建材 东莞网站建设茌平做网站

现在的情况是，在MySQL中有db1和db2两个数据库。项目使用Hibernate，可同时访问db1和db2，默认数据库为db1。表table2在db2中。且table2的主键名为ids，是自增长字段（Auto Increment）。

table2和ids的定义为：

@Entity
@Table(name = "table2", schema = "db2")
@Cache(usage = CacheConcurrencyStrategy.READ_WRITE)
public class Table2 implements java.io.Serializable {private static final long serialVersionUID = 48L;@Id@Column(name = "ids")@GeneratedValue(generator = "idGenerator", strategy = GenerationType.IDENTITY)@GenericGenerator(name = "idGenerator", strategy = "increment")private Integer ids;

当向table2中保存数据时，会报错。原因是生成ids时，系统会查询table2中ids的最大值。语句是：

select max(ids) from table2

由于默认数据库是db1，因此查询的是db1.table2表。但table2表实际上在db2中，所以系统找不到该表，从而报错。

这就需要在@GenericGenerator中指定max查询语句的schema。通过检查错误信息，发现对应代码在org.hibernate.id.IncrementGenerator.configure()方法中，如下：

	public void configure(Type type, Properties params, Dialect dialect) throws MappingException {returnClass = type.getReturnedClass();ObjectNameNormalizer normalizer =( ObjectNameNormalizer ) params.get( PersistentIdentifierGenerator.IDENTIFIER_NORMALIZER );String column = params.getProperty( "column" );if ( column == null ) {column = params.getProperty( PersistentIdentifierGenerator.PK );}column = dialect.quote( normalizer.normalizeIdentifierQuoting( column ) );String tableList = params.getProperty( "tables" );if ( tableList == null ) {tableList = params.getProperty( PersistentIdentifierGenerator.TABLES );}String[] tables = StringHelper.split( ", ", tableList );final String schema = dialect.quote(normalizer.normalizeIdentifierQuoting(params.getProperty( PersistentIdentifierGenerator.SCHEMA )));final String catalog = dialect.quote(normalizer.normalizeIdentifierQuoting(params.getProperty( PersistentIdentifierGenerator.CATALOG )));StringBuilder buf = new StringBuilder();for ( int i=0; i < tables.length; i++ ) {final String tableName = dialect.quote( normalizer.normalizeIdentifierQuoting( tables[i] ) );if ( tables.length > 1 ) {buf.append( "select max(" ).append( column ).append( ") as mx from " );}buf.append( Table.qualify( catalog, schema, tableName ) );if ( i < tables.length-1 ) {buf.append( " union " );}}if ( tables.length > 1 ) {buf.insert( 0, "( " ).append( " ) ids_" );column = "ids_.mx";}sql = "select max(" + column + ") from " + buf.toString();}

在初始化Table2实体类时，该方法就会执行。作用是生成对应数据库的select max语句。

在IncrementGenerator的注释中，有一段话：

Mapping parameters supported, but not usually needed: tables, column. (The tables parameter specified a comma-separated list of table names.)

说明在@GenericGenerator中，可以设置参数。在IncrementGenerator.configure()方法中，可以将这些参数读出来。读取参数的方法为params.getProperty("参数名")。例如：

params.getProperty( "column" )

就是读取column参数的值。对应读取schema的语句为：

final String schema = dialect.quote(normalizer.normalizeIdentifierQuoting(params.getProperty( PersistentIdentifierGenerator.SCHEMA ))
);

schema的参数名就是PersistentIdentifierGenerator.SCHEMA，也就是"schema"。其他预置保留参数的值大都在org.hibernate.id.PersistentIdentifierGenerator中定义。如：

public interface PersistentIdentifierGenerator extends IdentifierGenerator {/*** The configuration parameter holding the schema name*/public static final String SCHEMA = "schema";/*** The configuration parameter holding the table name for the* generated id*/public static final String TABLE = "target_table";/*** The configuration parameter holding the table names for all* tables for which the id must be unique*/public static final String TABLES = "identity_tables";/*** The configuration parameter holding the primary key column* name of the generated id*/public static final String PK = "target_column";/*** The configuration parameter holding the catalog name*/public static final String CATALOG = "catalog";/*** The key under whcih to find the {@link org.hibernate.cfg.ObjectNameNormalizer} in the config param map.*/public static final String IDENTIFIER_NORMALIZER = "identifier_normalizer";

在@GenericGenerator中设置parameter的方法为：

@Id
@Column(name = "ids")
@GeneratedValue(generator = "idGenerator", strategy = GenerationType.IDENTITY)
@GenericGenerator(name = "idGenerator", strategy = "increment", parameters = @Parameter(name = PersistentIdentifierGenerator.SCHEMA, value = "db2"))
private Integer ids;

这样就能在table2前加上正确的schema名称（db2），生成正确的查询语句：

select max(ids) from db2.table2

如果有多个参数，可以写为：

@GenericGenerator(name = "idGenerator", strategy = "increment", parameters = {
            @Parameter(name = PersistentIdentifierGenerator.SCHEMA, value = "db2"),
            @Parameter(name=PersistentIdentifierGenerator.CATALOG, value = "db2")

})