Message: 5
Date: Fri, 18 May 2012 16:45:47 +0100
From: Tim Brody <tdb2@ecs.soton.ac.uk>
Subject: [EP-tech] Re: query regarding compatibility to Zotero
To: eprints-tech@ecs.soton.ac.uk
Message-ID: <1337355947.2325.77.camel@chassis.ecs.soton.ac.uk">1337355947.2325.77.camel@chassis.ecs.soton.ac.uk>
Content-Type: text/plain; charset="utf-8"

On Fri, 2012-05-18 at 16:36 +0100, Tim Brody wrote:

On Thu, 2012-05-17 at 13:19 +0100, Hugh O Neill wrote:

Hi there,

I'm trying to get our prints repository (doras.dcu.ie) to be read by a

citation manager, Zotero.

It does not seem to be translated.

I enquired from the Zotero team as to whether a translator could be

developed by them to read eprints. This is their response:

"Unfortunately, as far as I can tell, there is no standard for eprints

metadata. ......  If there is a standard of any sort, or if you can

find a link to the default fields that ship with EPrints, we'd be glad

to support it."

I've looked through the eprints site, but I can't find the relevant

documentation.

Could someone point me in the right direction with this?

Zotero are correct that every EPrints install has its own schema -

although most are fairly close to the default configuration. Therefore

reading the raw EPrints data is probably not useful.

What does Zotero look for in the page?

Basic metadata can be extracted from the <meta> tags in the page header.

Or, it could read from one of the Export plugins via the given <link>s

in the header.

(3.3 also sets a <meta generator> which makes it easier for scrapers to

recognise EPrints pages)

Replying to myself ... I think Zotero will spot a "COinS". In
eprint_render add something like this (untested):

my $plugin = $session->plugin( "Export::COinS" );
my $uri = $plugin->output_dataobj( $eprint );
$page->appendChild( $session->make_element( "span",
class => "Z3988",
title => $uri
) );

But it looks like it ought to spot <meta> for Dublin Core too:
http://www.zotero.org/support/dev/exposing_metadata

Perhaps you need to ask Zotero why it doesn't spot the existing DC
metadata?

/Tim.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 490 bytes
Desc: This is a digitally signed message part
Url : http://mailman.ecs.soton.ac.uk/pipermail/eprints-tech/attachments/20120518/91ab9036/attachment-0001.bin