Removing duplicate entries/stories from a RSS feed?

[gaikokujinkyofusho]

Hi, I have been enjoying being able to subscribe to RSS
(http://kinja.com/user/thedigestibleaggie) for awhile and have come up
with a fairly nice list of feeds but I have run into an annoying
(though not critical) problem, duplicate stories. Apparently there is
overlap with some of the sites I subscribe to so I get duplicate
stories. Does anyone know of some sort of filter (software or online
service) that can remove duplicate stories? Any help or suggestions
would really be appreciated!

Cheers

-Gaiko

 

[Paul Lutus]

Hi, I have been enjoying being able to subscribe to RSS
(http://kinja.com/user/thedigestibleaggie) for awhile and have come up
with a fairly nice list of feeds but I have run into an annoying
(though not critical) problem, duplicate stories. Apparently there is
overlap with some of the sites I subscribe to so I get duplicate
stories. Does anyone know of some sort of filter (software or online
service) that can remove duplicate stories? Any help or suggestions
would really be appreciated!

Write a script in a language that supports associative arrays (as do Java,
Perl, Ruby, Python, and even JavaScript). Key the associative array to a
unique key created out of elements in the various RSS feed items. Fill the
associative array using the generated key.

Unfortunately, it is rare for two RSS feed items to be truly identical.
Often, they tell the same story with small differences in wording (to avoid
accusations of plagiarism) and of course the URL is normally different.

Without some complex coding to detect items that are almost the same, the
above method will remove only genuinely identical items from different RSS
feeds.

 

[Andy Dingley]

Unfortunately, it is rare for two RSS feed items to be truly identical.

They don't need to be - they have a link URL embedded in them and
that's perfectly adequate for disambiguation.

You might even have a feed with deliberate isPermaLink properties on
each item, for exactly that purpose.