Shell Programming and Scripting

What about this then:

Code:

A=( $valFile )

You can then blindly use A[0] or check/error when [ ${#A[@]} -ne 0 ]

Code:

[ -f "${A[0]}" ] && valid="$(<"${A[0]}")" || valid="V"

In theory, you should be able to close stderr with

Code:

2>&-

But that doesn't seem permanent - tracing bash execution shows that stderr is saved off using dup2, then "restored" using another dup2 call after it's closed.

Maybe you should just start your shell with stderr redirected to /dev/null?

I would avoid throwing out all errors to avoid 1 particular message that can be coded for anyway .

You don't want to miss any important error messages that can help you debug/correct future issues coming out of the script.

You know, $( ) isn't exactly a "built in" either, a fork and a subshell is involved.

How much data is in each of these files? Are they single lines? You might be able to use read instead of <, which WOULD be a builtin. It could be dramatically faster when there's thousands to do. Or perhaps readarray if there's multiple lines.

Reading thousands of lines into the shell at once may not be the best plan in any case, however, especially if the files can be used directly when you need them instead.