Cryptology Pattern Search with Perl

a7Hzq .#5r< kÜ\as TâÆK$ ûj(Ö2 ñw%h: Úk{4R f~`z8 ¤˜Æ+Ô „&¢Dø

Introduction: Numerical Patterns of Strings

We want to describe letter patterns by numbers. Here is an example: The word statistics defines the pattern 1232412451.

For the mathematical background of the following algorithm see here.

Algorithm

Goal: Determine the numerical pattern of a string.

Input: String as list string = (a₁,...,a_q).

Output: Numerical pattern as list pattern = (n₁,...,n_q).
   Initial value: empty list.

Auxiliary variables:

• n = current number; initial value: n = 0.
• assoc = liste of processed different letters;
  the index i belongs to the letter assoc[i];
  Initial value: empty list.

Procedure:
   Loop over the letters of string; current letter is x:
      If there is an i with x = assoc[i]
         then append i to pattern;
      else:
         increment n,
         append n to pattern;
         append x to assoc.

Here is a Perl program that implements this algorithm. It only serves for understanding the following more complex program.

Building a Regular Search Expression for a Pattern

The most efficient approach to pattern searching with Perl uses so called regular expressions. Here is the mninimum explanation for the following program:

Pattern Search Program

The complete program, written in the typical, highly condensed style usual with Perl, is here.

You may call it online from a web form and search in a dictionary or a text of your choice.

Exercise. Try to understand this program.

Exercise. Try the programm.