public override void Assignment(Token varName, bool withArrayIndex)
{
Output.WriteLine(
"ASSIGNMENT TO {0}{1}",
varName.Value,
withArrayIndex ? "[...]" : "");
}
public override void Assignment(Token varName, bool withArrayIndex)
{
//This is slightly tricky. We need to have the address of the
//variable to which we're doing the assignment on the stack,
//and then everything is really easy. If it's an array, we need
//the address of the array element to which we're doing the
//assignment (its index is on the stack following the LHS value).
if (MethodSymTable.HasSymbol(varName.Value))
{
Symbol symbol = MethodSymTable.GetSymbol(varName.Value);
Contract.Assert(symbol.Kind == SymbolKind.Local ||
symbol.Kind == SymbolKind.Parameter);
Output.WriteLine(" __PUSH((__WORD)&{0});", symbol.Name);
}
else
{
Contract.Assert(ClassSymTable.HasSymbol(varName.Value));
Symbol symbol = ClassSymTable.GetSymbol(varName.Value);
if (symbol.Kind == SymbolKind.Static)
{
Output.WriteLine(" __PUSH((__WORD)&{0});", FormatStaticName(symbol.Name));
}
else if (symbol.Kind == SymbolKind.Field)
{
Output.WriteLine(" __PUSH((__WORD)&((({0}*)THIS)->{1}));", _currentClassName, symbol.Name);
}
}
//If it's an array, obtain the address of the right element by
//adding the index which is on the stack. We need a scratch
//location because issuing two __POP() calls in the same statement
//does not guarantee left-to-right evaluation.
if (withArrayIndex)
{
//The array address is now on the stack, but we really need the
//address of the first element. Hence the dereference:
Output.WriteLine(" __SCRATCH2 = *(__WORD*)__POP();");
//This is the RHS value that we ought to put in the array element:
Output.WriteLine(" __SCRATCH1 = __POP();");
//Finally, the top of the stack contains the value of the array
//indexing expression, i.e. the element index:
Output.WriteLine(" * ( ((__WORD*)__SCRATCH2) + __POP() ) = __SCRATCH1;");
}
else
{
Output.WriteLine(" __SCRATCH1 = __POP();"); //This is the LHS
Output.WriteLine(" * ((__WORD*)__SCRATCH1) = __POP();");
}
}
public abstract void VariableRead(Token varName, bool withArrayIndex);
public abstract void Assignment(Token varName, bool withArrayIndex);
public void Advance()
{
EatWhitespace();
if (IsAtEnd)
{
_done = true;
return;
}
char nextChar = NextChar();
if (Syntax.IsSymbol(nextChar))
{
//This token is going to be a symbol. There are three special
//look-ahead cases for '<=', '>=', and '!='.
if ((new[] { '<', '>', '!' }.Contains(nextChar)) && LookAhead() == '=')
{
NextChar();//Eat the '='
_currentToken = new Token(TokenType.Symbol, nextChar + "=");
}
else
{
_currentToken = new Token(TokenType.Symbol, nextChar.ToString());
}
}
else if (Syntax.IsNumber(nextChar))
{
//This token is going to be an integer constant.
string intConst = nextChar.ToString();
intConst += EatWhile(Syntax.IsNumber);
int result;
if (!int.TryParse(intConst, out result))
{
throw new CompilationException(
"Integer constant must be in range [0,2147483648), but got: "
+ intConst, _currentLine);
}
_currentToken = new Token(TokenType.IntConst, intConst);
}
else if (Syntax.IsCharOrdinalStart(nextChar))
{
//This is an addition to the 'standard' Jack language. We
//support also character ordinals of the form 'H' or of the
//form '\nnn' where nnn are decimal digits. They are translated
//to integer constants as far as the parser is concerned.
char marker = NextChar();
if (marker == '\\')
{
string code = EatWhile(Syntax.IsNumber);
if (code.Length != 3)
{
throw new CompilationException("Expected: \\nnn where n are decimal digits", _currentLine);
}
int value = int.Parse(code);
if (value >= 256)
{
throw new CompilationException("Character ordinal is out of range [0,255]", _currentLine);
}
_currentToken = new Token(TokenType.IntConst, value.ToString());
}
else
{
_currentToken = new Token(TokenType.IntConst, ((int)marker).ToString());
}
NextChar();//Swallow the end of the character ordinal
}
else if (Syntax.IsStringConstantStart(nextChar))
{
//This token is going to be a string constant.
string strConst = EatWhile(c => !Syntax.IsStringConstantStart(c));
NextChar();//Swallow the end of the string constant
_currentToken = new Token(TokenType.StrConst, strConst);
}
else if (Syntax.IsStartOfKeywordOrIdent(nextChar))
{
//This is going to be a keyword or an identifier. We can't
//tell what it's going to be until we accumulate more of it.
//We need to read until we encounter whitespace.
string keywordOrIdent = nextChar.ToString();
keywordOrIdent += EatWhile(Syntax.IsPartOfKeywordOrIdent);
if (Syntax.IsKeyword(keywordOrIdent))
{
_currentToken = new Token(TokenType.Keyword, keywordOrIdent);
}
else
{
_currentToken = new Token(TokenType.Ident, keywordOrIdent);
}
}
else
{
throw new CompilationException("Unexpected character: " + nextChar, _currentLine);
}
}
private void VerifyArrayAccessAllowed(Token varName)
{
Symbol variable = GetClosestSymbol(varName.Value);
if (variable.Type != "Array")
{
ThrowCompilationException("Only arrays may be indexed with '[]'");
}
}
//sub-call ::= <sub-name> '(' <expr-list> ')' |
// (<class-name> | <var-name>) '.' <sub-name> '(' <expr-list> ')'
//expr-list ::= (<expression> (',' <expression>)*)?
private void ParseSubCall(Token firstPart = null)
{
//NOTE: Because the parser may inadvertently swallow the first part
//of the subroutine call expression (i.e. the subroutine name,
//class name, or variable name), there's an option for this method
//to accept it explicitly instead of reading it from the tokenizer.
//
//After this method completes, the code generator should have
//available to it the result of the subroutine's invocation (e.g.
//on the stack or in a predefined register). If the subroutine
//call is not part of an expression (i.e. it's part of a 'do'
//statement), then this result should be discarded via an explicit
//call to the code generator.
if (firstPart == null)
{
firstPart = NextToken();
}
if (firstPart.Type != TokenType.Ident)
{
ThrowCompilationException("Expected an identifier, got: " + firstPart.Value);
}
string classNameOfSubroutine = null;
string subroutineName = null;
bool isInstanceMethod = false;
Symbol classOrVarName = GetClosestSymbol(firstPart.Value);
if (classOrVarName == null)
{
if (LookAheadToken.Value == "(")
{
//If this is the case, this is a direct method call within
//the current class. The current subroutine has to be
//non-static (i.e. a method or a constructor) for us to
//dispatch a method call. Because we're a one-pass compiler,
//we can't verify that the method actually exists.
if (_currentSub.Kind != SubroutineKind.Constructor &&
_currentSub.Kind != SubroutineKind.Method)
{
ThrowCompilationException("Method calls are allowed only within a method or a constructor");
}
classNameOfSubroutine = _currentClassName;
subroutineName = firstPart.Value;
isInstanceMethod = true;
}
else if (LookAheadToken.Value == ".")
{
//This is a function (or constructor) call on a class name.
//We don't know about all class names, so we have to assume
//that this class exists. The assembler will tell us if
//we're wrong. Alternatively, we could implement a multi-
//pass compiler that goes over only declarations, or require
//all subroutines and classes to be declared before they
//can be called.
//NOTE: If we want to do inlining as an optimization, we
//must perform two passes (or else we don't know what's in
//the subroutine we're placing inline :-)).
Match(new Token(TokenType.Symbol, "."));
Token subName = NextToken();
if (subName.Type != TokenType.Ident)
{
ThrowCompilationException("Expected an identifier, got: " + subName.Value);
}
classNameOfSubroutine = firstPart.Value;
subroutineName = subName.Value;
isInstanceMethod = false;
}
else
{
ThrowCompilationException("Unexpected: " + LookAheadToken.Value);
}
}
else
{
//This is a method call on a variable. We must be able to tell
//the class name by looking at the variable's type. Note that
//built-in types that we recognize as keywords don't have any
//methods, so this is an error we can flag easily.
classNameOfSubroutine = classOrVarName.Type;
if (Syntax.IsKeyword(classNameOfSubroutine))
{
ThrowCompilationException("Can't call methods on built-in types");
}
Match(new Token(TokenType.Symbol, "."));
Token subName = NextToken();
if (subName.Type != TokenType.Ident)
{
ThrowCompilationException("Expected an identifier, got: " + subName.Value);
}
subroutineName = subName.Value;
isInstanceMethod = true;
}
Match(new Token(TokenType.Symbol, "("));
//If this is an instance method, we need to supply 'this' as the
//first parameter to be extracted on the other side.
if (isInstanceMethod)
{
if (classNameOfSubroutine == _currentClassName)
{ //'this' is actually our very own instance, as we're calling
//an instance method on ourselves.
_codeGenerator.This();
}
else
{ //'this' is actually the variable on which we're making
//the call, so we need that pushed.
_codeGenerator.VariableRead(firstPart, false);
}
}
ParseExpressionList();
Match(new Token(TokenType.Symbol, ")"));
_codeGenerator.Call(classNameOfSubroutine, subroutineName);
//TODO: Detect the situation where a void subroutine is called
//inside an expression where a value has to be provided. This is
//impossible without having the subroutine's declaration in front
//of us. To alleviate this, we can have all subroutines, even void
//ones, return some arbitrary value (e.g. 0).
}
private void MatchCurrent(Token token)
{
if (_tokenizer.LookAheadToken.Type != token.Type ||
_tokenizer.LookAheadToken.Value != token.Value)
{
Expected(token.Value);
}
}
private void Match(Token token)
{
if (!_tokenizer.HasNext)
Expected(token.Value);
Token actual = _tokenizer.Next();
if (actual.Type != token.Type ||
actual.Value != token.Value)
{
Expected(token.Value);
}
}
public override void VariableRead(Token varName, bool withArrayIndex)
{
Output.WriteLine(
"READ FROM {0}{1}",
varName.Value, withArrayIndex ? "[...]" : "");
}
public override void VariableRead(Token varName, bool withArrayIndex)
{
//Put the value of the variable on the top of the stack. If it's
//an array, the value is the address of the array's first element.
if (MethodSymTable.HasSymbol(varName.Value))
{
Symbol symbol = MethodSymTable.GetSymbol(varName.Value);
Contract.Assert(symbol.Kind == SymbolKind.Local ||
symbol.Kind == SymbolKind.Parameter);
Output.WriteLine(" __PUSH({0});", symbol.Name);
}
else
{
Contract.Assert(ClassSymTable.HasSymbol(varName.Value));
Symbol symbol = ClassSymTable.GetSymbol(varName.Value);
if (symbol.Kind == SymbolKind.Static)
{
Output.WriteLine(" __PUSH({0});", FormatStaticName(symbol.Name));
}
else if (symbol.Kind == SymbolKind.Field)
{
Output.WriteLine(" __PUSH((({0}*)THIS)->{1});", _currentClassName, symbol.Name);
}
}
//If it's an array, dereference it using []. We need a scratch
//location because issuing two __POP() calls in the same statement
//does not guarantee left-to-right evaluation.
if (withArrayIndex)
{
Output.WriteLine(" __SCRATCH1 = __POP();");
Output.WriteLine(" __PUSH( ((__WORD*)__SCRATCH1)[ __POP() ] );");
}
}