FullText Indexes

The SB-Tree index algorithm provides support for FullText indexes. These indexes allow you to index text as a single word and its radix. FullText indexes are like having a search engine on your database.

NOTE: Bear in mind that there is a difference between FULLTEXT without the LUCENE operator, which uses a FullText index with the SB-Tree index algorithm and FULLTEXT with the LUCENE operator, which uses a FullText index through the Lucene Engine.

For more information on the latter, see Lucene FullText Index.

Creating FullText Indexes

If you want to create an index using the FullText SB-Tree index algorithm, you can do so using the CREATE INDEX command.

  1. orientdb> CREATE INDEX City.name ON City(name) FULLTEXT

This creates a FullText index on the property name of the class City, using the default configuration.

FullText Index Parameters

In the event that the default FullText Index configuration is not sufficient to your needs, there are a number of parameters available to fine tune how it generates the index.

Parameter Default Description
indexRadix TRUE Word prefixes will be also index
ignoreChars " Chars to skip when indexing
separatorChars \r\n\t:;,.|+*/\=!?[](.md)
minWordLength 3 Minimum word length to index
stopWords the in a at as and or for his her him this that what which while up with be was were is Stop words escluded from indexing

To configure a FullText Index, from version 1.7 on, you can do so through the OrientDB console or the Java API. When configuring the index from the console, use the CREATE INDEX command with the METADATA operator.

  1. orientdb> CREATE INDEX City.name ON City(name) FULLTEXT METADATA 
  2.           {"indexRadix": true, "ignoreChars": "&", "separatorChars": " |()", 
  3.           "minWordLength": 4, "stopWords": ["the", "of"]}

Alternatively, you can configure the index in Java.

  1. OSchema schema = db.getMetadata().getSchema();
  2. OClass city = schema.getClass("City");
  3. ODocument metadata = new ODocument();
  4. metadata.field("indexRadix", true);
  5. metadata.field("stopWords", Arrays.asList(new String[] { "the", "in", "a", "at" }));
  6. metadata.field("separatorChars", " :;?[](.md)");
  7. metadata.field("ignoreChars", "$&");
  8. metadata.field("minWordLength", 5);
  9. city.createIndex("City.name", "FULLTEXT", null, metadata, null, new String[] { "name" });