Re: Better way to process records in bash?

## Ron Johnson (ronljohnsonjr@gmail.com):

> I need to process table records in a bash script.  Currently, I read them
> using a while loop and redirection.  The table isn't that big (30ish
> thousand rows), and performance is adequate, but am always looking for
> "better".

Use python, or any other sane language, and don't do bash in python.
Anything non-trivial is unreasonably hard to get really right when
using shell.

Take this example (even when allowing some fuzz with the variables):
> Here's the current code:
> declare f1 f3 f8
> while IFS='|' read f1 f3 f8; do
>     something f8 f3 f1
> done < <(psql -XAt -c "select f1, f3, f8 from some.table_name;")

What happens when f3
 a) is NULL?
 b) contains a '|' character?
 c) contains spaces? (this one depends on how exactly you're handling
    the shell variables).
This might be fine if your data is only NOT NULL numbers, but that's
already quite a limitation on your data.