Blacklight uses Apache Solr as its data index 1, but does not care how you run or index data into your instance. For this workshop, we’re starting with the same set of configuration files that Blacklight uses for its own testing. In this section, we’ll take a look at some commonly used tools and approaches.
solr_wrappersolr_wrapper is a tool that can be used to download, run, and configure a Solr instance, and is often used as part of an application’s test suite to ensure a clean slate.
In a separate terminal window within the application’s directory, run:
% bundle exec solr_wrapper
Solr should now be running on http://localhost:8983/. If we take a look at the solr admin, we can see there’s an empty blacklight-core collection with our preferred solr configuration.
Now that we have Solr up and running, let’s index some data. Blacklight provides some fixture objects (that coincidentally match our current solr schema) and makes it easy to quickly index data:
% bin/rake blacklight:index:seed
Submit another query using the solr admin UI; we should now have 30 documents available.
We can also index some data ourselves. One tool we commonly use to transform data into solr documents is traject. In this example, we’ve pre-configured a traject indexer that handles MARC records in app/models/marc_indexer.rb 2.
The blacklight-marc gem includes some example MARC data (this data is actually the source of the fixture data above).
% find $( bundle show blacklight-marc ) -name "*.mrc"
/path/to/gems/blacklight-marc-7.0.0/test_support/data/test_data.utf8.mrc
% bin/rake solr:marc:index MARC_FILE=/path/to/blacklight-marc-7.0.0/test_support/data/test_data.utf8.mrc
Now that we have some data in Solr, we can query solr and get results back. Try some of these:
These query parameters, and more, are documented in the Solr Reference Guide. Blacklight, for the most part, is just presenting a framework and a basic user interface for querying solr, so it is important to get to know how to manipulate Solr for indexing and querying to provide a good search experience.
Although some institutions have had luck creating adapters to work with ElasticSearch or bespoke search APIs, covering these use cases is outside the scope of this workshop. ↩
The details about the MARC format, traject, and this indexer are outside the scope of this tutorial; for now, it’s enough to know we have a way to get some data into the index. ↩