I wrote into the Safe team to ask the question but thought it might be valuable to ask here since I haven't come across the question in the Q & A forum yet.

Our team is working to process large amounts of data (such as parcels for the U.S.) and we've used the GeometryValidator Transformer as part of our process. We've analyzed repaired data and determined that we're ok with the automated repairs performed by the transformer and have loaded those to our database.

However, we're uncertain how to move forward with data that has pass through the rejected port. While the data contains new issue attributes and traits, it's hard to visually determine (in some cases) what issues were detected and how to correct them. We've had some luck exposing the issues attribute and exporting the .ffs to .gdb so we can analyze, and possible edit the un-repaired data. However, sub-issues or issue traits we've been unsuccessful in exposing so it limits our ability to narrow the problem down. More importantly, we can't always visually see the problem.

Are there steps that we can take to make it easier to visually see the detected issues and are their known ways to automate the correction of geometry validation issues other than what is performed when the data passes through the transformer? Due to the scale of our data, it may not be possible for us to correct each issue manually. If necessary though, we'd like to better see what corrections need to be made. Some of our data is beyond our ability to repair (random stray arcs or lines straying out in space) but we'd like to preserve and use as much of the data as possible.