如何在java中调用包含用户定义类型的oracle存储过程?

时间:2010-09-02 10:08:27

标签: java oracle stored-procedures jdbc

在Oracle DB中:

我有以下存储过程:

procedure getInfo ( p_ids IN IDS_TABLE, p_details OUT cursor )

类型IDS_TABLE是:

create or replace type IDS_TABLE as table of IDS    

create or replace type IDS as object ( id1 NUMBER, id2 NUMBER, id3 NUMBER )

如何在Java中调用getInfo?

4 个答案:

答案 0 :(得分:36)

手动设置Oracle SQL对象和Java对象之间的链接并非易事。特别是,用户定义对象的数组(或嵌套表)比标准数据类型的数组从Java传递到Oracle更复杂。换句话说,使用签名调用过程更容易:

(TABLE OF NUMBER, TABLE OF NUMBER, TABLE OF NUMBER)`

比签名为:

的程序
(TABLE OF (NUMBER, NUMBER, NUMBER))   <- your case

您可以在程序周围编写一个包装器,将第二种情况转换为第一种情况。


话虽如此,映射你的程序到目前为止并非不可能。以下示例为largely inspired by a post by Tom Kyte。 Tom描述了如何使用TABLE OF NUMBER映射oracle.sql.ARRAY。在您的情况下,我们还必须使用oracle.sql.STRUCT来映射IDS SQL对象。

您可能还想浏览Oracle JDBC文档,特别是章节Working with Oracle Object Types

首先是类似于你的设置:

SQL> CREATE OR REPLACE TYPE IDS AS OBJECT ( id1 NUMBER, id2 NUMBER, id3 NUMBER );
  2  /
Type created

SQL> CREATE OR REPLACE TYPE IDS_TABLE AS TABLE OF IDS;
  2  /
Type created

SQL> CREATE OR REPLACE PROCEDURE getInfo(p_ids IN IDS_TABLE) IS
  2  BEGIN
  3     FOR i IN 1 .. p_ids.COUNT LOOP
  4        dbms_output.put_line(p_ids(i).id1
  5                             || ',' || p_ids(i).id2
  6                             || ',' || p_ids(i).id3);
  7     END LOOP;
  8  END getInfo;
  9  /     
Procedure created

这是java程序:

SQL> CREATE OR REPLACE
  2  AND COMPILE JAVA SOURCE NAMED "ArrayDemo"
  3  as
  4  import java.io.*;
  5  import java.sql.*;
  6  import oracle.sql.*;
  7  import oracle.jdbc.driver.*;
  8  
  9  public class ArrayDemo {
 10  
 11     public static void passArray() throws SQLException {
 12  
 13        Connection conn =
 14           new OracleDriver().defaultConnection();
 15  
 16  
 17        StructDescriptor itemDescriptor =
 18           StructDescriptor.createDescriptor("IDS",conn);
 19  
 20        Object[] itemAtributes = new Object[] {new Integer(1),
 21                                               new Integer(2),
 22                                               new Integer(3)};
 23        STRUCT itemObject1 = new STRUCT(itemDescriptor,conn,itemAtributes);
 24  
 25        itemAtributes = new Object[] {new Integer(4),
 26                                      new Integer(5),
 27                                      new Integer(6)};
 28        STRUCT itemObject2 = new STRUCT(itemDescriptor,conn,itemAtributes);
 29  
 30        STRUCT[] idsArray = {itemObject1,itemObject2};
 31  
 32        ArrayDescriptor descriptor =
 33           ArrayDescriptor.createDescriptor( "IDS_TABLE", conn );
 34  
 35        ARRAY array_to_pass =
 36           new ARRAY( descriptor, conn, idsArray );
 37  
 38        OraclePreparedStatement ps =
 39           (OraclePreparedStatement)conn.prepareStatement
 40           ( "begin getInfo(:x); end;" );
 41  
 42        ps.setARRAY( 1, array_to_pass );
 43        ps.execute();
 44  
 45     }
 46  }
 47  /
Java created

我们称之为:

SQL> CREATE OR REPLACE
  2  PROCEDURE show_java_calling_plsql
  3  AS LANGUAGE JAVA
  4  NAME 'ArrayDemo.passArray()';
  5  /
Procedure created

SQL> exec show_java_calling_plsql ;
1,2,3
4,5,6

PL/SQL procedure successfully completed

答案 1 :(得分:1)

如果您正在使用 Spring ,则可能需要查看提供SqlArrayValue类型的Spring Data JDBC Extensions

7.2.1 Setting ARRAY values using SqlArrayValue for an IN parameter章解释了如何使用数组参数调用过程。

答案 2 :(得分:0)

这是一个很好的例子。如果你看到 java.sql.SQLException:无效的名称模式:。检查您在Oracle中声明的类型的范围。我正在使用Oracle 11g,并且必须在模式级别声明我的类型的字符串数组和对象表。花了大约3个小时才发现。

oracle.sql.StructDescriptor docObjDescriptor = StructDescriptor.createDescriptor("SSIADM.DOCUMENT_OBJECT",conn);
  String[] strArray = new String[] {"doc1","file1"};             
  oracle.sql.STRUCT DocObject1 = new STRUCT(docObjDescriptor,conn,strArray);

   strArray = new String[] {"doc2","file2"};
   oracle.sql.STRUCT DocObject2 = new STRUCT(docObjDescriptor,conn,strArray);

    oracle.sql.STRUCT[] docObjArray = {DocObject1,DocObject2};

    arrDesc = ArrayDescriptor.createDescriptor("DOCUMENT_TABLE", conn);
    oracle.sql.ARRAY array = new ARRAY(arrDesc, conn, docObjArray);

答案 3 :(得分:0)

我使用的解决方案让Spring解析对象而不必手动创建STRUCT数组。 不幸的是,它仍然不依赖于环境。

存储过程DAO:

package ****.dao.storedProcedures;

import java.sql.Array;
import java.sql.Connection;
import java.sql.PreparedStatement;
import java.sql.SQLException;
import java.sql.Types;
import java.util.Date;
import java.util.HashMap;
import java.util.Map;

import org.apache.commons.lang3.Validate;
import org.apache.commons.logging.Log;
import org.apache.commons.logging.LogFactory;
import org.springframework.beans.factory.InitializingBean;
import org.springframework.jdbc.core.SqlOutParameter;
import org.springframework.jdbc.core.SqlParameter;
import org.springframework.jdbc.core.SqlTypeValue;
import org.springframework.jdbc.object.StoredProcedure;

import ****.persistent.ComplexTypeObj;
import ****.persistent.InnerType;
import oracle.sql.ARRAY;
import oracle.sql.ArrayDescriptor;

public class SaveStoredProc extends StoredProcedure implements InitializingBean {

    public static final String IT_COMPLEX_TYPE = "it_complex_type";

    public SaveStoredProc() {
    }

    @Override
    public void afterPropertiesSet() {
        Validate.notNull(getJdbcTemplate());
        super.setFunction(true);
        super.declareParameter(new SqlOutParameter(RESULT, Types.NUMERIC));
        super.declareParameter(new SqlParameter(IT_COMPLEX_TYPE, Types.OTHER, ComplexTypeObj.ORACLE_OBJECT_NAME));
        compile();
    }

    public long execute(final ComplexTypeObj complexTypeObj) {
        Map<String, Object> inParameters = new HashMap<String, Object>();
        inParameters.put(IT_COMPLEX_TYPE, new ComplexSqlTypeValue(complexTypeObj));

        @SuppressWarnings("unchecked")
        Map<String, Object> resp = super.execute(inParameters);

        return ((Number)resp.get(RESULT)).longValue();
    }

    private static final class ComplexSqlTypeValue implements SqlTypeValue {
        private final Log logger = LogFactory.getLog(getClass());

        private final ComplexTypeObj complexTypeObj;

        public ComplexSqlTypeValue(ComplexTypeObj complexTypeObj) {
            this.complexTypeObj = complexTypeObj;
        }

        @Override
        public void setTypeValue(PreparedStatement ps, int paramIndex, int sqlType, String typeName) throws SQLException {
            Connection conn = ps.getConnection();
            try {
                conn = conn.unwrap(oracle.jdbc.OracleConnection.class);
            } catch (Exception e) {
                logger.debug("Could not unrap connection");
            }

            Map<String, Class<?>> typeMap = conn.getTypeMap();
            typeMap.put(typeName, ComplexTypeObj.class); //The name of the outer object type.
            typeMap.put(InnerType.ORACLE_OBJECT_NAME, InnerType.class); //The name of the inner object type.

            ArrayDescriptor des = ArrayDescriptor.createDescriptor(InnerType.ORACLE_LIST_NAME, conn); //The name of the inner list type.
            Array objArray = new ARRAY(des, conn, complexTypeObj.getInnerList().toArray());
            complexTypeObj.setInnerArray(objArray);

            ps.setObject(paramIndex, complexTypeObj);
        }
    }
}

外型:

import java.sql.*;
import java.util.*;

public class OuterType extends BaseSQLData implements SQLData {

    public static final String ORACLE_OBJECT_NAME = "T_OUTER_TYPE";

    private List<InnerType> innerList;
    private Array innerArray;

    public OuterType() {
        this.innerList = new ArrayList<InnerType>();
    }

    public String getSQLTypeName() throws SQLException {
        return ORACLE_OBJECT_NAME;
    }

    @Override
    public void writeSQL(SQLOutput stream) throws SQLException {
        stream.writeArray(innerArray);
    }

内部类型:

public final class InnerType extends BaseSQLData {
    public static final String ORACLE_OBJECT_NAME = "T_INNER_TYPE";
    public static final String ORACLE_LIST_NAME = "T_INNER_TYPE_LIST";

    private String valueA;
    private Long   valueB = 0;

    public String getSQLTypeName() throws SQLException {
        return ORACLE_OBJECT_NAME;
    }

    @Override
    public void readSQL(SQLInput stream, String typeName) throws SQLException {
        throw new UnsupportedOperationException("This class doesn't support read opperations.");
    }

    @Override
    public void writeSQL(SQLOutput stream) throws SQLException {
        stream.writeString(valueA);
        stream.writeBigDecimal(valueB == null ? null : new BigDecimal(valueB.toString()));
    }