How do you handle error rows in a VGL pipeline?

In a VGL (Visual Graph Language) pipeline, error rows are typically handled by adding a dedicated error handling node. When a transformation fails, the row is redirected to an "Error" branch that writes the record to an error table or file along with metadata such as error code, message, and timestamp. The main flow continues processing the remaining rows. You can also use VGL’s built‑in validation steps to flag rows that violate business rules and route them to the error branch. Periodically, a cleanup job can review the error table, attempt automatic fixes (e.g., trimming whitespace), or trigger alerts for manual intervention. This approach keeps the pipeline robust while preserving problematic data for later analysis.