Code:
import re
class AdvancedUserRecipe1234495609(BasicNewsRecipe):
title = u'Physicsworld'
oldest_article = 7
max_articles_per_feed = 100
no_stylesheets = True
use_embedded_content = False
remove_javascript = True
remove_tags_before = dict(name='h1')
remove_tags_after = [dict(name='div', attrs={'id':'shareThis'})]
preprocess_regexps = [
(re.compile(r'<div id="shareThis">.*</body>', re.DOTALL|re.IGNORECASE),
lambda match: '</body>'),
]
feeds = [
(u'Headlines News', u'http://feeds.feedburner.com/PhysicsWorldNews')
]
Note that to ensure that calibre can get all the article, you need to login. Making a custom recipe with login is, however, beyond my skill.