blob: f341d59337bf89d64f8abd64d2cfbd28b4688cf9 (
plain) (
tree)
|
|
body: //div[@class="entry-content"]
# Remove text ‘Tweet’
strip: //div[@class="entry-content"]/div[last()]
title: h1[@class="entry-title"]
# If the Instapaper text parser worked with HTML5 tags, we would use:
date: //time[@class="entry-date"]
# But since it does not, use this more complicated rule:
date: //div[@class="entry-meta"]/a[@rel="bookmark"]
# Unfortunately, the following rule is overridden by the automatically found author.
author: ("Benoit Maison")
test_url: http://www.benoitmaison.org/2011/12/06/why-siri-had-to-start-in-beta/
|
