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

Ensure deduplication by simplifying keep_only_tags

Browse Source
This commit is contained in:
Henrik Holm 2025-05-04 13:56:30 +02:00
parent be586a4b69
commit a76310665b
No known key found for this signature in database
1 changed files with 7 additions and 59 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

66
recipes/fokus.recipe
View File

@ -1,6 +1,5 @@
#!/usr/bin/env python
# vim:fileencoding=utf-8
import re
from datetime import datetime, timezone
from mechanize import Request
@ -37,42 +36,14 @@ class Fokus(BasicNewsRecipe):
div.calibre-nuked-tag-figure { font-size: small; text-align: center; }
'''
remove_tags = [
dict(name='div', attrs={'class': 'External-ad'}),
dict(name='header', attrs={'class': 'Header'}),
dict(name='div', attrs={'class': 'Header-expanded'}),
dict(name='div', attrs={'class': 'Overlay'}),
dict(name='div', attrs={'class': 'Search-expanded'}),
dict(name='section', attrs={'class': 'Site__footer'}),
dict(name='div', attrs={'class': 'Toaster'}),
dict(name='div', attrs={'class': 'fbc-badge'}),
dict(name='div', attrs={'class': 'Posts-by-related-cat'}),
dict(name='div', attrs={'class': 'finite-scroll'}),
dict(name='div', attrs={'class': 'Sidebar'}),
dict(name='div', attrs={'id': 'single-comments'}),
dict(name='footer', attrs={'class': 'Single__footer'}),
dict(name='div', attrs={'class': 'Social-share'}),
dict(name='div', attrs={'class': 'mediaconnect-paywall'}),
dict(name='svg', attrs={'class': 'icon'}),
dict(name='figure', attrs={'class': 'wp-block-audio'}),
]
remove_tags_after = [
dict(name='div', class_='Single__content'),
]
keep_only_tags = [
dict(name='h1', class_='Single__title'), # Title.
dict(name='h1', class_='Longread__title'), # Alt. title.
dict(name='p', class_='Single__lead'), # Lead text.
dict(name='p', class_='Longread__lead'), # Alt. lead text.
dict(name='figure', class_='Single__thumbnail'), # Image.
dict(name='figure', class_='Longread__thumbnail'), # Alt. image.
# dict(name='p', class_='Meta__author'), # Author.
# dict(name='time', class_='Meta__updated'), # Last updated.
# Main article.
dict(name='div', class_='sesamy-protected-content'),
dict(name='div', class_='wp-block-core-paragraph'),
dict(name='h1', class_='Single__title'), # Title of "Single" type articles.
dict(name='h1', class_='Longread__title'), # Title of "Longread" type articles.
dict(name='p', class_='Single__lead'), # Lead text of "Single" type articles.
dict(name='p', class_='Longread__lead'), # Lead text of "Longread" type articles.
dict(name='figure', class_='Single__thumbnail'), # Image of "Single" type articles.
dict(name='figure', class_='Longread__thumbnail'), # Image of "Longread" type articles.
dict(name='div', class_='sesamy-protected-content'), # Article body.
]
def get_cover_url(self) -> str:
@ -319,26 +290,3 @@ class Fokus(BasicNewsRecipe):
self.log(f'A total of {num_articles} articles belonging to {len(section_to_articles)} sections were kept.')
return feeds
def postprocess_html(self, soup, _, read_more_regex: re.Pattern = re.compile(r'^Läs även:')):
# When scraped, Fokus sometimes returns a duplicate of the article, i.e., the exact same article concatenated
# to itself. To avoid duplication in the output file, remove all <div> tags after the <p> tag whose content is
# "***" (the Fokus indication of the end of the article). "***" is sometimes followed by one or multiple
# "Läs även:" ("Read also:") paragraphs. If such paragraphs are encountered, do not delete them.
has_reached_end, has_reached_read_more = False, False
for div_tag in soup.find_all('div', class_='wp-block-core-paragraph'):
if has_reached_read_more:
div_tag.decompose()
elif has_reached_end and div_tag.find('strong', text=read_more_regex):
# If the end has been reached, check if this is a "Läs även:" paragraph. If yes, keep it.
continue
elif has_reached_end and not has_reached_read_more:
# If the end of the article has been reached, and if no more "Läs även:" paragraphs follow, delete this
# tag and set `has_reached_read_more` to `True`.
has_reached_read_more = True
div_tag.decompose()
elif div_tag.find('p', text='***'):
# The end of the article has been reached.
has_reached_end = True
return soup