From a84bdaebeb0611031d7fb01ea305fe51fce9dc21 Mon Sep 17 00:00:00 2001 From: Kovid Goyal Date: Sat, 6 Oct 2012 21:55:54 +0530 Subject: [PATCH] Update The Sun UK --- recipes/the_sun.recipe | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) diff --git a/recipes/the_sun.recipe b/recipes/the_sun.recipe index d93ac2c49b..a5c4dbdd10 100644 --- a/recipes/the_sun.recipe +++ b/recipes/the_sun.recipe @@ -8,13 +8,11 @@ class AdvancedUserRecipe1325006965(BasicNewsRecipe): title = u'The Sun UK' description = 'Articles from The Sun tabloid UK' __author__ = 'Dave Asbury' - # last updated 25/7/12 + # last updated 6/10/12 added starsons remove article code language = 'en_GB' oldest_article = 1 - max_articles_per_feed = 12 + max_articles_per_feed = 15 remove_empty_feeds = True - no_stylesheets = True - masthead_url = 'http://www.thesun.co.uk/sol/img/global/Sun-logo.gif' encoding = 'UTF-8' @@ -23,13 +21,9 @@ class AdvancedUserRecipe1325006965(BasicNewsRecipe): - #preprocess_regexps = [ - # (re.compile(r'
', re.IGNORECASE | re.DOTALL), lambda match: '')] - - extra_css = ''' body{ text-align: justify; font-family:Arial,Helvetica,sans-serif; font-size:11px; font-size-adjust:none; font-stretch:normal; font-style:normal; font-variant:normal; font-weight:normal;} - ''' + ''' keep_only_tags = [ dict(name='div',attrs={'class' : 'intro'}), dict(name='h3'), @@ -52,6 +46,15 @@ class AdvancedUserRecipe1325006965(BasicNewsRecipe): (u'Showbiz', u'http://www.thesun.co.uk/sol/homepage/showbiz/rss'), (u'Woman', u'http://www.thesun.co.uk/sol/homepage/woman/rss'), ] +# starsons code + def parse_feeds (self): + feeds = BasicNewsRecipe.parse_feeds(self) + for feed in feeds: + for article in feed.articles[:]: + # print 'article.title is: ', article.title + if 'Web porn harms kids' in article.title.upper() or 'The-Sun-says' in article.url: + feed.articles.remove(article) + return feeds def get_cover_url(self): soup = self.index_to_soup('http://www.politicshome.com/uk/latest_frontpage.html') @@ -69,6 +72,7 @@ class AdvancedUserRecipe1325006965(BasicNewsRecipe): cov2 = str(cov) cov2=cov2[27:-18] #cov2 now is pic url, now go back to original function + # print "**** cov2 =",cov2,"****" br = browser() br.set_handle_redirect(False) try: