sed 's/./&\n/g' ids.txt | sort | uniq # Produce a list of all the unique characters used in a file. Useful if you need to try to figure out the expected character set and can't find a documented standard.

@climagic can't you use sort -u instead of having to pipe the output through uniq?