How will we actually search for OMR things with solr

Understanding how we'll be using solr will inform how we need to parse and prepare the data from the mei files.

We could add a Neume model, whose fields include folio, manuscript, and string versions of lists of pitches, contours, intervals etc, along with some id we can use to get unsearchable data like regions. Or we could somehow attach the data at a higher level, the chant, folio or manuscript model. How we do it will necessitate different forms of organizing the data.

Search results are often returned as a list of chants, so mapping neumes to chants when parsing the mei files could be good.

It's also possible that solr has searching capabilities that could simplify the process for matching patterns across neumes. The website mentions "grouping" and "joins", that sounds maybe useful.

Start by examining how MEI data is currently parsed and how Solr search results are represented. Compare whether neumes should be attached to neumes, chants, folios, or manuscripts, including how chant-level results would map back to neumes. Done means documenting a decided data organization and search approach.

Written by the indexing model from the issue text.