Getting "unruly": Finding base words

[tibit]

With your command you're not lowercase'ing stuff like german umlauts... (Ä --> ä, Ö --> ö etc.). But I'm not sure if the corresponding rule (toggle) does it... I have to check it out.

Based on the epixoip's code, you could use 'sed' instead understand foreign characters like German, French, Turkisch ..

The code then becomes

Code:

sed 's/[[:upper:]]*/\L&/g' infile | sed 's/^[^[:lower:]]*//g; s/[^[:lower:]]*$//g; y/112345677890@\$\!\#/ilzeasbzvbgoasih/; s/[^[:lower:]]//g; /^$/d' >outfile