August

August was also marked by the summer holidays. Beside several smaller bugfixes there were feature developments in two areas.

Developments

Crowdsourcing

User-generated content captured in crowdsourcing, i.e. information marked on the image and provided with metadata, is now completely read from the Solr Index for display in the record. This makes the ad independent of the crowdsourcing module itself. Previously, this was only the case for the full texts and transcriptions. This adaptation allows to share the results of user-generated content across multiple instances if they use the same Solr Index.

SEO

The topic search engine optimization continues and the existing functionality was further developed. With the HighWire Press Tags as well as the DublinCore Tags, a hand was once again put on to output more metadata and to mark them up in the DublinCore format even more specifically.

The works listed on the Sitelinks page now always refer to the Bibliographic Data page.

A lot of work has also been done to generate the sitemap, so that it scales well with large databases. The following pages will be added to the sitemap:

  • all CMS pages

  • per record:

    • the page with the first picture or the representative

    • the page "Bibliographic data".

    • the "Table of contents" page

    • all pages with full text

The generation of the sitemap is done by calling the REST interface. The following endpoint is available for this:

/rest/sitemap/update/

In order for search engines to better understand the order of pages, rel="next"- and rel="prev"-links in the header were added to factory pages with paginator.

Cronjob

There are certain tasks that should take place regularly in the Goobi viewer. These have been compiled and combined into a cron file. A template for this can be found in the update instructions.