Thank you for your comment. I think X-SQL based extraction could also be useful: its ideal for high-complexity data-extraction pipelines with multiple-dozen entities and several hundred fields per entity. What are your thoughts on this idea?