Sometimes when adding resources to CKAN, users mistakenly add a link to the page containing links to the data rather than the data itself.
This extension attempts, for a given resource to find the actual resources on the page. This will only cover some cases and isn’t a solution that will fix every problem.
By using an assortment of CSS, XPath and Regex on a per url (or per publisher) basis it is possible to find the important links in a HTML page.
To install the extension …
Running the following command in the ckanext-linkfinder directory will run all of the available tests.