Online Documentation for Data Pump for Oracle
Configuration file format
The configuration (template) file used by Data Pump for Oracle is divided into several sections, each corresponding to a particular group of settings specified at different steps of the GUI application wizard.
[#General#]
This section stores general information about the utility:
Parameter |
Description |
Product |
internal product name |
Version |
major version |
[Source]
This section stores the connection properties used to access the data source. The connection string is built according to parameters specified in the Data Link Properties dialog.
Parameter |
Description |
ConnectionString |
connection string for the data source |
[#Comment#]
This section stores the template file comment as specified optionally in the Save template options dialog:
Parameter |
Description |
Line<N> |
comment text |
where N stands for the comment line identifier
Example:
Line0=Data Pump for Oracle
Line1=Template file #1
Line2=Pumping tables from MS Access to Oracle
[Options]
This section stores data conversion and script execution options. The parameters correspond to the values specified at Step 4 and Step 8 of the Wizard application.
Parameter |
Description |
CommitAfter |
number of records after which the COMMIT statement is inserted |
DisplayErrorMessages |
0 = error messages are hidden in the import log 1 = error messages are displayed in the import log |
DetailErrorMessages |
0 = simple error messages in the import log 1 = detailed error messages in the import log |
ShowTableProgress |
0 = simple data import progress in the import log 1 = detailed data import progress in the import log |
ConvertNames |
0 = object names are not converted 1 = object names are converted to conform to SQL92 naming rules |
NamesFormat |
nfAsIs = object names are left without changes nfLowerCase = object names are converted to lower case nfUpperCase = object names are converted to upper case |
ConvertDefaultValues |
0 = default values of the source database are not converted 1 = default values of the source database are converted to default values for the destination database |
EnableScriptComments |
0 = script comments are disabled 1 = script comments are included into the body of the script |
DropIfExistDbStatement |
the parameter is not used by Data Pump for Oracle |
StopScriptOnError |
0 = script execution is not stopped on errors 1 = script execution is stopped if an error occurs |
ShowExecutedStatements |
0 = the Script execution information area does not display any information upon script execution at Step 6 1 = records for successfully executed statements are listed in the Script execution information area at Step 6 |
AlwaysSkipSuccExecuted |
0 = upon subsequent script execution at Step 6 you are prompted to choose whether successfully executed statements should be skipped or not 1 = successfully executed statements are skipped upon subsequent script execution at Step 6 |
ClearTablesBeforeImport |
0 = tables in the Oracle database are not cleared before data import at Step 8 1 = tables in the Oracle database are cleared before data import at Step 8 |
TrimStringSpaces |
0 = source data strings 1 = unused space at the end of source data strings is cut off |
QuoteNamesInImportSelQuery |
0 = quoting source object identifiers is disabled 1 = quoting source object identifiers is enabled |
QuoteTemplate |
qtCustom = custom quoting characters (QuoteCharLeft, QuoteCharRight) qtDefault = the default quotes for the data source in use qtDoubleQuote = "..." qtSingleQuote = '...' qtApostrophe = `...` qtSquareBracket = [...] qtRoundBracket = (...) qtBrace = {...} qtAngleBracket = <...> |
QuoteCharLeft |
specifies the left quoting character (if QuoteTemplate = qtCustom) |
QuoteCharRight |
specifies the right quoting character (if QuoteTemplate = qtCustom) |
SchemasMapping |
smExisting = the tables are placed into an existing schema (SchemasMappingExist) |
SchemasMappingExist |
name of the existing schema in the Oracle database to pump the source tables into |
SchemasMappingNew |
the parameter is not used by Data Pump for Oracle |
GenerateDropTableStmt |
0 = no DROP statements are generated 1 = the DROP statements are added for tables in the script |
[Target]
This section stores connection parameters for the target Oracle database. The parameters correspond to the values entered at Step 1 of the Wizard application and are obligatory.
Parameter |
Description |
Host |
the parameter is not used by Data Pump for Oracle |
Login |
Oracle login (if NTAuth = 0) |
Password |
password to identify the login (encrypted) |
RemoteIndex |
the parameter is not used by Data Pump for Oracle |
DataBase |
target database name as specified in the TNS file |
DatabaseActions |
cdtSelectExisting = an existing database is selected for pumping data |
DatabaseHome |
Oracle Home storage |
ConnectAs |
caNormal = connect as Normal caSYSDBA = connect as SYSDBA caSYSOPER = connect as SYSOPER |
NTAuth |
type of authentication: 0 = Server authentication 1 = Windows authentication |
OptimizeGoalType |
0 = Unchanged 1 = Choose 2 = FirstRows 3 = AllRows 4 = Rule |
TunnelType |
indicates whether SSH tunneling is used for connection or not (TunnelType = ttNotUse) |
SSHHostName |
name of the host where SSH server is running |
SSHPort |
port on which SSH server is activated |
SSHUserName |
user on the machine where SSH server is running |
SSHPassword |
password to identify SSH server user (encrypted) |
SSHKeyFile |
path to the Private Key used for the SSH connection (if SSHUseKeyFile = True) |
SSHUseKeyFile |
True = SSH Private Key is used False = SSH Private Key is not used |
[{TBL}<table_name>]
Sections of this type contain table properties (those that were changed at Step 5 of the Wizard application) and data import options specified at Step 7.
Note: Sections of this type are only available in dynamic templates (for details see Save template options).
Parameter |
Description |
Name |
table name |
NameSpace |
Oracle schema to save the table into |
Excluded |
1 = the table is excluded from data import at Step 7 |
WhereClause |
text of the WHERE condition for pumping data |
[{FLD}<field_name>]
Sections of this type contain field properties (those that were changed at Step 5 of the Wizard application).
Note: Sections of this type are only available in dynamic templates (for details see Save template options).
Parameter |
Description |
Name |
field name |
Type |
data type applied to the field |
Default |
value that inserted records get by default |
DataScale |
scale for the field values |
DataPrecision |
precision for the field values |
NotNULL |
0 = nullable 1 = NOT NULL |
[{IND}<index_name>]
Sections of this type contain index properties (those that were changed at Step 5 of the Wizard application).
Note: Sections of this type are only available in dynamic templates (for details see Save template options).
Parameter |
Description |
Name |
index name |
Unique |
0 = non-unique index 1 = unique index |
Logging |
0 = creation of the index is not logged 1 = creation of the index is logged in the redo log file |
Reverse |
0 = bytes of the index block are stored in regular order 1 = bytes of the index block are stored in reverse order, excluding the rowid |
SortingType |
1 = SORT is added to the index definition 2 = the rows are stored in the database in ascending order, so that Oracle does not have to sort the rows when creating the index |
[{FK}<key_name>]
Sections of this type contain key properties (those that were changed at Step 5 of the Wizard application).
Note: Sections of this type are only available in dynamic templates (for details see Save template options).
Parameter |
Description |
Name |
key name |
DeleteAction |
'on delete' rule (for foreign keys): 0 = NONE 1 = SET NULL 2 = CASCADE |
[TypeMapping]
This section stores source-target type mapping options. The parameters correspond to the values specified in the Type mapping dialog.
Note: This section is only available in dynamic templates (for details see Save template options).
[ExcludedFromStructure]
This section stores the list of source objects excluded from structure conversion at Step 3 of the Wizard application.
Note: This section is only available in dynamic templates (for details see Save template options).
Parameter |
Description |
{NSP}<schema_name> |
excluded schema (if available) |
{TBL}<table_name> |
excluded table |
{IND}<index_name> |
excluded index |
{FK}<key_name> |
excluded key |
[SelectedTables]
This section stores the list of source and target tables, as specified at Step 3 and Step 5 of the Wizard application.
Note: This section is only available in fixed templates (for details see Save template options).
Parameter |
Description |
TableCount |
number of tables selected for pumping |
SrcTableNS<N> |
source schema name (if available) |
SrcTableNM<N> |
source table name |
TrgTableNS0<N> |
target schema name |
TrgTableNM<N> |
target table name |
where N stands for the table identifier
Example:
[SelectedTables]
TableCount=2
SrcTableNS0=dbo
SrcTableNM0=EMPLOYEE
TrgTableNM0=EMPLOYEE
SrcTableNS1=dbo
SrcTableNM1=DEPARTMENT
TrgTableNM1=DEPARTMENT
[Table<N>]
Sections of this type contain the list of source and target fields of the table (N stands for the table identifier in [SelectedTables]), as specified at Step 3 and Step 5 of the Wizard application.
Note: This section is only available in fixed templates (for details see Save template options).
Parameter |
Description |
SrcField<N> |
source field name |
TrgField<N> |
target field name |
FieldCount |
number of table fields selected for pumping |
where N stands for the field identifier
Example:
[Table1]
SrcField0=DepartmentID
TrgField0=DEPARTMENTID
SrcField1=Name
TrgField1=NAME
SrcField2=GroupName
TrgField2=GROUPNAME
SrcField3=ModifiedDate
TrgField3=MODIFIEDDATE
FieldCount=4
[FixedInfo]
This section stores the result script as presented at Step 6 of the Wizard application.
Note: This section is only available in fixed templates (for details see Save template options).
Parameter |
Description |
Line<N> |
script text |
where N stands for the script line identifier