Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-08 10:44:09 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix recipe for nature.com

Browse Source
This commit is contained in:
ping 2022-06-10 12:21:48 +08:00
parent 609b431c91
commit ef487db79c
No known key found for this signature in database
GPG Key ID: 6CCF56BCEDD24084
1 changed files with 13 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
recipes/nature.recipe
View File

@ -1,7 +1,7 @@
#!/usr/bin/env python
from collections import defaultdict
from calibre.web.feeds.news import BasicNewsRecipe
from calibre.web.feeds.news import BasicNewsRecipe, classes
BASE = 'https://www.nature.com'
@ -39,11 +39,15 @@ class Nature(BasicNewsRecipe):
no_javascript = True
no_stylesheets = True
keep_only_tags = [
dict(name='div', attrs={'data-component': check_words('article-container')})
]
keep_only_tags = [dict(name="article")]
remove_tags = [dict(attrs={'class': check_words('hide-print')})]
remove_tags = [
classes(
"u-hide-print hide-print c-latest-content__item c-context-bar "
"c-pdf-button__container u-js-hide"
),
dict(name="img", attrs={"class": ["visually-hidden"]}),
]
def parse_index(self):
soup = self.index_to_soup(BASE + '/nature/current-issue')
@ -57,11 +61,9 @@ class Nature(BasicNewsRecipe):
failed, img src might have changed, use default width 200
"""
pass
section_tags = soup.find(
'div', {'data-container-type': check_words('issue-section-list')}
)
section_tags = section_tags.findAll(
'div', {'class': check_words('article-section')}
section_tags = soup.find_all(
"section", attrs={"data-container-type": "issue-section-list"}
)
sections = defaultdict(list)