For bash or ksh, with strings of [a-z]. Using base-36 to pluck the bitstrings out of huff[], has the added bonus of flattening the case for you. As such:
$ _decode $(_encode tHeQuIcKbRoWnFoX) thequickbrownfox $
Enjoy!
_encode() {
typeset in="${1:?args}"
typeset out=""
typeset huff=(
1110 101100 01000 11111 # a b c d
011 00110 111100 0101 # e f g h
1100 001011001 0010111 10111 # i j k l
00111 1010 1101 101101 # m n o p
001011010 1000 1001 000 # q r s t
01001 001010 111101 001011011 # u v w x
00100 001011000 # y z
)
typeset -i i=0
while (( i < ${#in} )); do
out+="${huff[36#${in:i++:1} - 10]}";
done
echo "${out}"
}
_decode() {
typeset d="${1:?args}"
typeset out=""
while [[ "$d" ]]; do
d="${d/#000/t}" d="${d/#011/e}" d="${d/#0101/h}" \
d="${d/#1000/r}" d="${d/#1001/s}" d="${d/#1010/n}" \
d="${d/#1100/i}" d="${d/#1101/o}" d="${d/#1110/a}" \
d="${d/#00100/y}" d="${d/#00110/f}" d="${d/#00111/m}" \
d="${d/#01000/c}" d="${d/#01001/u}" d="${d/#10111/l}" \
d="${d/#11111/d}" d="${d/#001010/v}" d="${d/#101100/b}" \
d="${d/#101101/p}" d="${d/#111100/g}" d="${d/#111101/w}" \
d="${d/#0010111/k}" d="${d/#001011000/z}" d="${d/#001011001/j}" \
d="${d/#001011010/q}" d="${d/#001011011/x}"
out+="${d:0:1}"
d="${d/#?/}"
done
echo "${out}"
}
