atr: transform characters from the std input

usage: atr [options] [input_set] [output_set]

the sets are character sets, where the definition is:
- simple character: h
- char range: first and last char, separated by a colon: a-k
- predefined ranges:
  - alnum: alphanumeric (letters and digits)
  - alpha: alphabetic (letters)
  - digit: numeric characters
  - lower: lower-case chars
  - upper: upper-case chars
  - space: whitespaces
  - xdigit: hexadecimal characters
The previous items can be combined with the comma. Range examples:
a-f,m,n * digit,p-z * a,b,c,k,1,3,5,7 * K-P,T-Z

options:
-d: delete characters in the first set from the output
-s: squeeze repeated chars
-c: use the complementer set of that in the parameter

None of the previous switches means simple replacement: the
characters in the first set will be replaced by the corresponding
one in the second set. The sets must have the same number of
items (special case: the output set consists only of a simple
character: each char from the first set will be replaced by this
one).

The replaced/modified strings ar written to the std out.

Examples:
acat test.txt | atr a,b,c z,y,x
  -> 'a' -> 'z', 'b' -> 'y', 'c' -> 'x'

... | atr upper lower
  -> 'a' -> 'A' and so on

... | atr a-f,m,n +
  -> 'a,b,c,d,e,f,m,n' -> '+'

... | atr -d a,d
  -> 'apple data' -> 'pple ata'

... | atr -s
  -> 'employee foo' -> 'employe fo'

... | atr -c f-h #
  -> each input char, except f,g,h, will be repalced by sharp sign