from calibre.web.feeds.news import BasicNewsRecipe
import re

class ReutersJa(BasicNewsRecipe):

    title = 'Reuters(Japan)'
    description = 'Global news in Japanese'
    __author__ = 'Hiroshi Miura'
    use_embedded_content   = False
    language = 'ja'
    max_articles_per_feed = 10
    remove_javascript = True

    feeds = [ ('Top Stories', 'http://feeds.reuters.com/reuters/JPTopNews?format=xml'),
                  ('World News', 'http://feeds.reuters.com/reuters/JPWorldNews?format=xml'),
                  ('Business News', 'http://feeds.reuters.com/reuters/JPBusinessNews?format=xml'),
                  ('Technology News', 'http://feeds.reuters.com/reuters/JPTechnologyNews?format=xml'),
                  ('Oddly Enough News', 'http://feeds.reuters.com/reuters/JPOddlyEnoughNews?format=xml')
         ]

    remove_tags_before = {'class':"article primaryContent"}
    remove_tags = [ dict(id="banner"),
                    dict(id="autilities"),
                    dict(id="textSizer"),
                    dict(id="shareFooter"),
                    dict(id="relatedNews"),
                    dict(id="editorsChoice"),
                    dict(id="ecArticles"),
                    {'class':"secondaryContent"},
                    {'class':"module"},
                     ]
    remove_tags_after = {'class':"assetBuddy"}

    def print_version(self, url):
        m = re.search('(.*idJPJAPAN-[0-9]+)', url)
        return m.group(0)+'?sp=true'