Data transformation

Now that we understand data mapping, we can perform data transformation in our GoldenGate configuration. Using the built-in string manipulation and number conversion functions, with the COLMAP option, it is possible to achieve the most common data transformations: truncation, concatenation, substitution, conversion, and case changes. Note that the target column is always on the left in the COLMAP statement.

Data transformation does come at a price though—performance. When conducting complex or numerous simple transformations with high data volume and throughput, latency will prevail. It may not be as significant a bottleneck as a serial process, but the CPU consumption on the target system will increase.

Truncation and extraction

Starting with truncation, Oracle performs implicit data type conversion between a source and target column. However, should the target column scale be smaller than its source, the data is truncated on the right.

GoldenGate offers the @SUBEXT function...