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

News download: Remove all link and meta tags

Browse Source 
These can have various ill effects
This commit is contained in:
Kovid Goyal 2021-08-13 20:44:08 +05:30
parent 72bfb00300
commit bb71434797
No known key found for this signature in database
GPG Key ID: 06BC317B515ACE7C
1 changed files with 6 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
src/calibre/web/feeds/news.py
View File

@ -1007,14 +1007,15 @@ class BasicNewsRecipe(Recipe):
for o in soup.findAll(onload=True): for o in soup.findAll(onload=True):
del o['onload'] del o['onload']
for script in list(soup.findAll('noscript')):
script.extract()
for attr in self.remove_attributes: for attr in self.remove_attributes:
for x in soup.findAll(attrs={attr:True}): for x in soup.findAll(attrs={attr:True}):
del x[attr] del x[attr]
for base in list(soup.findAll(['base', 'iframe', 'canvas', 'embed', for bad_tag in list(soup.findAll(['base', 'iframe', 'canvas', 'embed',
'command', 'datalist', 'video', 'audio'])): 'command', 'datalist', 'video', 'audio', 'noscript', 'link', 'meta'])):
base.extract() # link tags can be used for preloading causing network activity in
# calibre viewer. meta tags can do all sorts of crazy things,
# including http-equiv refresh, viewport shenanigans, etc.
bad_tag.extract()
# srcset causes some viewers, like calibre's to load images from the # srcset causes some viewers, like calibre's to load images from the
# web, and it also possible causes iBooks on iOS to barf, see # web, and it also possible causes iBooks on iOS to barf, see
# https://bugs.launchpad.net/bugs/1713986 # https://bugs.launchpad.net/bugs/1713986