Load RDF into Solr-cloud for a SearchWorks-preview-dev app

[dazza-codes]

See similar project from Cornell in the LD4L-1 phase:

• https://github.com/ld4l/ld4l_browser_data/tree/master/lib/ld4l_browser_data/indexing
• https://github.com/ld4l/ld4l_blacklight_search

Solr reference materials:

• https://cwiki.apache.org/confluence/display/solr/Introduction+to+Solr+Indexing
• https://cwiki.apache.org/confluence/display/solr/Indexing+and+Basic+Data+Operations

Related projects in DLSS:

• https://github.com/sul-dlss/sul-solr-configs
  • sw-preview-prod collection could be a good starting point
  • searchworks-rdf-dev will be our new collection, see https://github.com/sul-dlss/sul-solr-configs/pull/93
• https://github.com/sul-dlss/solrmarc-sw
  • generates Solr-doc data that is not compatible with Solr-cloud
  • https://github.com/sul-dlss/searchworks_marc_indexing
  • start of a port of solrmarc-sw to use traject (see below)
  • https://github.com/traject/traject - for Solr-doc data compatible with Solr-cloud
• https://github.com/sul-dlss?utf8=%E2%9C%93&q=index
• https://github.com/sul-dlss/triannon/blob/master/app/services/triannon/solr_writer.rb
  • @ndushay worked on getting triannon RDF into Solr
• https://github.com/sul-dlss/dor_indexing_app
  • intermediary to Solr API
  • uses background jobs

[dazza-codes]

@eefahy and @cbeer has some suggestions on using ldpath from Marmotta:

• http://marmotta.apache.org/ldpath/

That project recommends using Apache Stanbol for semantic search:

• https://stanbol.apache.org/

[dazza-codes]

The json-ld format might be well suited to converting RDF into a Solr-doc. If the json-ld fields can use the Solr-schema fields (and the json-ld context can define these fields OK), then the body of the json-ld represention is the Solr-doc that can be input to Solr-cloud.

[dazza-codes]

Adding assignee to flag folks who can input to some of this, not to demand work on it from anyone in particular.