PASSTHRU | NOPASSTHRU
Valid for
Extract
Description
Use PASSTHRU
and NOPASSTHRU
to control whether a data-pump Extract processes files in pass-through mode or normal mode. In pass-through mode, the Extract process does not look up file and table definitions, either from the database or from a data-definitions file. In normal mode the definitions are used to perform mapping and conversion functions.
Using pass-through mode, you can cascade the captured data to a data pump on an intermediary system that has no database installed on it. Source and target file names and structures must be identical; no filtering, column mapping, SQLEXEC
functions, transformation, or other functions requiring data manipulation or translation can be used.
You can use PASSTHRU
to allow Extract to move data after a file is purged or renamed. By default, Extract does not output data for purged or renamed source files, because no definition is available for the file. However, there are instances when this is appropriate, as when Extract distributes data created by Logger to other systems or acts as a data pump.
To ensure your trails process correctly, you must set the PASSTHRU
parameter before you set your EXTTRAIL
or RMTTRAIL
parameters. Later in the parameter file, you can apply NOPASSTHRU
then set up your where clauses or filters on files that require them. Use PASSTHRU
only when Extract is distributing data already in a Oracle GoldenGate trail (LOGTRAILSOURCE
/EXTTRAILSOURCE
). Do not use PASSTHRU
in combination with CUSEREXIT
or COBOLUSEREXIT
.
Enter PASSTHRU
above any FILE
entries that may fit this description.
Default
NOPASSTHRU
Syntax
PASSTHRU | NOPASSTHRU
Example
The following example passes through all data from $DATA.DAT
even if the source file no longer exists. Only account codes less than 100 are distributed from the ACCOUNT
file.
EXTTRAIL \XYZ.$DATA3.GGSDAT.ET PASSTHRU FILE $DATA1.DAT.*; NOPASSTHRU FILE $DATA2.DAT2.ACCOUNT, DEF ACCOUNT-REC, WHERE (ACCOUNT-CODE < 100);