[SOLVED] cop5556 Assignment 4 symbol table

Assignment 4

Due: 20 March at 11:59pm

Implement a LeBlanc-Cook symbol table by completing the given class. You should create unit tests for the given methods.

Create a Visitor class called TypeCheckVisitor by completing the given class. Your visitor should traverse the AST generated by your parser from Assignment 3 and perform type checking according to the rules described below. Your Visitor class will both decorate the tree and check conditions along the way. If a condition is violated, throw a TypeCheckException.

Global rules: no variable may be declared more than once in the same scope.

The name of the program is not a variable nameit is used to name the class implemented by the program but does not preclude the same name begin declared as a variable.

SymbolTable symbab = new SymbolTable()

Program = List<ParamDec> Block

ParamDec = type ident symtab.insert(ident.getText(), ParamDec);

Block = symtab.enterScope() List<Dec> List<Statement> symtab.leaveScope()

Dec = type ident symtab.insert(ident.getText(), Dec);

Statement = SleepStatement | WhileStatement | IfStatement | Chain

| AssignmentStatement

SleepStatement = Expression condition: Expression.type==INTEGER

AssignmentStatement = IdentLValue Expression

condition: IdentLValue.type== Expression.type

Chain = ChainElem | BinaryChain

ChainElem ::= IdentChain | FilterOpChain | FrameOpChain | ImageOpChain

IdentChain = ident

condition: ident has been declared and is visible in the current scope

IdentChain.type <- ident.type

ident.type <- symtab.lookup(ident.getText()).getType()

FilterOpChain = filterOp Tuple

condition: Tuple.length == 0

FilterOpChain.type <- IMAGE

FrameOpChain = frameOp Tuple

if (FrameOP.isKind(KW_SHOW, KW_HIDE) {

condition: Tuple.length == 0

FrameOpChain.type <- NONE

}

else if (FrameOp.isKind(KW_XLOC, KW_YLOC){

condition: Tuple.length == 0

FrameOpChain.type <- INTEGER

}

else if(FrameOp.isKind(KW_MOVE){

condition: Tuple.length == 2

FrameOpChain.type <- NONE

}

else there is a bug in your parser

ImageOpChain = imageOp Tuple

if (imageOp.isKind(OP_WIDTH, OP_HEIGHT){

condition: Tuple.length == 0

ImageOpChain.type <- INTEGER

}

else if (imageOP.isKind(KW_SCALE)){

condition: Tuple.length==1

ImageOpChain.type <- IMAGE

}

BinaryChain = Chain (arrow | bararrow) ChainElem

Legal combinations shown:

BinaryChain		Chain	op	ChainElem
type <-IMAGE	type =URL		arrow	type = IMAGE
type <-IMAGE	type = FILE		arrow	type = IMAGE
type <-INTEGER	type = FRAME		arrow	instanceof FrameOp & firstToken { KW_XLOC, KW_YLOC}
type <-FRAME	type = FRAME		arrow	instanceof FrameOp & firstToken { KW_SHOW, KW_HIDE, KW_MOVE}
type <-INTEGER	type = IMAGE		arrow	instanceof ImageOpChain) && firstToken { OP_WIDTH, OP_HEIGHT}
type <-FRAME	type = IMAGE		arrow	type = FRAME
type <-NONE	type = IMAGE		arrow	type = FILE
type <-IMAGE	type = IMAGE		arrow | barrow	instanceof FilterOpChain & firstToken {OP_GRAY, OP_BLUR, OP_CONVOLVE}
type <-IMAGE	type = IMAGE		arrow	instanceof ImageOpChain & firstToken {KW_SCALE}
type <-IMAGE	type = IMAGE		arrow	instanceof IdentChain

WhileStatement = Expression Block

condition: Expression.type = Boolean

IfStatement = Expression Block

condition: Expression.type = Boolean

Expression = IdentExpression | IntLitExpression | BooleanLitExpression| ConstantExpression | BinaryExpression

IdentExpression = ident

condition: ident has been declared and is visible in the current scope

IdentExpression.type <- ident.type

IdentExpression.dec <- Dec of ident

IdentLValue = ident

condition: ident has been declared and is visible in the current scope

IdentLValue.dec <- Dec of ident

IntLitExpression = intLit

IntLitExpression.type <- INTEGER

BooleanLitExpression = booleanLiteral

BooleanLitExpression.type <- BOOLEAN

ConstantExpression = screenWidth | screenHeight

ConstantExpression.type <- INTEGER

BinaryExpression = Expression op Expression

Legal combinations shown:

BinaryExpression.type	Expression0.type	op	Expression1.type
INTEGER	INTEGER	PLUS, MINUS	INTEGER
IMAGE	IMAGE	PLUS, MINUS	IMAGE
INTEGER	INTEGER	TIMES,DIV	INTEGER
IMAGE	INTEGER	TIMES	IMAGE
IMAGE	IMAGE	TIMES	INTEGER
BOOLEAN	INTEGER	LT,GT,LE,GE	INTEGER
BOOLEAN	BOOLEAN	LT,GT,LE,GE	BOOLEAN
BOOLEAN		EQUAL, NOTEQUAL	condition: Expression0.type = Expression1.type

Tuple = List<Expression>

condition: for all expression in List<Expression>: Expression.type = INTEGER

op = relOp | weakOp | strongOp

type = integer | image | frame | file | boolean | url

You will need to add an attribute along with get and set methods to some of the AST classes to record the type. In particular, add a TypeName field to Expression, Dec, and Chain. TypeName is defined in class cop5556sp17.AST and provided for you. Note that it is inherited by all of the subclasses of the given classes. If your Parser and Scanner have been implemented correctly, you shouldnt need to modify them.

A couple of junit tests have been provided in TypeCheckVisitorTest.java. You need to add more. As before you can work incrementally by calling the parser with a different method than parse.

Turn in a jar file containing your source code TypeCheckVisitor.java, SymbolTable.java, Parser.java, Scanner.java, all of the AST nodes, Type.java, and your TypeCheckVisitorTest.java.

Your TypeCheckVisitorTest.java will not be graded, but may be looked at in case of academic honesty issues. We will subject your classes to our set of junit tests and your grade will be determined solely by how many tests are passed. Name your jar file in the following format:

firstname_lastname_ufid_hw4.jar

Additional requirements:

• Your code must remain package cop5556sp17(case sensitive): do not create additional packages.
• The provided class Type is in package cop5556sp17.AST. As before, all of the classes in this package should be in a folder called AST inside the cop5556sp17 folder
• Names (of classes, method, variables, etc.) in the provided AST classes must not be changed.
• Unless otherwise specified, your code should not import any classes other than those from the standard Java distribution or those that are part of the project.

Submission Checklist

See the checklist from Assignment 1.

Comments and suggestions:

• Work incrementally, starting with small constructs and moving to bigger ones, and adding tests each time. As in previous assignments, you can call the routines corresponding to fragments of the grammar in Junit tests.
• You can use the firstToken field in the case of errors to provide