Reply to thread

Message: [QUOTE="Stefan Ram, post: 2984394"] I have not fully understood your design, maybe I read not careful enough. I am also writing something like this, but have not yet published the full code. At least, I can try to quote relevant parts of the code here. Since I have not looked at this code for several month, it looks strange to me, too, but at least i already have written some documentation earlier. I am working with two stacks: on operator stack and one value stack (the »designation stack«). Operators and Values from the source code are considered to be »Tokens«. And a token is expected to know how to »enroll« itself to the stacks. This is what happens when parsing the source code to an internal representation. The interface »Token«: interface Token extends de.dclj.ram.notation.programming.Designation { /** Enroll this token to the parser. Every token knows how to enroll itself to the parser. The stacks might be somewhat reduced, and then the token might push itself on one of the stacks. @param operatorStack the operator stack to be used for enrolling @param designationStack the designation stack to be used for enrolling */ public void enroll ( final PlacementHint placementHint, final OperatorStack operatorStack, final DesignationStack designationStack ); /** Whether this token is a Prefix context. +1 means that the next operator must be prefix. This is the case: At the start of a full expression. After opening parentheses (like, for example, »(«, »[«, »{«). After prefix or infix operators (like, for example, »+«, »*«, »/«, »-«, »,«, »;«) After prefixed function symbols (like, for example, »SIN«, »LOG«) 0 means that the next operator must be infix. This is the case: After a numeral (like, for example, »0«) After a closing parentheses (like, for example »)«, »]«, »}«) */ public boolean isPrefixContext(); } When the token is an operator, it usually is enrolled as follows. The method »acceptOperator«: /** An operator is accepted by a stack pair given a placement hint. The operator is pushed on the operator stack. Possibly the stacks will be reduced before this happens (depending on the priority of the given operators). @param operatorToken an operator to be accepted @param placementHint an instance of OperatorMustBePrefixPlacementHint if this operator token appears in a context where a prefix operator is expected @param operatorStack a stack of operators @param designationStack a stack of designations */ public static void acceptOperator ( final OperatorToken operatorToken, final PlacementHint placementHint, final OperatorStack operatorStack, final DesignationStack designationStack ) { int rightPriority; if( placementHint instanceof OperatorMustBePrefixPlacementHint ) { rightPriority = getOperatorRightPriorityInPrefixContext( operatorToken ); } else { rightPriority = getOperatorRightPriorityInInfixContext( operatorToken ); } while( isReducible( operatorStack, rightPriority )) { OperatorToken operator = operatorStack.pop(); operator.reduce( placementHint, operatorStack, designationStack ); } operatorStack.accept( operatorToken ); } This takes care of the fact that an operator character, like »-« might denote both a prefix operator and an infix operator. Therefore, context is used to figure out which to assume and then the appropriate priority is retrieved. The above while loops reduces the stack as long as indicated by the priority of the operators seen. Eventually, the operator then is pushed onto the operator stack by the final »accept«. A value, like a literal, is just pushed onto the operand stack: The methode »acceptLiteral«: public static void acceptLiteral ( final LiteralToken literalToken, final PlacementHint placementHint, final OperatorStack operatorStack, final DesignationStack designationStack ) { designationStack.accept( literalToken.value() ); } Some parameters are not used above, but given for consistency with the »acceptOperator« call. So, back to the »reduce« operation: Every operator token knows how to reduce itself to a designation of its operation (this is not the evaluation yet, but an internal representation of this application of this oprator including its operands): The interface »OperatorToken«: /** An operator token. There will be several kinds of operator tokens. */ interface OperatorToken extends Token { /** Reduces the stack by this operator token. It is assume that this token already has been removed from the operator stack. Now it will usually pop some designations from the designation stack, possibly evaluate some of them (»strict« arguments) and then build a reduction result and push it onto the designation stack. For example, »+« might take two designations, which might might be the number 2 and the number 3 and then push the number 5 on the designation stack. @param placementHint A placement hint indicates whether the reduction takes place in a context where certain types of operators are required or permissible see {@link PlacementHint}. */ /* todoc: @params */ public void reduce ( final PlacementHint placementHint, final OperatorStack operatorStack, final DesignationStack designationStack ); } An example is the »PRINT« keyword in a BASIC-like language. It takes its operand from the designation stack and pushes a new »print designation« on the designation stack. The method »reduce«: public void reduce ( final PlacementHint placementHint, final OperatorStack operatorStack, final DesignationStack designationStack ) { designationStack.accept ( new PrintDesignation ( designationStack.popOrProduceNull() )); } The »print designation« is an object representing the print operation including its argument. It has an »eval« operation, that can be called to actually do the printing: The method »eval«: public de.dclj.ram.notation.programming.EvaluableDesignation eval( de.dclj.ram.notation.programming.World world ) { Value value =( Value )Evaluator.completeEvaluation ( argumentDesignation, world ); if( value instanceof NumericValue ) { java.lang.System.out.println ( value.getPrintRepresentation() + " " ); } else { java.lang.System.out.println ( value.getPrintRepresentation() ); } return new Ok(); } The result is a new designation indicating the success of the operation. In my implementation everything is considered to be built of operators and operands. Each operator is defined by two classes. The intention is to make extensions to the language easy by requiring them just to add two new classes per operator. For example, »if ... then ...« also would be considered to be an operator. For example, the operator »-« is being specified as follows. The class pair for the operator »-«: /** A minus-operator token represents the operator "-". */ class MinusOperatorToken extends DefaultInfixOperatorToken implements PrefixOrInfixOperatorToken { public java.lang.String fixtext(){ return "-"; } public int infixLeftPriority(){ return 640; } public int infixRightPriority(){ return 639; } public int prefixLeftPriority(){ return 680; } public int prefixRightPriority(){ return 999 /* 681 */; } public boolean isPrefix(){ return this.isPrefix; } public void setPrefix(){ this.isPrefix = true; } public void reduce ( final PlacementHint placementHint, final OperatorStack operatorStack, final DesignationStack designationStack ) { de.dclj.ram.notation.programming.EvaluableDesignation rightDesignation = designationStack.pop(); if( placementHint instanceof OperatorMustBePrefixPlacementHint || this.isPrefix() ) { designationStack.accept ( new DifferenceDesignation ( new NumericValue( "0" ), rightDesignation )); } else { de.dclj.ram.notation.programming.EvaluableDesignation leftDesignation = designationStack.pop(); designationStack.accept ( new DifferenceDesignation ( leftDesignation, rightDesignation )); }} public MinusOperatorToken newInstance(){ return new MinusOperatorToken(); } private boolean isPrefix = false; public java.lang.String toString() { return this.getClass().getName(); }} /** A difference designation designates the difference of two designations. */ class DifferenceDesignation implements de.dclj.ram.notation.programming.EvaluableDesignation { final de.dclj.ram.notation.programming.EvaluableDesignation leftDesignation; final de.dclj.ram.notation.programming.EvaluableDesignation rightDesignation; public DifferenceDesignation ( final de.dclj.ram.notation.programming.EvaluableDesignation leftDesignation, final de.dclj.ram.notation.programming.EvaluableDesignation rightDesignation ) { this.leftDesignation = leftDesignation; this.rightDesignation = rightDesignation; } public de.dclj.ram.notation.programming.EvaluableDesignation eval( de.dclj.ram.notation.programming.World world ) { final java.lang.Object l = Evaluator.completeEvaluation( leftDesignation, world ); final java.lang.Object r = Evaluator.completeEvaluation( rightDesignation, world ); final java.math.BigDecimal left = new java.math.BigDecimal((( NumericValue )l ).asDecimal(), (( BasicWorld )world ).config().getMathContext() ); final java.math.BigDecimal right = new java.math.BigDecimal((( NumericValue )r ).asDecimal(), (( BasicWorld )world ).config().getMathContext() ); final java.math.BigDecimal difference = left.subtract ( right, (( BasicWorld )world ).config().getMathContext() ); return new NumericValue( difference.toString() ); } public de.dclj.ram.notation.programming.EvaluableDesignation continue_( final de.dclj.ram.notation.programming.World world ) { return null; } public java.lang.String toString() { return "DIFFERENCE-DESIGNATION( \"" + leftDesignation + "\",\"" + rightDesignation + "\")" ; }} [/QUOTE]

Verification