How to catch XML's attribute of elements in PKP Harvester?

[sophonie]

Hi,
In PKP Harvester this url http://harvester.aquitaine.fr/index.php/record/view/268
the fields locinsee and Références géospatiales are empty but in the OAI Repository
a value is in. In fact this value is an attribute of these elements so that
<ap:locinsee code_insee="33254"/>
<ap:spatial latitude="0.425698" longitude="32.14587"/>

I want show these values in PKP Harvester. But How to configure PKP for catch theses attributes?
Thanks.
Sophonie

[asmecher]

Hi Sophonie,

Are you handy with MySQL? I'd suggest checking the records table to see what data is there. Find the record_id from in URL while browsing one of the records in question, then use the record_id (e.g. 12345 below) to query the records table:

Code: Select all

SELECT contents FROM records WHERE record_id = 12345;

If you see the missing data there, it's been harvested OK but is probably not being processed. Next check the parsed_contents field:

Code: Select all

SELECT parsed_contents FROM records WHERE record_id = 12345;

If the data appears in the first field but not the second, you'll need to look at the DublinCorePlugin (implemented in plugins/schemas/dc/DublinCorePlugin.inc.php). Specifically, look at the parseContents function, which turns the raw XML (as contained in "contents") into a more compact format that's used for presenting the data (as contained in "parsed_contents").

Regards,
Alec Smecher
Public Knowledge Project Team