As always, the correct answer depends on what you have installed, what you’re familiar with, how much data you have, and whether you’re going to be doing it on a regular basis and want to automate it.

Sure, but I think there is a widespread hesitation to use a db unless one's data reaches a higher level of size and complexity, which is misguided.

What I'm saying is that it's good to start with a database from the start, even if the problem seems too trivial for it. That way, if your problem grows into something that actually needs it, you already have it in a suitable database and the code to deal with that.

xsv can be inserted into a shell pipeline at the drop of a hat, regardless of how much data you have. Using a DB is typically much more effort. Just look at the other comments in this thread. Tons of complaints about it being slow and trying to finagle indices to make it better.

Sometimes functionality isn't the only thing that's important. Expression can be just as important.