Doing data analysis with bash seems interesting. It's perhaps the fastest way to do quick queries. But I am wondering, how much data analysis can you do with "BASHing data"?
The main problem you will hit eventually is subtle bugs, like not handling CSV escaping correctly when, for example, the field value happens to contain a comma.
Quick and dirty and probably fine for a quick PoC if that's what you're used to, but otherwise you're probably better off with a real language with libraries designed for what you're doing.
Quick and dirty and probably fine for a quick PoC if that's what you're used to, but otherwise you're probably better off with a real language with libraries designed for what you're doing.