A not very famous writer Arthur Conan Moyle has finally found out why his books
are not as popular as he believes they would deserve. He noticed that his works are
getting a bit boring due to frequent repetitions of same or similar pieces of
text. He decided that the best way how to improve the quality of his books is
to simply throw away everything after the first repetition - the books then
get an interesting open-ended feeling.

First he attempted to look for exactly same pieces of text, but this failed,
since the repeated texts often do not match precisely - some of letters that are
lowercase in the first place may be uppercase in the second and vice versa,
punctuation may be a bit different, and even the words in sentences may be in
slightly different order. To overcome these problems, he devised the following
more involved criterion for recognizing duplicates (a positive integer k is a parameter
of his criterion; by changing it he affects how long repeated pieces are still acceptable):

Alphabetic characters are the letters 'a'-'z' and 'A'-'Z'. We do not distinguish
case of the letters, i.e. 'a' is supposed to be the same letter as 'A'.

Two strings S₁ and S₂ are k-identical up to permutation of letters if:

• Both S₁ and S₂ start and end with an alphabetic character
• Both S₁ and S₂ contain exactly k alphabetic characters
• For each alphabetic character c, the string S₁ contains the same number of occurrences of c as the string S₂.

In other words, if the strings S₁ and S₂ are k-identical up to permutation of letters,
then the alphabetic characters in them are the same, but their ordering may be different.

Your task is to write a program that separates a longest initial part that does not
contain two substrings k-identical up to permutation of letters from several of
the ACM's books.