Yeah I also do the same, import the csv (it doesn't have to be comma separated) file into PostgreSQL and analyze it using SQL. Alternatively, I will load it into IPython/pandas and work with it like that.

This is however useful for one off, throwaway query that offers familiar SQL syntax, if you don't want to use awk that is.