sed/grep to extract word between 2 characters

Question

My sentence is

abcde^10=fgi^11=jkl^12=lmn 

The sequence can be jumbled also for eg.

abcde^11=jkl^12=lmn^10=fgi

The sequence can also be :

12=lmn^11=jkl^10=fgi^abcde 

I want to print lmn.

I have gotten till thus far:

~$ echo abcde^10=fgi^11=jkl^12=lmn|grep -o '\^12=[^)]*\^' -> this doesnt print anything 

~$ echo abcde^11=jkl^12=lmn^10=fgi|grep -o '\^12=[^)]*\^' -> this prints ^12=lmn^

Answer 1

With GNU awk for multi-char RS:

$ echo 'abcde^10=fgi^11=jkl^12=lmn' | awk -v RS='[\n^]' -F'=' '$1==12{print $2}'
lmn

$ echo '12=lmn^11=jkl^10=fgi^abcde' | awk -v RS='[\n^]' -F'=' '$1==12{print $2}'
lmn

or using any awk:

$ echo 'abcde^10=fgi^11=jkl^12=lmn' | awk -v RS='^' -F'[=\n]' '$1==12{print $2}'
lmn

$ echo 'abcde^10=fgi^11=jkl^12=lmn' | awk -v RS='^' -F'[=\n]' '$1==12{print $2}'
lmn