Reference no: EM132375291

Assignment

1. Purpose

The purpose of this assignment is to write a Java program which will allow a s-grammar (see slide 15 of the CFL slide set) G = (V,T,S,P) and a string to be inputted. The program should check that the grammar is a valid s-grammar and then parse the string to determine whether the string is a member of the language that is defined by the grammar.

All elements in the set V of variables and set T of terminals consist of single characters. Variables in V consist of upper-case alphabetic characters. Terminals in T consist of lower-case alphabetic characters and special characters like { = + etc. The set P of production rules do not include ,-productions, but otherwise can include any valid s-grammar rules.

2. Program Details
The program must have a GUI interface which will allow the grammar to be inputted. The program should then check that the grammar is a valid s-grammar. If it is, strings should be inputted and the program should then parse them, i.e. check whether they belong to the language that the grammar defines, by performing a leftmost derivation on them. The program must also have a facility to display each step of the leftmost derivation as it is occurring.

You might find it useful to check out JFlap's Grammar option which has quite a nice interface. Any s-grammar (like grammars G1 and G2 in section 3 below) is actually usable in JFLap as an LL(1) grammar, since s-grammars are special cases of LL(1) grammars.

In addition, s-grammars and strings to be parsed must be able to be read in from a data file.

You should use a common internal representation for s-grammars and strings to be parsed, so that it will make no difference to your program whether they are inputted via the GUI interface or via a data file.

3. Testing

You must test your code on the following two s-grammars, though of course you can (and should) create or find others as well on which to test your code. Note that all productions in these grammars follow the requirements for s-grammars in that
• there is a single variable on the left hand side of a production.
• the right hand side of a production starts with a single terminal and is followed by 0 or more variables.
• if there are two or more productions with the same variable on the left hand side, the right hand of each of these productions starts with a different terminal.

G₁ is a s-grammar
G₁ = ( { S,A,B }, { a,b }, S, P )

for the language L = {aⁿbⁿ⁺¹ : n $1}, with the elements of P being the productions

S          6    aAB

A          6    aAB | b

B           6    b

G₂ is a s-grammar

G₂ = ( { S,R,L,C,A,Y,P,Z,X,T,D }, { {,},x,=,y,+,z,i,t,e,d }, S, P }

for a fragment of a programming language, with the elements of P being the productions

S          6    {LR

R          6    }

L          6    xAYPZ | iXTLC

C          6    d | eLD

A         6    =

Y         6    y

P          6    +

Z          6    z

X         6    x

T          6    t

D         6    d

A program fragment takes the form

{ Stmt }

where Stmt can have the one of the forms

(i) x = y + z

(ii) if x then Stmt endif

(iii) if x then Stmt else Stmt endif

E.g a program fragment might be

{ if x then
x = y + z
else
if x then
x = y + z
endif endif }

Obviously, since both variables and terminals are specified to consist of single characters only, any multi-character variables or terminals like Stmt or endif must be represented by single characters. The following scheme is used

word            symbol

Stmt               L

if                         i

then               t

else                   e

endif                 d

The above program fragment is then represented by the string

{ixtx=y+zeixtx=y+zdd}

You could decide in your implementation that whitespace is not significant within strings to be parsed. If that were the case, then the string might have the (slightly) more user-friendly format

{ i x t x=y+z e i x t x=y+z d d }

4. Data Structures and Algorithms

In any leftmost derivation, the essential requirement is to replace the leftmost variable in the current sentential form with the right hand side of a production with that variable on the left hand side. In an s-grammar, the choice of which production to use is made very simple. If a is the next symbol in the input string being parsed, and A is the leftmost variable in the current sentential form, we have to find a production with A on the left hand side whose right hand side starts with a. By the nature of an s-grammar there can be only one such production. If there is no such production the string cannot be a member of the language that the grammar defines.

Design decisions you should consider include
- how sentential forms should be represented so that it is easy to find and replace the leftmost variable in the sentential form.
- how productions should be represented so that it is easy to find the production that is to be used in the replacement of the leftmost variable in the sentential form. The restriction that variables must consist of uppercase single characters means there can be no more than 26 variables here, but in a real grammar there may be dozens or even hundreds of variables. A method of representing and retrieving productions efficiently is therefore important.