Methods for retrieving and converting data types

With the IMS Universal JDBC driver, you can use the ResultSet interface (java.sql.ResultSet) to retrieve and convert the data from the type that is defined in the database metadata to the type that is required by your Java™ application. Similarly, with the IMS Universal DL/I driver, you can use the Path interface to perform data retrieval and conversion to Java data types.

The following table shows the available get methods in the ResultSet interface (for the IMS Universal JDBC driver) or the Path interface (for the IMS Universal DL/I driver) for accessing data of a certain Java data type.

The "No Truncation or Data Loss" column indicates the data types that are designed to be accessed with the given getXXX method. No truncation or data loss occurs when using those methods for those data types. The data types that are in the Legal without Data Integrity column are all other legal calls; however, data integrity cannot be ensured when using the given getxxx method to access those data types. If a data type is not in either column, using the given getXXX method for that data type will result in an exception.

Table 1. ResultSet.getXXX and Path.getXXX methods to retrieve data types
ResultSet.getXXX Method or Path.getXXX Method	Data Type (any not listed result in an exception)
ResultSet.getXXX Method or Path.getXXX Method	No Truncation or Data Loss	Legal without Data Integrity
getByte	TINYINT UTINYINT	SMALLINT INTEGER BIGINT FLOAT DOUBLE BIT CHAR VARCHAR PACKEDDECIMAL¹ ZONEDDECIMAL¹
getShort	SMALLINT USMALLINT	TINYINT INTEGER BIGINT FLOAT DOUBLE BIT CHAR VARCHAR PACKEDDECIMAL¹ ZONEDDECIMAL¹
getInt	INTEGER UINTEGER	TINYINT SMALLINT BIGINT FLOAT DOUBLE BIT CHAR VARCHAR PACKEDDECIMAL¹ ZONEDDECIMAL¹
getLong	BIGINT UBIGINT	TINYINT SMALLINT INTEGER FLOAT DOUBLE BIT CHAR VARCHAR PACKEDDECIMAL¹ ZONEDDECIMAL¹
getFloat	FLOAT	TINYINT SMALLINT INTEGER BIGINT DOUBLE BIT CHAR VARCHAR PACKEDDECIMAL¹ ZONEDDECIMAL¹
getDouble	DOUBLE	TINYINT SMALLINT INTEGER BIGINT FLOAT BIT CHAR VARCHAR PACKEDDECIMAL¹ ZONEDDECIMAL¹
getBoolean	BIT	TINYINT SMALLINT INTEGER BIGINT FLOAT DOUBLE CHAR VARCHAR PACKEDDECIMAL¹ ZONEDDECIMAL¹
getString	CHAR VARCHAR	TINYINT SMALLINT INTEGER BIGINT FLOAT DOUBLE BIT PACKEDDECIMAL¹ ZONEDDECIMAL¹ BINARY DATE TIME TIMESTAMP
getBigDecimal	BINARY³ PACKEDDECIMAL¹ ZONEDDECIMAL¹	TINYINT SMALLINT INTEGER BIGINT FLOAT DOUBLE BIT CHAR VARCHAR
getClob	CLOB²	all others result in an exception
getBytes	BINARY	SMALLINT INTEGER BIGINT FLOAT DOUBLE BIT CHAR VARCHAR PACKEDDECIMAL1 ZONEDDECIMAL1 DATE TIME TIMESTAMP
getDate	DATE	CHAR VARCHAR TYIMESTAMP
getTime	TIME	CHAR VARCHAR TIMESTAMP
getTimestamp	TIMESTAMP	CHAR VARCHAR DATE TIME

Note:

PACKEDDECIMAL and ZONEDDECIMAL are data type extensions for IMS Universal JDBC driver and the IMS Universal DL/I driver. All other types are standard SQL types defined in SQL92. Restriction: PACKEDDECIMAL and ZONEDDECIMAL data types do not support the Sign Leading or Sign Separate modes. For these two data types, sign information is always stored with the Sign Trailing method.
The CLOB data type is supported only for the retrieval and storage of XML data.
The BINARY data type is valid only for decimal data used with a binary type converter.

If the field type is either PACKEDDECIMAL or ZONEDDECIMAL, the type qualifier is the COBOL PICTURE string that represents the layout of the field. All COBOL PICTURE strings that contain valid combinations of 9s, Ps, Vs, and Ss are supported. Expansion of PICTURE strings is handled automatically. For example, '9(5)' is a valid PICTURE string. For zoned decimal numbers, the decimal point can also be used in the PICTURE string. PIC 9(06)V99 COMP and PIC 9(06)V99 COMP-4 are valid PICTURE clauses for BINARY decimal data.

If the field contains DATE, TIME, or TIMESTAMP data, the type qualifier specifies the format of the data. For example, a type qualifier of ddMMyyyy indicates that the data is formatted as follows:

11122015 is December 11, 2015

For DATE and TIME types, all formatting options in the java.text.SimpleDateFormat class are supported.

For the TIMESTAMP type, the formatting option 'f' is available for nanoseconds. TIMESTAMP can contain up to nine 'f's and replaces the 'S' options for milliseconds. Instead, 'fff' indicates milliseconds of precision. An example TIMESTAMP format is as follows:

yyyy-mm-dd hh:mm:ss.fffffffff