I attempted something similar with large rosbag files, but the performance gain wasn’t enough to make the initial processing worth it. That said, I’d like to tr

I attempted something similar with large rosbag files, but the performance gain wasn’t enough to make the initial processing worth it. That said, I’d like to try again with duckdb as that seems like it might be better suited to time series analysis.