Cutlery/calibre
1
0
Fork 0
mirror of https://github.com/kovidgoyal/calibre.git synced 2025-07-08 18:54:09 -04:00
Code Issues Packages Projects Releases Wiki Activity

Merge branch 'master' of https://github.com/unkn0w7n/calibre

Browse Source
This commit is contained in:
Kovid Goyal 2025-01-05 14:51:57 +05:30
commit d2ff5bb2dd
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 14 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
recipes/substack.recipe
View File

@ -20,7 +20,7 @@ import re
from mechanize import Request from mechanize import Request
from calibre.web.feeds.news import BasicNewsRecipe from calibre.web.feeds.news import BasicNewsRecipe, classes
class Substack(BasicNewsRecipe): class Substack(BasicNewsRecipe):
@ -30,13 +30,23 @@ class Substack(BasicNewsRecipe):
oldest_article = 7 oldest_article = 7
language = 'en' language = 'en'
max_articles_per_feed = 100 max_articles_per_feed = 100
auto_cleanup = True
auto_cleanup_keep = '//*[@class="subtitle"]'
needs_subscription = 'optional' needs_subscription = 'optional'
use_embedded_content = False use_embedded_content = False
masthead_url = 'https://substack.com/img/substack_wordmark.png' masthead_url = 'https://substack.com/img/substack_wordmark.png'
cover_url = 'https://substack.com/img/substack.png' cover_url = 'https://substack.com/img/substack.png'
extra_css = '.captioned-image-container, .image-container {font-size: small;}' extra_css = '.captioned-image-container, .image-container, .image-caption {font-size: small;}'
remove_empty_feeds = True
remove_attributes = ['style', 'height', 'width']
no_stylesheets = True
keep_only_tags = [
classes('post-title post-subtitle subtitle available-content')
]
remove_tags = [
dict(name=['svg', 'source']),
classes('subscribe-widget button-wrapper')
]
recipe_specific_options = { recipe_specific_options = {
'auths': { 'auths': {
@ -113,8 +123,4 @@ class Substack(BasicNewsRecipe):
res = w res = w
for img in soup.findAll('img', attrs={'src': True}): for img in soup.findAll('img', attrs={'src': True}):
img['src'] = re.sub(r'w_\d+', 'w_' + res, img['src']) img['src'] = re.sub(r'w_\d+', 'w_' + res, img['src'])
for src in soup.findAll(['source', 'svg']):
src.extract()
for but in soup.findAll(attrs={'class': ['button-wrapper']}):
but.extract()
return soup return soup