Neat. I did a similar thing for analysing the output of commands that can produce JSON (SLURM in this case). Worked pretty well but I immediately ran into PRQL's lack of support for DuckDB's struct types.
Yes, there's definitely some downsides (which honestly, we didn't realize when we started to build this). For us, it was the lack of a PIVOT statement so I had to make a modified dialect of PRQL that supports PIVOT, then split on the final PRQL into PIVOT and non-PIVOT chunks, and then convert it to chained SQL CTEs. Annoying work for sure.
As a result, I'll be adding support for PIVOTs very soon, either in the main repo or just in a temporary fork.
