SED multiple characters inside a getops

Question

I'm trying to create a script that can use a command such as

script -c abcd test.txt

and it will get rid of any characters a or b or c or d inside test.txt.

This is the code I have so far:

while getopts ":c:" opt; do
  case $opt in
    c)
      sed 's/'$OPTARG'//g'
      ;;
    \?)
      echo "Invalid option: -$OPTARG" >&2
      exit 1
      ;;
  esac
done

But this only gets rid of "abcd" inside my file, not a or b or c or d, not acdb etc. How do I make SED read the OPTARG as seperate characters?

Answer 1

Use tr instead:

tr -d "$OPTARG"

will delete all characters in OPTARG.

Or if you insist on using sed, use character sets:

sed "s/[$OPTARG]//g"

Answer 2

You need to group the characters together so that they're a character class.

ross@ross-laptop ~> cat foo
agkdjbdefghijglkthdn

ross@ross-laptop ~> sed 's/[abcd]//g' foo
gkjefghijglkthn

Hope that helps.