Grep help

I am not sure if this is the right spot for this question, I am sorry if I got the wrong place.


I am trying to create a grep that will do the following:

I have a file with a list of words, I am trying to create a grep to find all the words that (for example) have the letters A,B,C or D in it but no other letters.

So it would return a list of words like CAB,DAB etc. but not BAT.

Hope that makes sense.

Any help would be appreciated.

Though not being the most elegant solution,
grep '[A-D][^a-zE-Z0-9][A-D]' filename(s)
lists the lines with only the correct words but listing *only* the words and not the lines is sth else, if you're asking that..

a sed approach:
first sed collect all strings containig ABCD
the second delete all strings NOT containing ABCD

sed -ne '/[ABCD]/p' inputfilename | sed -e '/[^ABCD]/d'
if you use variables:
GetIt=ABCD
sed -ne "/[$GetIt]/p" inputfilename | sed -e "/[^$GetIt]/d"

grep is sure able to do something similar, for this kind
of jobs i prefer sed.