The following table lists the data types that can be specified as mapping keys.

Table 4-6 Data types that can be specified as mapping keys (when the source database is HiRDB)

Classification	Data type that can be specified as mapping key	Column definition length
Numeric type	INTEGER	--
	SMALLINT	--
	DECIMAL(m,n)	1 m 38, 0 n 38, n m
	FLOAT	--
	SMALLFLT	--
Character type	CHAR(n)	n 255
	VARCHAR(n)	n 255
	NCHAR(n)	n 127
	NVARCHAR(n)	n 127
	MCHAR(n)	n 255
	MVARCHAR(n)	n 255
Date type	DATE	--
	TIME	--
	TIMESTAMP(p)	p = 0, 2, 4, 6
	INTERVAL YEAR TO DAY	--
	INTERVAL HOUR TO SECOND	--

Legend:

--: Not applicable

Note:

A repetition key cannot be specified as a mapping key.

• Specify a mapping key in the key or ukey clause of the extraction statement in the extraction definition.
• You can specify up to 16 columns for one mapping key.
• Once you specify a column in the key clause of an extraction definition statement in the extraction definition, you must not update the mapping key. If you update the mapping key, the target database will be updated based on the new key value; as a result, the wrong row might be updated in the target database, and conformity between the source and target databases might be lost. To update a mapping key, specify the applicable column in the ukey clause.
• The data corresponding to all mapping key rows that are manipulated by INSERT, UPDATE, and DELETE is sent to the target Datareplicator.

• For details about how to specify a mapping key when the source database is XDM/SD E2, see the VOS3 XDM/DS manual.
• The following data is sent to the target Datareplicator:
  

  During ERASE of a record subject to extraction processing

  The record subject to extraction processing and the data corresponding to the mapping key at the higher levels

  During ERASE of a record at a higher level than the record subject to extraction processing

  The deleted record and the data corresponding to the mapping key at the higher levels

  During STORE or MODIFY of a record subject to extraction processing

  The record subject to extraction processing and the data corresponding to the mapping key at the higher levels

• For details about how to specify a mapping key when the source database is XDM/RD E2, see the VOS3 XDM/DS manual.
• The data corresponding to all mapping key rows that are manipulated by INSERT, UPDATE, and DELETE is sent to the target Datareplicator.

• For details about how to specify a mapping key when the source database is ADM, see the VOS3 XDM/DS manual.
• The following data is sent to the target Datareplicator:
  Data corresponding to the mapping keys for ISRT, REPL, and DLET calls
  

(a) Data linkage using XDM/DS

• For details about how to specify a mapping key when the source database is PDM2 E2, see the VOS3 XDM/DS manual.

(b) Data linkage using SAM files

• When the source database is PDM2 E2, specify the mapping key in the update information definition; for details about how to specify the update information definition, see 5.11 Update information definition.
• The contents of the SAM file transferred from the source database is stored in an import information queue file at the target Datareplicator.

• For details about how to specify a mapping key when the source database is TMS-4V/SP, see the VOS3 XDM/DS manual.
• The data corresponding to all mapping key rows that are manipulated by INSERT, UPDATE, and DELETE is sent to the target Datareplicator.

• When the source database is RDB1 E2, mapping keys are created by RDB1 E2's update information extraction facility.
• The contents of the SAM file transferred from the source database are stored in an import information queue file at the target Datareplicator.

If a mapping key includes a SMALLFLT or FLOAT column of the round number type, the data format depends on the source system. The round number type might result in SQLCODE 100 (no row satisfies the conditions) during SQL statement execution. For this reason, do not specify in a mapping key a column of the FLOAT or SMALLFLT type.