Package org.h2.fulltext

Class FullTextLucene

java.lang.Object
org.h2.fulltext.FullText
org.h2.fulltext.FullTextLucene
public class FullTextLucene extends FullText
This class implements the full text search based on Apache Lucene. Most methods can be called using SQL statements as well.

  • Field Details

    • STORE_DOCUMENT_TEXT_IN_INDEX

      protected static final boolean STORE_DOCUMENT_TEXT_IN_INDEX
      Whether the text content should be stored in the Lucene index.

  • Constructor Details

    • FullTextLucene

      public FullTextLucene()

  • Method Details

    • init

      public static void init(Connection conn) throws SQLException
      Initializes full text search functionality for this database. This adds the following Java functions to the database:
      • FTL_CREATE_INDEX(schemaNameString, tableNameString, columnListString)
      • FTL_SEARCH(queryString, limitInt, offsetInt): result set
      • FTL_REINDEX()
      • FTL_DROP_ALL()
      It also adds a schema FTL to the database where bookkeeping information is stored. This function may be called from a Java application, or by using the SQL statements: 
       CREATE ALIAS IF NOT EXISTS FTL_INIT FOR
      "org.h2.fulltext.FullTextLucene.init";
 CALL FTL_INIT();
      Parameters:
      conn - the connection
      Throws:
      SQLException - on failure

    • createIndex

      public static void createIndex(Connection conn, String schema, String table, String columnList) throws SQLException
      Create a new full text index for a table and column list. Each table may only have one index at any time.
      Parameters:
      conn - the connection
      schema - the schema name of the table (case sensitive)
      table - the table name (case sensitive)
      columnList - the column list (null for all columns)
      Throws:
      SQLException - on failure

    • dropIndex

      public static void dropIndex(Connection conn, String schema, String table) throws SQLException
      Drop an existing full text index for a table. This method returns silently if no index for this table exists.
      Parameters:
      conn - the connection
      schema - the schema name of the table (case sensitive)
      table - the table name (case sensitive)
      Throws:
      SQLException - on failure

    • reindex

      public static void reindex(Connection conn) throws SQLException
      Re-creates the full text index for this database. Calling this method is usually not needed, as the index is kept up-to-date automatically.
      Parameters:
      conn - the connection
      Throws:
      SQLException - on failure

    • dropAll

      public static void dropAll(Connection conn) throws SQLException
      Drops all full text indexes from the database.
      Parameters:
      conn - the connection
      Throws:
      SQLException - on failure

    • search

      public static ResultSet search(Connection conn, String text, int limit, int offset) throws SQLException
      Searches from the full text index for this database. The returned result set has the following column:
      • QUERY (varchar): the query to use to get the data. The query does not include 'SELECT * FROM '. Example: PUBLIC.TEST WHERE ID = 1
      • SCORE (float) the relevance score as returned by Lucene.
      Parameters:
      conn - the connection
      text - the search query
      limit - the maximum number of rows or 0 for no limit
      offset - the offset or 0 for no offset
      Returns:
      the result set
      Throws:
      SQLException - on failure

    • searchData

      public static ResultSet searchData(Connection conn, String text, int limit, int offset) throws SQLException
      Searches from the full text index for this database. The result contains the primary key data as an array. The returned result set has the following columns:
      • SCHEMA (varchar): the schema name. Example: PUBLIC
      • TABLE (varchar): the table name. Example: TEST
      • COLUMNS (array of varchar): comma separated list of quoted column names. The column names are quoted if necessary. Example: (ID)
      • KEYS (array of values): comma separated list of values. Example: (1)
      • SCORE (float) the relevance score as returned by Lucene.
      Parameters:
      conn - the connection
      text - the search query
      limit - the maximum number of rows or 0 for no limit
      offset - the offset or 0 for no offset
      Returns:
      the result set
      Throws:
      SQLException - on failure

    • convertException

      protected static SQLException convertException(Exception e)
      Convert an exception to a fulltext exception.
      Parameters:
      e - the original exception
      Returns:
      the converted SQL exception

    • getIndexAccess

      protected static org.h2.fulltext.FullTextLucene.IndexAccess getIndexAccess(Connection conn) throws SQLException
      Get the index writer/searcher wrapper for the given connection.
      Parameters:
      conn - the connection
      Returns:
      the index access wrapper
      Throws:
      SQLException - on failure

    • getIndexPath

      protected static String getIndexPath(Connection conn) throws SQLException
      Get the path of the Lucene index for this database.
      Parameters:
      conn - the database connection
      Returns:
      the path
      Throws:
      SQLException - on failure

    • removeIndexAccess

      protected static void removeIndexAccess(String indexPath) throws SQLException
      Close the index writer and searcher and remove them from the index access set.
      Parameters:
      indexPath - the index path
      Throws:
      SQLException - on failure

    • search

      protected static ResultSet search(Connection conn, String text, int limit, int offset, boolean data) throws SQLException
      Do the search.
      Parameters:
      conn - the database connection
      text - the query
      limit - the limit
      offset - the offset
      data - whether the raw data should be returned
      Returns:
      the result set
      Throws:
      SQLException - on failure