2014-07-13 08:15:40 +00:00
|
|
|
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.
|
2013-12-06 09:13:03 +00:00
|
|
|
author: ("Benoit Maison")
|
|
|
|
|
test_url: http://www.benoitmaison.org/2011/12/06/why-siri-had-to-start-in-beta/
|
