Package dk.netarkivet.harvester.datamodel

Class HarvestDBConnection

java.lang.Object

dk.netarkivet.harvester.datamodel.HarvestDBConnection

public final class HarvestDBConnection
extends Object

This class handles connections to the harvest definition database, and also defines basic logic for checking versions of tables.

The statements to create the tables are located in:

• Derby: scripts/sql/createfullhddb.sql
• MySQL: scripts/sql/createfullhddb.mysql
• PostgreSQL: scripts/postgresql/netarchivesuite_init.sql

The implementation relies on a connection pool. Once acquired through the get() method, a connection must be explicitly returned to the pool by calling the release(Connection) method.

THis class is intended to be used statically, and hence cannot be instantiated and is final.

Method Summary

Modifier and Type	Method	Description
static void	cleanup()	Closes the underlying data source.
static void	executeSql(String dbm, String tableName, int version)	Execute the sql to update a given table to a given version.
static void	executeSql(Connection conn, String dbm, String tableName, int version)	Look for the file sql-migration/${dbm}/${tableName}.${version}.sql inside in the harvester-core.jar file.
static Connection	get()	Get a connection to the harvest definition database from the pool.
static String	getDBUrl()	Method for retrieving the url for the harvest definition database.
static void	release(Connection connection)	Helper method to return a connection to the pool.
protected static void	updateTable(String table, int newVersion, String... updates)	Update a table by executing all the statements in the updates String array.
static void	updateTable(Connection c, String table, int newVersion, String... updates)
protected static void	updateTableVersion(String table, int newVersion, String... updates)
static void	updateTableVersion(Connection c, String table, int newVersion)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

get

public static Connection get()

Get a connection to the harvest definition database from the pool. The pool is configured via the following configuration properties:

• @see CommonSettings.DB_POOL_MIN_SIZE
• @see CommonSettings.DB_POOL_MAX_SIZE
• @see CommonSettings.DB_POOL_ACQ_INC

Note that the connection obtained must be returned to the pool by calling release(Connection).

Returns:

a connection to the harvest definition database

Throws:

IOFailure - if we cannot connect to the database (or find the driver).

updateTable

protected static void updateTable(String table,
                                  int newVersion,
                                  String... updates)

Update a table by executing all the statements in the updates String array. If newVersion=1 then the table is created. Note that this method does not make any checks that the SQL statements in the updates parameter actually update or create the correct table.

Parameters:

table - The table to update

newVersion - The version that the table should end up at

updates - The SQL update statements that makes the necessary updates.

Throws:

IOFailure - in case of problems in interacting with the database

updateTable

public static void updateTable(Connection c,
                               String table,
                               int newVersion,
                               String... updates)

updateTableVersion

protected static void updateTableVersion(String table,
                                         int newVersion,
                                         String... updates)

updateTableVersion

public static void updateTableVersion(Connection c,
                                      String table,
                                      int newVersion)

getDBUrl

public static String getDBUrl()

Method for retrieving the url for the harvest definition database. This url will be constructed from the base-url, the machine, the port and the directory. If the database is internal, then only the base-url should have a value.

Returns:

The url for the harvest definition database.

cleanup

public static void cleanup()

Closes the underlying data source.

release

public static void release(Connection connection)

Helper method to return a connection to the pool.

Parameters:

connection - a connection

executeSql

public static void executeSql(String dbm,
                              String tableName,
                              int version)

Execute the sql to update a given table to a given version. The necessary scripts are bundled into the root of the harvester-core.jar in the directory sql-migration. The source of these scripts are in one of these directories: $BASEDIR/deploy/deploy-core/scripts/postgresql/migration/ $BASEDIR/deploy/deploy-core/scripts/mysql/migration/ $BASEDIR/deploy/deploy-core/scripts/derby/migration/ To allow the user to update table 'eav_attribute' in postgresql to version 1 the file $BASEDIR/deploy/deploy-core/scripts/postgresql/migration/eav_attribute.1.sql must exist The postgresql files are during the build-phase put into the sql-migration/postgresql directory The same holds for mysql and derby.

Parameters:

dbm - the type of DBMS (mysql, postgresql,derby)

tableName - The given table to update

version - The table version to update to

executeSql

public static void executeSql(Connection conn,
                              String dbm,
                              String tableName,
                              int version)

Look for the file sql-migration/${dbm}/${tableName}.${version}.sql inside in the harvester-core.jar file.

Parameters:

conn - a valid database connection

dbm - the name of the DBMS used

tableName - the name of the table being updated

version - The new version of the table