File content is not being extracted and included in the search indexes

When a file is uploaded into the repository (or projected via the file system connector), the "jcr:data" binary value of the "jcr:content" node is being ignored by the Lucene search engine. The content of the file should be run through a text extractor (such as Tika) to pull out the terms that should be indexed.