Obtaining IndexReaders

As we have seen already, there are several prerequisites (for example, opening a directory, setting up an analyzer, and writing to index) to prepare an index for a search. Lucene provides the IndexReader class to access a point-in-time view of an index. It means that you can concurrently write to an index, while an existing IndexReader is reading without exposing any uncommitted data to the active IndexReader. This is an important concept to keep in mind because this architecture allows the possibility of providing a seamless transition between index versions by opening a new IndexReader while the old IndexReader is still servicing a search. The DirectoryReader is a subclass of IndexReader, which is the class that provides the facility to actually open a directory containing an index, and returns an IndexReader. DirectoryReader also has a more optimized index opening method called openIfChanged that will reuse the existing DirectoryReader for faster reopening of an...