
parent b74453ce94
commit 84772e8b14
No known key found for this signature in database
1 changed files with 7 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

27
recipes/fokus.recipe
View File

@ -131,27 +131,14 @@ class Fokus(BasicNewsRecipe):
if url.startswith('/'):
url = f'{self.main_url}{url}'
if title_tag := a_tag.find('h2', {'class': 'Blurb__title'}):
if title_tag := a_tag.find('h2', {'class': 'PostBlurb__title'}):
title = self.tag_to_string(title_tag).strip()
if time_tag := a_tag.find('time', {'class': 'Blurb__date'}):
swedish_date_str = self.tag_to_string(time_tag).rstrip()
desc = ''
if desc_tag := a_tag.find('div', {'class': 'PostBlurb__excerpt'}):
desc = self.tag_to_string(desc_tag).strip()
# Skip articles older than `self.oldest_article`.
datetime_str = time_tag['datetime']
datetime_time = datetime.strptime(datetime_str, '%Y-%m-%dT%H:%M:%S%z')
now = datetime.now(timezone.utc)
delta = now - datetime_time
if delta.days > self.oldest_article:
self.log.debug(f"\tSkipping article as it is too old: '{title}'")
return
return {'url': url, 'title': title, 'description': desc}
desc = ''
if desc_tag := a_tag.find('div', {'class': 'Blurb__summary'}):
desc = self.tag_to_string(desc_tag).strip()
if in_cooperation_with_tag := a_tag.find('p', {'class': 'Blurb__meta'}):
desc += f' ({self.tag_to_string(in_cooperation_with_tag).strip()})'
return {'url': url, 'title': title, 'description': desc, 'date': swedish_date_str}
return
def _get_article_blurbs(self, soup) -> dict[str, dict[str, str, str, str]]:
@ -169,13 +156,13 @@ class Fokus(BasicNewsRecipe):
def _log(article) -> None:
'''Log a digestible summary of the input `article` blurb.'''
log_message = f"\t{article['title']} : {article['date']} : {article['url']}"
log_message = f"\t{article['title']} : {article['url']}"
if article.get('description'):
log_message += f" : {article['description']}"
self.log.debug(log_message)
try:
article_blurbs = soup.find_all('article', {'class': 'Blurb'})
article_blurbs = soup.find_all('article', {'class': 'PostBlurb'})
except AttributeError:
article_blurbs = []