使用 JDBC 从存储过程中获取 Oracle 表类型
声明：本页面是StackOverFlow热门问题的中英对照翻译，遵循CC BY-SA 4.0协议，如果您需要使用它，必须同样遵循CC BY-SA许可，注明原文地址和作者信息，同时你必须将它归于原作者（不是我）：StackOverFlow
原文地址: http://stackoverflow.com/questions/6410452/
Warning: these are provided under cc-by-sa 4.0 license. You are free to use/share it, But you must attribute it to the original authors (not me):
StackOverFlow
Fetch Oracle table type from stored procedure using JDBC
提问by Lukas Eder
I'm trying to understand different ways of getting table data from Oracle stored procedures / functions using JDBC. The six ways are the following ones:
我试图了解使用 JDBC 从 Oracle 存储过程/函数获取表数据的不同方法。这六种方式如下：
- procedure returning a schema-level table type as an OUT parameter
- procedure returning a package-level table type as an OUT parameter
- procedure returning a package-level cursor type as an OUT parameter
- function returning a schema-level table type
- function returning a package-level table type
- function returning a package-level cursor type
- 将模式级表类型作为 OUT 参数返回的过程
- 将包级表类型作为 OUT 参数返回的过程
- 将包级游标类型作为 OUT 参数返回的过程
- 返回模式级表类型的函数
- 返回包级表类型的函数
- 返回包级游标类型的函数
Here are some examples in PL/SQL:
以下是 PL/SQL 中的一些示例：
-- schema-level table type
CREATE TYPE t_type AS OBJECT (val VARCHAR(4));
CREATE TYPE t_table AS TABLE OF t_type;
CREATE OR REPLACE PACKAGE t_package AS
-- package level table type
TYPE t_table IS TABLE OF some_table%rowtype;
-- package level cursor type
TYPE t_cursor IS REF CURSOR;
END library_types;
-- and example procedures:
CREATE PROCEDURE p_1 (result OUT t_table);
CREATE PROCEDURE p_2 (result OUT t_package.t_table);
CREATE PROCEDURE p_3 (result OUT t_package.t_cursor);
CREATE FUNCTION f_4 RETURN t_table;
CREATE FUNCTION f_5 RETURN t_package.t_table;
CREATE FUNCTION f_6 RETURN t_package.t_cursor;
I have succeeded in calling 3, 4, and 6 with JDBC:
我已经成功地使用 JDBC 调用了 3、4 和 6：
// Not OK: p_1 and p_2
CallableStatement call = connection.prepareCall("{ call p_1(?) }");
call.registerOutParameter(1, OracleTypes.CURSOR);
call.execute(); // Raises PLS-00306. Obviously CURSOR is the wrong type
// OK: p_3
CallableStatement call = connection.prepareCall("{ call p_3(?) }");
call.registerOutParameter(1, OracleTypes.CURSOR);
call.execute();
ResultSet rs = (ResultSet) call.getObject(1); // Cursor results
// OK: f_4
PreparedStatement stmt = connection.prepareStatement("select * from table(f_4)");
ResultSet rs = stmt.executeQuery();
// Not OK: f_5
PreparedStatement stmt = connection.prepareStatement("select * from table(f_5)");
stmt.executeQuery(); // Raises ORA-00902: Invalid data type
// OK: f_6
CallableStatement call = connection.prepareCall("{ ? = call f_6 }");
call.registerOutParameter(1, OracleTypes.CURSOR);
call.execute();
ResultSet rs = (ResultSet) call.getObject(1); // Cursor results
So obviously, I'm having trouble understanding
很明显，我很难理解
- How to retrieve schema-level and package-level table types from OUT parameters in stored procedures
- How to retrieve package-level table types from stored functions
- 如何从存储过程中的 OUT 参数检索架构级和包级表类型
- 如何从存储的函数中检索包级表类型
I can't seem to find any documentation on this, as everyone always uses cursors instead of table types. Maybe because it's not possible? I prefer table types, though, because they are formally defined and can be discovered using the dictionary views (at least the schema-level table types).
我似乎找不到任何关于此的文档，因为每个人都总是使用游标而不是表类型。也许是因为不可能？不过，我更喜欢表类型，因为它们是正式定义的，可以使用字典视图（至少是模式级别的表类型）发现。
Note: obviously, I could write a wrapper function returning the OUT parameters and package-level table types. But I'd prefer the clean solution.
注意：显然，我可以编写一个包装函数来返回 OUT 参数和包级表类型。但我更喜欢干净的解决方案。
采纳答案by Vincent Malgrat
You can't access PLSQL objects (cases 2 & 5 = package-level objects) from java, see "java - passing array in oracle stored procedure". You can however access SQL types (case 1 and 4).
您无法从 java 访问 PLSQL 对象（情况 2 和 5 = 包级对象），请参阅“java - 在 oracle 存储过程中传递数组”。但是，您可以访问 SQL 类型（情况 1 和 4）。
To get OUT parameters from PL/SQL to java, you can use the method described in one of Tom Kyte's threadusing OracleCallableStatement. Your code will have an additional step since you're retrieving a table of Object instead of a table of VARCHAR.
要将 OUT 参数从 PL/SQL 获取到 java，您可以使用使用 OracleCallableStatement的 Tom Kyte 线程之一中描述的方法。由于您正在检索 Object 表而不是 VARCHAR 表，因此您的代码将有一个额外的步骤。
Here's a demo using Table of SQL Object, first the setup:
这是一个使用 SQL 对象表的演示，首先是设置：
SQL> CREATE TYPE t_type AS OBJECT (val VARCHAR(4));
2 /
Type created
SQL> CREATE TYPE t_table AS TABLE OF t_type;
2 /
Type created
SQL> CREATE OR REPLACE PROCEDURE p_sql_type (p_out OUT t_table) IS
2 BEGIN
3 p_out := t_table(t_type('a'), t_type('b'));
4 END;
5 /
Procedure created
The actual java class (using dbms_output.put_lineto log because I will call it from SQL, use System.out.printlnif called from java):
实际的 java 类（dbms_output.put_line用于记录，因为我将从 SQL 调用它，System.out.println如果从 java调用则使用）：
SQL> CREATE OR REPLACE
2 AND COMPILE JAVA SOURCE NAMED "ArrayDemo"
3 as
4 import java.sql.*;
5 import oracle.sql.*;
6 import oracle.jdbc.driver.*;
7
8 public class ArrayDemo {
9
10 private static void log(String s) throws SQLException {
11 PreparedStatement ps =
12 new OracleDriver().defaultConnection().prepareStatement
13 ( "begin dbms_output.put_line(:x); end;" );
14 ps.setString(1, s);
15 ps.execute();
16 ps.close();
17 }
18
19 public static void getArray() throws SQLException {
20
21 Connection conn = new OracleDriver().defaultConnection();
22
23 OracleCallableStatement cs =
24 (OracleCallableStatement)conn.prepareCall
25 ( "begin p_sql_type(?); end;" );
26 cs.registerOutParameter(1, OracleTypes.ARRAY, "T_TABLE");
27 cs.execute();
28 ARRAY array_to_pass = cs.getARRAY(1);
29
30 /*showing content*/
31 Datum[] elements = array_to_pass.getOracleArray();
32
33 for (int i=0;i<elements.length;i++){
34 Object[] element = ((STRUCT) elements[i]).getAttributes();
35 String value = (String)element[0];
36 log("array(" + i + ").val=" + value);
37 }
38 }
39 }
40 /
Java created
Let's call it:
让我们称之为：
SQL> CREATE OR REPLACE
2 PROCEDURE show_java_calling_plsql
3 AS LANGUAGE JAVA
4 NAME 'ArrayDemo.getArray()';
5 /
Procedure created
SQL> EXEC show_java_calling_plsql;
array(0).val=a
array(1).val=b
回答by Tanmay kumar shaw
You Can Also Use the below one
你也可以使用下面的
public List<EmployeeBean> fetchDataFromSPForRM(String sInputDate) {
List<EmployeeBean> employeeList = new ArrayList<EmployeeBean>();
Connection dbCon = null;
ResultSet data = null;
CallableStatement cstmt = null;
try {
dbCon = DBUtil.getDBConnection();
String sqlQuery = "{? = call PKG_HOLD_RELEASE.FN_RM_PDD_LIST()}";
cstmt = dbCon.prepareCall(sqlQuery);
cstmt.registerOutParameter(1, OracleTypes.CURSOR);
cstmt.execute();
data = (ResultSet) cstmt.getObject(1);
while(data.next()){
EmployeeBean employee = new EmployeeBean();
employee.setEmpID(data.getString(1));
employee.setSubBusinessUnitId((Integer)data.getObject(2));
employee.setMonthOfIncentive((Integer)data.getObject(3));
employee.setPIPStatus(data.getString(5));
employee.setInvestigationStatus(data.getString(6));
employee.setEmpStatus(data.getString(7));
employee.setPortfolioPercentage((Integer)data.getObject(8));
employee.setIncentive((Double)data.getObject(9));
employee.setTotalSysemHoldAmt((Double)data.getObject(10));
employee.setTotalManualHoldAmt((Double)data.getObject(11));
employeeList.add(employee);
}
} catch (SQLException e) {
e.printStackTrace();
}finally{
try {
if(data != null){
data.close();
data = null;
}
if(cstmt != null){
cstmt.close();
cstmt = null;
}
if(dbCon != null){
dbCon.close();
dbCon = null;
}
} catch (SQLException e) {
e.printStackTrace();
}
}
return employeeList;
}
