diff --git a/recipes/readers_digest.recipe b/recipes/readers_digest.recipe index 1055aaf00a..760432af7c 100644 --- a/recipes/readers_digest.recipe +++ b/recipes/readers_digest.recipe @@ -7,6 +7,12 @@ import re from calibre.web.feeds.recipes import BasicNewsRecipe +def classes(classes): + q = frozenset(classes.split(' ')) + return dict(attrs={ + 'class': lambda x: x and frozenset(x.split()).intersection(q)}) + + class ReadersDigest(BasicNewsRecipe): title = 'Readers Digest' @@ -20,6 +26,7 @@ class ReadersDigest(BasicNewsRecipe): language = 'en' remove_javascript = True + remove_attributes = ['style'] extra_css = ''' h1 {font-family:georgia,serif;color:#000000;} .mainHd{font-family:georgia,serif;color:#000000;} @@ -51,7 +58,10 @@ class ReadersDigest(BasicNewsRecipe): # cover_url = 'http://www.rd.com/images/logo-main-rd.gif' - keep_only_tags = dict(id='main') + keep_only_tags = [ + classes('entry-header rd-article-dek'), + dict(name='article', attrs={'class': lambda x: x and 'post-' in x}), + ] remove_tags = [ dict(attrs={'class': re.compile('rd-article--sidebar.*')}), dict(attrs={'class': ['rd-article--footer', 'rd-article--sharing']}),