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

Merge branch 'fix_mediapart_and_bloomberg' of https://github.com/LAntoine/calibre

Browse Source
This commit is contained in:
Kovid Goyal 2024-07-14 08:42:24 +05:30
commit e3237e49ca
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
2 changed files with 16 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
recipes/bloomberg.recipe
View File

@ -207,3 +207,11 @@ class Bloomberg(BasicNewsRecipe):
article.summary = self.tag_to_string(soup.find('div', attrs={'class':'subhead'})) article.summary = self.tag_to_string(soup.find('div', attrs={'class':'subhead'}))
article.text_summary = self.tag_to_string(soup.find('div', attrs={'class':'subhead'})) article.text_summary = self.tag_to_string(soup.find('div', attrs={'class':'subhead'}))
article.title = article.title.replace(' - Bloomberg', '') article.title = article.title.replace(' - Bloomberg', '')
def get_browser(self):
# -- Handle Google's cookies consent page
br = BasicNewsRecipe.get_browser(self)
br.open('https://news.google.com')
br.select_form(action="https://consent.google.com/save")
br.submit()
return br

15
recipes/mediapart.recipe
View File

@ -43,7 +43,7 @@ class Mediapart(BasicNewsRecipe):
] ]
remove_tags = [ remove_tags = [
classes('action-links media--rich read-also login-subscribe print-source_url'), classes('action-links media--rich read-also login-subscribe print-source_url screen-reader-only'),
dict(name='svg'), dict(name='svg'),
] ]
@ -92,17 +92,18 @@ class Mediapart(BasicNewsRecipe):
def get_browser(self): def get_browser(self):
# -- Handle login # -- Handle login
def is_form_login(form):
return "id" in form.attrs and form.attrs['id'] == "logFormEl"
br = BasicNewsRecipe.get_browser(self) br = BasicNewsRecipe.get_browser(self)
if self.username is not None and self.password is not None: if self.username is not None and self.password is not None:
br.open('https://www.mediapart.fr/login') br.open('https://www.mediapart.fr/login')
br.select_form(predicate=is_form_login) br.select_form(nr=0)
br['name'] = self.username br['email'] = self.username
br['password'] = self.password br['password'] = self.password
br.submit() br.submit()
# -- Handle Google's cookies consent page
br.open('https://news.google.com')
br.select_form(action="https://consent.google.com/save")
br.submit()
return br return br
def default_cover(self, cover_file): def default_cover(self, cover_file):