Crawl all JSON-LD via the Omeka S API to make a compressed N-triple file for import.
Run php composer.phar update to get the EasyRDF module (which is installed in the vendor directory).
More info about PHP Composer: https://getcomposer.org/
crawler.php crawls all Omeka S items (and related resources) in JSON-LD format. The JSON-LD is converted to N-triples and stored in the nt directory. The hashes directory is home for the MD5 hashes of the N-triple files and are used to skip the unnecessary conversion if there are no changes in contents.
If the ARK module is used and dcterms:identifier triples are found, then an additional owl:sameAs triple is added to 'connect' a https://n2t.net/ark:/99999/xxxxxx like URI to the Omeka S API URI.
The Bash-script convert.sh collects all N-triple files from the nt directory, sorts them (making lines uniq) and Gzip's the resulting file.
Crawling all items can take hours, conversion only minutes.
A Omeka S module which makes the N-triple file on each item creation/update and deletes the N-triple file on item delete would be more efficient.
For shorter (re)crawls the $next=$omeka_base.'/api/items?per_page=100&sort_by=id&sort_order=asc&page=1'; can be adjusted to narrow down the crawl to a specific item site.