javacc-7.0.4.grammars.OberonParser.jj Maven / Gradle / Ivy

Go to download
/*
 *               JavaCC source of a
 *               parser for the Oberon-2 Programming Language
 *
 *       Author: James Power,
 *  Affiliation: Dept. of Computer Science, National University of Ireland, 
 *               Maynooth, Co. Kildare, Ireland.
 * Last Revised: 17 November 1998
 *
 *     Based on: "The Programming Language Oberon"
 *               H. Mossenbock, N. Wirth, ETH Zurich, 1992-1996
 *
 * For more information on Oberon see http://www.oberon.ethz.ch/language.html
 */



PARSER_BEGIN(OberonParser)
public final class OberonParser { // Standard parser class setup...

  public static void main(String args[]) { 
    OberonParser parser;
    java.io.InputStream input;

    if (args.length==1) {
      System.out.print(args[args.length-1] + ": ");
      try {
        input = new java.io.FileInputStream(args[args.length-1]);
      } catch (java.io.FileNotFoundException e) {
        System.out.println("File not found.");
        return;
      }
    } else if (args.length==0) {
      System.out.println("Reading from standard input...");
      input = System.in;
    } else {
      System.out.println("Usage: java OberonParser [inputfile]");
      return;
    }
    try {
      parser = new OberonParser(input);
      parser.translation_unit();
      System.out.println("Successful parse!");
    } catch (ParseException e) {
      String msg = e.getMessage();
      msg = msg.substring(0,msg.indexOf("\n"));
      System.out.println("Syntax Error: "+msg);
    }
  }
}
PARSER_END(OberonParser)

/***************************************/
/********** TOKEN DEFINITIONS **********/
/***************************************/

TOKEN_MGR_DECLS :
{
  static int commentNesting;  // Current level of nesting of comments
}


/*** Skip whitespace and comments ***/
SKIP :
{
  " "
| "\t"
| "\n"
| "\r"
| "(*"  { commentNesting=1; } : IN_COMMENT
}
 MORE:
{
  < ~[] >
}
 SKIP:
{
  < "(*" > { commentNesting++; } 
| < "*)" > { commentNesting--; if (commentNesting==0) SwitchTo(DEFAULT); }
}


/*** The keywords ***/
TOKEN :
{
  
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
| 
}

TOKEN  :
{
  < DOTDOT : ".." >
}
/* The hack: if we see digits followed by "..", trap it here, otherwise
 * the digits and first dot will be gobbled up to make a REAL below,
 * and the second dot will give a syntax error
 */
TOKEN :
{
< DIGIT_DOT_HACK : ()+ ".."  > 
  {
    // Create the new token and initialise its fields properly
    Token newTok = new Token();
    newTok.kind = DOTDOT;
    newTok.beginLine = newTok.endLine = matchedToken.endLine;
    newTok.beginColumn = matchedToken.endColumn-1;
    newTok.endColumn = matchedToken.endColumn;
    newTok.image="..";
    newTok.next =  matchedToken.next;
    newTok.specialToken = null;
    // Now fix up matchedToken:
    matchedToken.kind = INTEGER;
    matchedToken.endColumn -= 2;
    matchedToken.image = matchedToken.image.substring(0,matchedToken.image.length()-2);
    matchedToken.next = newTok;
  }
}


/*** Basic constants: numbers, idents, chars and strings ***/
TOKEN  :
{
  < #DIGIT : ["0"-"9"] >
| < #SCALEFACTOR : ("E" | "D") ("+" | "-")? ()+ >
| < #HEXDIGIT : ["0"-"9","A"-"F"] >
| < INTEGER : ()+ |  ()* "H" >
| < REAL : ()+ "." ()* ()? >

| < #LETTER : ["A"-"Z","a"-"z"] >
| < IDENT :  ( | )* >

| < CHARACTER :  ()* "X" >

| < STRING : "\"" (~["\""])* "\"" | "'" (~["'"])* "'" >
}


/**************************************/
/********** START OF GRAMMAR **********/
/**************************************/

void translation_unit() : {}
{
  Module()
}

void Module() : {}
{
    ";" 
  (ImportList())? 
  DeclSeq() 
  ( StatementSeq())?
    "."
}
void ImportList() : {}
{
   ImportSpec() ("," ImportSpec())* ";"
}
void ImportSpec() : {}
{
   (":=" )?
}

void DeclSeq() : {}
{
  ( VarTypeDecl() )*
  ( ProcedureDecl() )*
}
void VarTypeDecl() : {}
{
   (ConstDecl() ";")* 
|  (TypeDecl() ";")* 
|  (VarDecl() ";")* 
}
void ConstDecl() : {}
{
  IdentDef() "=" ConstExpr()
}
void TypeDecl() : {}
{
  IdentDef() "=" Type()
}
void VarDecl() : {}
{
  IdentList() ":" Type()
}
void ProcedureDecl() : {}
{
   (ForwardDecl() | ProcDecl()) ";"
}
void ProcDecl() : {}
{
  (Receiver())? IdentDef() (FormalPars())? ";" 
  DeclSeq() ( StatementSeq())?  
}
void ForwardDecl() : {}
{
  "^" (Receiver())? IdentDef() (FormalPars())?
}
void FormalPars() : {}
{
  "(" (FPSection() (";" FPSection())*)? ")" (":" Qualident())?
}
void FPSection() : {}
{
  ()?  ("," )* ":" Type()
} 
void Receiver() : {}
{
  "(" ()?  ":"  ")"
}
void Type() : {}
{
  Qualident()
  |  (ConstExpr() ("," ConstExpr())*)?  Type() 
  |  ("("Qualident()")")? FieldList() (";" FieldList())* 
  |   Type()
  |  (FormalPars())?
}
void FieldList() : {}
{
  (IdentList() ":" Type())?
}
void StatementSeq() : {}
{
  optStatement() (";" optStatement())*
}
void optStatement() : {}
{
  (Statement())?
}
void Statement() : {}
{
    Designator() ((":=" Expr()) | Arguments() )
  |  Expr()  StatementSeq() 
    ( Expr()  StatementSeq())*
    ( StatementSeq())?
     
  |  Expr()  Case() 
    ("|" Case())*
    ( StatementSeq())?
     
  |  Expr()  StatementSeq()  
  |  StatementSeq()  Expr() 
  |   ":=" Expr()  Expr() 
    ( ConstExpr())? 
     StatementSeq()  
  |  StatementSeq() 
  |  Guard()  StatementSeq() 
    ("|" Guard()  StatementSeq())*
    ( StatementSeq())?
    
  |  
  |  (Expr())?
}
void Case() : {}
{
  (CaseLabels() ("," CaseLabels())* ":" StatementSeq())?
} 
void CaseLabels() : {}
{
  ConstExpr() ( ConstExpr())?
}
void Guard() : {}
{
  Qualident() ":" Qualident()
}
void ConstExpr() : {}
{
  Expr()
}
void Expr() : {}
{
  SimpleExpr() (Relation() SimpleExpr())?
}
void SimpleExpr() : {}
{
  ("+" | "-")? Term() (AddOp() Term())*
} 
void Term() : {}
{
  Factor() (MulOp() Factor())*
}
void Factor() : {}
{
  Designator() Arguments() | Number() |  |  
                |  | Set() | "(" Expr() ")" | "~" Factor()
}
void Set() : {}
{
  "{" (Element() ("," Element())* )? "}"
} 
void Element() : {}
{
  Expr() ( Expr())?
}
void Relation() : {}
{
  "=" | "#" | "<" | "<=" | ">" | ">=" |  | 
}
void AddOp() : {}
{
  "+" | "-" | 
}
void MulOp() : {}
{
  "*" | "/" |  |  | "&"
} 

  /* NB: Because Arguments() always appears after Designator(), there 
   * will be a conflict if we get "(" Qualident() ")" - do we consume
   * this as part of the Designator(), or leave it for the Arguments()
   * The disambiguation in Designator() chooses the former, so the
   * semantic phase will have to disentangle this for function calls
   * with one (qualified) identifier as an argument.
   */

void Arguments() : {}
{
  ( "(" (ExprList())? ")" )?
} 
void Designator() : {}
{
   (LOOKAHEAD("." | "[" | "^" | ("(" Qualident() ")")) 
   ( "."  
  | "[" ExprList() "]" 
  | "^" 
  | "(" Qualident() ")")
  )*
} 
void ExprList() : {}
{
  Expr() ("," Expr())*
}
void IdentList() : {}
{
  IdentDef() ("," IdentDef())*
}
void Qualident() : {}
{
  // ( ".")? 
   ("." )?
}
void IdentDef() : {}
{
   ("*" | "-")?
}
void Number() : {}
{
   | 
}