Simple Lucene Abstractions

  <dependency>
      <groupId>io.github.iamnicknack.simple-lucene</groupId>
      <artifactId>simple-lucene-annotations</artifactId>
      <version>0.1.3</version>
  </dependency>

Up to date dependency details are available from https://mvnrepository.com/artifact/io.github.iamnicknack.simple-lucene

What this is

• An opinionated, low-level abstraction of Lucene which aims to simplify resource management and query execution.
• Facilitates modelling of domain types and programmatically define the structure of the Lucene index.

What this is not

• NOT a dedicated persistence layer.
• Hibernate provides a full-featured ORM.

What are these abstractions for?

If you come to Lucene from any of the well known distributions such as Elastic or Solr, the barrier to entry can seem quite high.

• Lucene requires us to pay close attention to resource usage when using searchers, readers, etc writers so as to avoid resource leakage.
• We need to choose appropriate field types and maintain instances of them correctly when updating the index in order for Lucene to do the work we’re expecting it to do at query time.
• Attention needs to be paid to the details of indexing data such as when to commit or updating taxonomy for a document.
• …

This is not an exhaustive list. We may not be experts on Lucene (yet) and the list of details for the developer to remember to account for each time Lucene is brought into a project will probably grow over time.

Re-writing features to support these types of scenarios each time an application wants to use Lucene might be a waste of development time and resources. By accepting some opinionated choices around field names and types, we can negate the need to duplicate code each time we build an application with Lucene.

Documentation and examples

Documentation is provided along with working examples in the examples/documentation module