The user can write in my Bash script a mac address in the following way:
read -p "enter mac-address " mac-address
Now i want to check in an if-statement, if this mac-address matches with a "specific" format. i.e. it should be FF:FF:FF:FF:FF:FF and not FFFFFFFFFFFF. Also the length should be correct: 6x2.
The lazy way is just to run
if [[ $mac_address == ??:??:??:??:??:?? ]]; then echo Heureka; fi
but this doesn't check whether it's a hex string. So if this is important
if [[ $mac_address =~ ^[0-9a-fA-F][0-9a-fA-F]:[0-9a-fA-F][0-9a-fA-F]:[0-9a-fA-F][0-9a-fA-F]:[0-9a-fA-F][0-9a-fA-F]:[0-9a-fA-F][0-9a-fA-F]:[0-9a-fA-F][0-9a-fA-F]$ ]]; then echo Heureka; fi
might be better. The later can be shortened to
if [[ $mac_address =~ ^([[:xdigit:]]{2}:){5}[[:xdigit:]]{2}$ ]]; then
echo Heureka;
fi
If the pattern matches I don't see a need to check for the correct length as well.
[a-f] also matches on â, é (and many other characters) with bash and in most of today's locales. Better to use [[:xdigit:]] as you do for the last example, or [0-9abcdefABCDEF]
[[ $mac_address: =~ ^([[:xdigit:]]{2}:){6}$ ]]
$mac_address with a literal, then I saw it.
I changed the name of the variable so that it consisted of legal characters, then used the =~ test operator to compare the value to the extended regular expression matching: at the beginning of the string ^, "2 hex digits and a colon" 5 times, followed by 2 hex digits, ending the string $:
#!/bin/bash
read -p "enter mac-address " mac_address
if [[ $mac_address =~ ^([[:xdigit:]][[:xdigit:]]:){5}[[:xdigit:]][[:xdigit:]]$ ]]
then
echo good
else
echo bad
fi
With standard sh syntax (so would work with bash and any other POSIX compatible shell):
x='[[:xdigit:]]'
mac_pattern="$x$x:$x$x:$x$x:$x$x:$x$x:$x$x"
printf >&2 'Please enter a MAC address: '; read mac
case $mac in
($mac_pattern) echo OK;;
(*) echo >&2 BAD;;
esac
