This is the main (end-user) documentation for the PostgreSQL Full Text Search module, explaining how this module works, how to setup and manage your FTS indexes, how to manage your database indexes and how to use Views to leverage the power of full text within your Drupal installation.
Note: Since the module is still in early stages of development, consider these pages to shift as new features are added, bugs are fixed or simply functionality is optimized.
Before we dive into a hands-on approach, a bit of background:
For those who haven't read the module's main page, PostgreSQL Full Text Search is an honest attempt to bring a native feature of the PostgreSQL RDBMS - professional full text search - into Drupal.
Born from wanting a more lightweight and easier to setup solution to Solr (and Tomcat, Java, ...) and from the (somewhat) sane reasoning that if one already has software on their server that is capable of full text searching (and that software happens to be your very own database you run Drupal on) you do not need yet another software chain contaminating your machine.
It is built around the idea that the content one wishes to index is mostly located in fields that are attached to entities (leveraging the Field API CRUD). Thus it gives you the ability to index the content(s) of any field (preferably text based fields) that you have stored on your entities (any entity).
A word about terminology used throughout the documentation (and inside the module):
When I refer to an "FTS index", I am talking about the stuff that PostgreSQL creates after crunching the data available in your chosen field. All full text solutions need to take your "normal" human readable text data and convert it to some other form to be able to perform full text search operations on it. PostgreSQL is no exception to that. To read in more detail about what is happing when you hit the "Index remaining items" button, read the "What is indexing" chapter.
Yet when I say "Database index", I am mumbling about an actual Gist or Gin index that can be made on top of the "FTS index" so PostgreSQL's planner can use that index to significantly speed up full text searches.
Good ... enough formalities, let us get this party started.