Oracle7 Server Distributed Systems Volume II: Replicated Data

DBMS_REPCAT.ADD_conflicttype_RESOLUTION

Purpose

To designate a method for resolving an update, delete, or uniqueness conflict. You must call these procedures from the master definition site. The procedure that you need to call is determined by the type of conflict that the routine is used to resolve.

Conflict Type Procedure Name
update ADD_UPDATE_RESOLUTION
uniqueness ADD_UNIQUE_RESOLUTION
delete ADD_DELETE_RESOLUTION

For additional information, refer to .

Syntax

The parameters for the ADD_UPDATE_RESOLUTION procedure are described in Table 12 - 70, and the exceptions are listed in Table 12 - 71. The syntax for this procedure is shown below:

DBMS_REPCAT.ADD_UPDATE_RESOLUTION(
		sname			IN	VARCHAR2, 
		oname			IN	VARCHAR2, 
		column_group		IN	VARCHAR2,
		sequence_no		IN	NUMBER,
		method			IN	VARCHAR2, 
		parameter_column_name	IN	VARCHAR2, 
		priority_group	IN	VARCHAR2 := NULL,
		function_name		IN	VARCHAR2 := NULL,
		comment		IN	VARCHAR2 := NULL)

The parameters for the ADD_DELETE_RESOLUTION procedure are described in Table 12 - 70, and the exceptions are listed in Table 12 - 71. The syntax for this procedure is shown below:

DBMS_REPCAT.ADD_DELETE_RESOLUTION(
	sname			IN	VARCHAR2, 
	oname			IN	VARCHAR2, 
	sequence_no		IN	NUMBER,
	parameter_column_name	IN	VARCHAR2, 
	function_name		IN	VARCHAR2 := NULL,
	comment			IN	VARCHAR2 := NULL)

The parameters for the ADD_UNIQUE_RESOLUTION procedure are described in Table 12 - 70, and the exceptions are listed in Table 12 - 71. The syntax for this procedure is shown below:

DBMS_REPCAT.ADD_UNIQUE_RESOLUTION(
	sname			IN	VARCHAR2, 
	oname			IN	VARCHAR2, 
	constraint_name		IN	VARCHAR2,
	sequence_no		IN	NUMBER,
	method			IN	VARCHAR2, 
	parameter_column_name	IN	VARCHAR2, 
	function_name		IN	VARCHAR2 := NULL,
	comment			IN	VARCHAR2 := NULL)

Parameter Description
sname The name of the schema containing the table to be replicated.
oname The name of the table for which you are adding a conflict resolution routine.
column_group The name of the column group for which you are adding a conflict resolution routine. Column groups are required for update conflict resolution routines only.
constraint_name The name of the unique constraint or unique index for which you are adding a conflict resolution routine. Use the name of the unique index if it differs from the name of the associated unique constraint. Constraint names are required for uniqueness conflict resolution routines only.
sequence_no The order in which the designated conflict resolution methods should be applied.
method The type of conflict resolution routine that you want to create. This can be the name of one of the standard routines provided with symmetric replication, or, if you have written your own routine, you should choose USER FUNCTION, and provide the name of your routine as the FUNCTION_NAME argument. The methods supported in this release are: MINIMUM, MAXIMUM, LATEST TIMESTAMP, EARLIEST TIMESTAMP, ADDITIVE, AVERAGE, PRIORITY GROUP, SITE PRIORITY, OVERWRITE, and DISCARD (for update conflicts) and APPEND SITE NAME, APPEND SEQUENCE NUMBER, and DISCARD (for uniqueness conflicts). There are no standard methods for delete conflicts, so this argument is not used.
parameter_column_name The name of the columns used to resolve the conflict. The standard methods operate on a single column. For example, if you are using the LATEST TIMESTAMP method for a column group, you should pass the name of the column containing the timestamp value as this argument. If your are using a USER FUNCTION, you can resolve the conflict using any number of columns. This argument accepts either a comma separated list of column names, or a PL/SQL table of type dbms_repcat.varchar2s. The single value `*' indicates that you want to use all of the columns in the table (or column group, for update conflicts) to resolve the conflict. If you specify `*', the columns will be passed to your function in alphabetical order.
priority_group If you are using the PRIORITY GROUP or SITE PRIORITY update conflict resolution method, you must supply the name of the priority group that you have created. Instructions for creating a priority group are included . If you are using a different method, you can use the default value for this argument, NULL. This argument is applicable to update conflicts only.
function_name If you selected the USER FUNCTION method, or if you are adding a delete conflict resolution routine, you must supply the name of the conflict resolution routine that you have written. If you are using one of the standard methods, you can use the default value for this argument, NULL.
comment This user comment is added to the RepResolution view.

Table 12 - 70. Parameters for ADD_UPDATE/DELETE_RESOLUTION

Exception Description
nonmasterdef The invocation site is not the masterdef site.
missingobject The given object does not exist as a table in the given schema using row-level replication.
missingschema The given schema does not exist.
missingcolumn The column that you specified as part of the PARAMETER_COLUMN_NAME argument does not exist.
missingprioritygroup The priority group that you specified does not exist for the table.
invalidmethod The resolution method that you specified is not recognized.
invalidparameter The number of columns that you specified for the PARAMETER_COLUMN_NAME argument is invalid. (The standard routines take only one column name.)
missingfunction The user function that you specified does not exist.
missingconstraint The constraint that you specified for a uniqueness conflict does not exist.
typefailure The datatype of one of the values that you specified for the PARAMETER_COLUMN_NAME argument is not one of the types supported for the given method.

Table 12 - 71. Exceptions for ADD_UPDATE/DELETE_RESOLUTION

Conflict Type	Procedure Name
update	`ADD_UPDATE_RESOLUTION`
uniqueness	`ADD_UNIQUE_RESOLUTION`
delete	`ADD_DELETE_RESOLUTION`

Parameter	Description
`sname`	The name of the schema containing the table to be replicated.
`oname`	The name of the table for which you are adding a conflict resolution routine.
`column_group`	The name of the column group for which you are adding a conflict resolution routine. Column groups are required for update conflict resolution routines only.
`constraint_name`	The name of the unique constraint or unique index for which you are adding a conflict resolution routine. Use the name of the unique index if it differs from the name of the associated unique constraint. Constraint names are required for uniqueness conflict resolution routines only.
`sequence_no`	The order in which the designated conflict resolution methods should be applied.
`method`	The type of conflict resolution routine that you want to create. This can be the name of one of the standard routines provided with symmetric replication, or, if you have written your own routine, you should choose USER FUNCTION, and provide the name of your routine as the FUNCTION_NAME argument. The methods supported in this release are: MINIMUM, MAXIMUM, LATEST TIMESTAMP, EARLIEST TIMESTAMP, ADDITIVE, AVERAGE, PRIORITY GROUP, SITE PRIORITY, OVERWRITE, and DISCARD (for update conflicts) and APPEND SITE NAME, APPEND SEQUENCE NUMBER, and DISCARD (for uniqueness conflicts). There are no standard methods for delete conflicts, so this argument is not used.
`parameter_column_name`	The name of the columns used to resolve the conflict. The standard methods operate on a single column. For example, if you are using the LATEST TIMESTAMP method for a column group, you should pass the name of the column containing the timestamp value as this argument. If your are using a USER FUNCTION, you can resolve the conflict using any number of columns. This argument accepts either a comma separated list of column names, or a PL/SQL table of type dbms_repcat.varchar2s. The single value `' indicates that you want to use all of the columns in the table (or column group, for update conflicts) to resolve the conflict. If you specify `', the columns will be passed to your function in alphabetical order.
`priority_group`	If you are using the PRIORITY GROUP or SITE PRIORITY update conflict resolution method, you must supply the name of the priority group that you have created. Instructions for creating a priority group are included . If you are using a different method, you can use the default value for this argument, NULL. This argument is applicable to update conflicts only.
`function_name`	If you selected the USER FUNCTION method, or if you are adding a delete conflict resolution routine, you must supply the name of the conflict resolution routine that you have written. If you are using one of the standard methods, you can use the default value for this argument, NULL.
`comment`	This user comment is added to the RepResolution view.

Exception	Description
nonmasterdef	The invocation site is not the masterdef site.
missingobject	The given object does not exist as a table in the given schema using row-level replication.
missingschema	The given schema does not exist.
missingcolumn	The column that you specified as part of the PARAMETER_COLUMN_NAME argument does not exist.
missingprioritygroup	The priority group that you specified does not exist for the table.
invalidmethod	The resolution method that you specified is not recognized.
invalidparameter	The number of columns that you specified for the PARAMETER_COLUMN_NAME argument is invalid. (The standard routines take only one column name.)
missingfunction	The user function that you specified does not exist.
missingconstraint	The constraint that you specified for a uniqueness conflict does not exist.
typefailure	The datatype of one of the values that you specified for the PARAMETER_COLUMN_NAME argument is not one of the types supported for the given method.