The following table shows the correspondence of data types between the source and target tables when the source database is HiRDB.

Table 4-8 Correspondence of data types between the source and target tables when the source database is HiRDB

Data type in source table	Data type in target table
INTEGER	INTEGER
SMALLINT	SMALLINT
DECIMAL	DECIMAL
LARGE DECIMAL	LARGE DECIMAL
FLOAT	FLOAT
SMALLFLT	SMALLFLT#1
CHAR	CHAR or MCHAR
VARCHAR	VARCHAR or MVARCHAR
NCHAR	NCHAR
NVARCHAR	NVARCHAR
MCHAR	CHAR or MCHAR
MVARCHAR	VARCHAR or MVARCHAR
DATE	DATE
TIME	TIME
TIMESTAMP	TIMESTAMP
INTERVAL YEAR TO DAY	INTERVAL YEAR TO DAY
INTERVAL HOUR TO SECOND	INTERVAL HOUR TO SECOND
SGMLTEXT#2	SGMLTEXT#2
FREEWORD	FREEWORD
XML	XML
BLOB#2	BLOB#2
BINARY	BINARY

Notes:

• An error occurs at the target HiRDB if you attempt to import data that is longer than the target table column or whose numeric value overflows.
• You can specify a repetition column. However, data extraction cannot be performed in units of elements in the repetition column.

#1

Data linkage is supported, but rounding errors occur because values are rounded.

#2

To use data linkage, specify ALL in the RECOVERY operand in the HiRDB table definition. For details about HiRDB table definition, see the manual HiRDB Version 9 SQL Reference.

The following table shows the correspondence of data types between the source and target tables when the source database is XDM/SD E2.

Table 4-9 Correspondence of data types between the source and target tables when the source database is XDM/SD E2

Data subject to extraction processing#1		Data type in target table
Data attribute	Data length
CHAR(n)	1 n 30,000	CHAR(n)
CHAR(n)	n > 30,000	VARCHAR(n)#3
NCHAR(n)	1 n 15,000	NCHAR(n)
NCHAR(n)	n > 15,000	NVARCHAR(n)#3
PACK(m,n)	--	DECIMAL(m + n,n)#4
UNPACK(m,n)#2	--	DECIMAL(m + n,n)#4
COMP(m,n)	1 m 4, n = 0	SMALLINT
COMP(m,n)	5 m 9, n = 0	INTEGER
COMP(m,n)	m = 0 or m > 10 or n 0	DECIMAL(29,n)#6
BINARY(m,n)	1 m 15, n = 0	SMALLINT
BINARY(m,n)	16 m 31, n = 0	INTEGER
BINARY(m,n)	m = 0 or m > 32 or n 0	DECIMAL(29,n)#5

Note:

An error occurs at the target HiRDB if you attempt to import data that is longer than the target table column or whose numeric value overflows.

#1: If the RESTRUCT statement is specified with XDM/DS, Datareplicator uses the attribute and length defined with the RESTRUCT statement.

#2: Data conversion on the UNPACK type results in an error if any of the following conditions is satisfied:

• The value of the zoned part is not (F)₁₆.
• The value of the sign part is not (F)₁₆, (C)₁₆, or (D)₁₆.
• The value of the numeric part is outside the range (0)₁₆ to (9)₁₆.

If an error occurs, Datareplicator outputs an error message and the unimported information, and then terminates import processing on the erroneous import group. If you are using a UOC routine during import processing, Datareplicator passes the extracted data as is as UNPACK data to the UOC routine without converting the data type, and then resumes import processing.

#3: Spaces are not suppressed.

#4: The sign is not normalized.

#5: BINARY type is converted to DECIMAL type if the number of decimal places is not zero. Because the maximum precision for the DECIMAL type supported by XDM/DS is 29 decimal places, the digits beginning with the 30^th decimal place are truncated.

#6: The maximum precision for the DECIMAL type supported by XDM/DS is 29.

The following table shows the correspondence of data types between the source and target tables when the source database is XDM/RD E2.

Table 4-10 Correspondence of data types between the source and target tables when the source database is XDM/RD E2

Data type in source table	Data type in target table
INTEGER	INTEGER
SMALLINT	SMALLINT
DECIMAL	DECIMAL
LARGE DECIMAL	LARGE DECIMAL
FLOAT	FLOAT
SMALLFLT	SMALLFLT#1
CHAR	CHAR or MCHAR
VARCHAR	VARCHAR
LONG VARCHAR
NCHAR	NCHAR
NVARCHAR	NVARCHAR
LONG NVARCHAR
MCHAR	CHAR or MCHAR
MVARCHAR	VARCHAR or MVARCHAR
LONG MVARCHAR
DATE	DATE
TIME	TIME
INTERVAL YEAR TO DAY	INTERVAL YEAR TO DAY
INTERVAL HOUR TO SECOND	INTERVAL HOUR TO SECOND
BLOB#2	BLOB#2

Notes:

• An error occurs at the target HiRDB if you attempt to import data that is longer than the target table column or whose numeric value overflows.
• You cannot specify repetition and array columns.

#1

Data linkage is supported, but rounding errors occur because values are rounded.

#2

To apply data linkage to columns of the BLOB type, specify ALL in the RECOVERY operand in the HiRDB table definition. For details about HiRDB table definition, see the manual HiRDB Version 9 SQL Reference.

The following table shows the correspondence of data types between the source and target tables when the source database is ADM.

Table 4-11 Correspondence of data types between the source and target tables when the source database is ADM

Data type under ADM		Data type at Datareplicator	Data type in target table
Cn		CHAR(n)	CHAR(n)
Xn		CHAR(n)	CHAR(n)
Pn	n 15	PACK(2n - 1,0)	DECIMAL(2n - 1,0)
	n > 15	CHAR(n)	CHAR(n)

Notes:

1. An error occurs at the target HiRDB if you attempt to import data that is longer than the target table column or whose numeric value overflows.
2. If you have used XDM/DS's RESTRUCT statement to define the UNPACK attribute as an ADM data type and the extracted update information contains data satisfying any of the following conditions, Datareplicator terminates import processing:
   - The value of the zoned part is not (F)₁₆.
   - The value of the sign part is not (F)₁₆, (C)₁₆, or (D)₁₆.
   - The value of the numeric part is outside the range (0)₁₆ to (9)₁₆.
   If an error occurs, Datareplicator outputs an error message and the unimported information, and then terminates import processing on the erroneous import group. If you are using a UOC routine during import processing, Datareplicator passes the extracted data as is as UNPACK data to the UOC routine without converting the data type, and then resumes import processing.
   

Table 4-12 Correspondence of data types between the extraction redefinition fields and the target table's data types

Field attribute in extraction redefinition statement	Data length	Data type in target table
CHAR(n)	n 30,000	CHAR(n)
NCHAR(n)	n 15,000	NCHAR(n)
PACK(m,n)	N/A	DECIMAL(m + n,n)
PACKNS(m,n)	N/A	DECIMAL(m + n,n)
UNPACK(m,n)	N/A	DECIMAL(m + n,n)
UNPACKNS(m,n)	N/A	DECIMAL(m + n,n)
COMP(m,n)	1 m 4, n = 0	SMALLINT
	5 m 9, n = 0	INTEGER
	m = 0 or m > 10 or n! = 0	DECIMAL(29,n)#
BINARY(m,n)	1 m 15, n = 0	SMALLINT
	16 m 31, n = 0	INTEGER
	Other than the above	FLOAT
FLOAT(n)	n = 4, 8	FLOAT

Note:

An error occurs at the target HiRDB if you attempt to import data that is longer than the target table column or whose numeric value overflows.

#

The maximum precision for the DECIMAL type supported by XDM/DS is 29.

Table 4-13 Correspondence of data types between PDM2 E2's field attributes and the target table's data types

Field attribute in PDM2 E2	Data length	Data type in target table
X(n)	n 30,000	CHAR(n)
P(m,n), AP(m,n)	n > 0, 2m - 1 > n	DECIMAL(2m - 1,n)
	n > 0, 2m - 1 n	DECIMAL(n,n)
	n 0	DECIMAL(2m - 1 - n,0)
Z(m,n), AZ(m,n)	n > 0, m > n	DECIMAL(m,n)
	n > 0, m n	DECIMAL(n,n)
	n 0	DECIMAL(m - n,0)
B(m,n), AB(m,n)	n > 0, 2.5m > n	DECIMAL(2.5m,n)
	n > 0, 2.5m n	DECIMAL(n,n)
	n 0	DECIMAL(2.5m - n,0)
F(n)	n = 4, 8	FLOAT
N(n)	n 30,000	NCHAR(n / 2)

Note:

An error occurs at the target HiRDB if you attempt to import data that is longer than the target table column or whose numeric value overflows.

When the source database is TMS-4V/SP, you must use a UOC routine provided by TMS-4V/SP/Data Linkage Support to define data types. For details about data types when the source database is TMS-4V/SP, see the VOS3 TMS-4V/SP/Data Linkage Support manual.

The following table shows the correspondence of data types between the source and target tables when the source database is RDB1 E2.

Table 4-14 Correspondence of data types between the source and target tables when the source database is RDB1 E2

Data type in source table	Data type in target table
INTEGER	INTEGER
SMALLINT	SMALLINT
DECIMAL	DECIMAL
FLOAT	FLOAT
SMALLFLT	SMALLFLT#
CHAR	CHAR or MCHAR
VARCHAR	VARCHAR
LONG VARCHAR	N/A
NCHAR	NCHAR
NVARCHAR	NVARCHAR

N/A: Not subject to import processing.

Notes:

• An error occurs at the target HiRDB if you attempt to import data that is longer than the target table column or whose numeric value overflows.
• A repetition column cannot be specified.

#

Data linkage is supported, but rounding errors occur because values are rounded.