 
  
  
  
  
 
We implement substrings of strings in the following way. Consider the following sequence of commands:
The first command defines a string whose name is `A'. The second command defines a 
second string, called `B', whose definition part includes a substring extracted from the 
string whose name is `A'. Indeed, the syntax `  ' means that we should replace 
`
 ' means that we should replace 
`  ' by the third, fourth, fifth, and sixth characters of the string `A'. Thus, `B' 
will be defined as `tuvcdefwxyz'.
 ' by the third, fourth, fifth, and sixth characters of the string `A'. Thus, `B' 
will be defined as `tuvcdefwxyz'.
In general, if
  is detected as input, then `name' is assumed to be the
name of a string and a and b to be integers (or equations which will be rounded to 
integers) marking the desired substring of this named string. The
  is detected as input, then `name' is assumed to be the
name of a string and a and b to be integers (or equations which will be rounded to 
integers) marking the desired substring of this named string. The   to the
  to the 
  characters inclusive of this named string then replace the piece of input in square 
brackets. The square brackets are also removed.
  characters inclusive of this named string then replace the piece of input in square 
brackets. The square brackets are also removed.
The rules governing
substrings are as follows. Suppose that we encounter the sequence of characters 
  in a command line. N should be the name of a string and a and b are equations 
which will be rounded to the nearest integer. Suppose that L is the length of the string
whose name is N.
 
in a command line. N should be the name of a string and a and b are equations 
which will be rounded to the nearest integer. Suppose that L is the length of the string
whose name is N.
 is the substring formed by the characters
  is the substring formed by the characters   of N  whenever
  of N  whenever   
  is simply removed.
  is simply removed.