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

Cleanup tags for removal

Browse Source
This commit is contained in:
a10kiloham 2018-07-18 12:43:29 +01:00 committed by GitHub
parent 889f07dcdc
commit b25718deca
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
recipes/times_online.recipe
View File

@ -8,13 +8,11 @@ import html5lib
from lxml import html from lxml import html
from calibre.web.feeds.news import BasicNewsRecipe from calibre.web.feeds.news import BasicNewsRecipe
def classes(classes): def classes(classes):
q = frozenset(classes.split(' ')) q = frozenset(classes.split(' '))
return dict( return dict(
attrs={'class': lambda x: x and frozenset(x.split()).intersection(q)}) attrs={'class': lambda x: x and frozenset(x.split()).intersection(q)})
class TimesOnline(BasicNewsRecipe): class TimesOnline(BasicNewsRecipe):
title = 'The Times & Sunday Times (UK)' title = 'The Times & Sunday Times (UK)'
__author__ = 'Bobby Steel' __author__ = 'Bobby Steel'
@ -87,12 +85,9 @@ class TimesOnline(BasicNewsRecipe):
return br return br
remove_tags = [ remove_tags = [
{'attrs': {'class': lambda x: x and 'is-hidden' in x}}, classes('is-hidden Toolbar Tooltip Topics Comments u-hide RelatedLinks ArticlePager Media-caption'),
{'attrs': {'class': lambda x: x and 'Toolbar' in x}},
{'attrs': {'class': lambda x: x and 'Tooltip' in x}},
{'attrs': {'class': lambda x: x and 'Topics' in x}},
{'name': ['object', 'link', 'iframe', 'base', 'meta', 'script']}, {'name': ['object', 'link', 'iframe', 'base', 'meta', 'script']},
{'attrs': {'class': ['u-hide', 'RelatedLinks', 'Comments Article-container','ArticlePager','Media-caption']}}] ]
remove_attributes = ['lang'] remove_attributes = ['lang']
keep_only_tags = [{ keep_only_tags = [{