solutions/cw4a/parser.sc
author Christian Urban <christian.urban@kcl.ac.uk>
Fri, 29 Nov 2024 18:59:32 +0000
changeset 976 e9eac62928f5
parent 910 b655ce68983f
permissions -rw-r--r--
updated

// CW3

import $file.lexer
import lexer._ 


case class ~[+A, +B](_1: A, _2: B)
type IsSeq[A] = A => Seq[_]

abstract class Parser[I : IsSeq, T] {
  def parse(ts: I): Set[(T, I)]

  def parse_all(ts: I) : Set[T] =
    for ((head, tail) <- parse(ts); if tail.isEmpty) yield head
}

class SeqParser[I : IsSeq, T, S](p: => Parser[I, T], q: => Parser[I, S]) extends Parser[I, ~[T, S]] {
  def parse(sb: I) = 
    for ((head1, tail1) <- p.parse(sb); 
         (head2, tail2) <- q.parse(tail1)) yield (new ~(head1, head2), tail2)
}

class AltParser[I : IsSeq, T](p: => Parser[I, T], q: => Parser[I, T]) extends Parser[I, T] {
  def parse(sb: I) = p.parse(sb) ++ q.parse(sb)   
}

class FunParser[I : IsSeq, T, S](p: => Parser[I, T], f: T => S) extends Parser[I, S] {
  def parse(sb: I) = 
    for ((head, tail) <- p.parse(sb)) yield (f(head), tail)
}

// New parser that takes as input a list of tokens
case class TokenListParser(ts: List[Token]) extends Parser[List[Token], List[Token]] {
    def parse(tsb: List[Token]) = {
        val (prefix, suffix) = tsb.splitAt(ts.length)
        if (prefix == ts) Set((prefix, suffix)) else Set()
    }
}

// Implicit definitions to go from a token 
// or a list of tokens to a TokenListParser
implicit def token2parser(t: Token) = TokenListParser(List(t))
implicit def tokenList2parser(ts: List[Token]) = TokenListParser(ts)

implicit def ParserOps[I : IsSeq, T](p: Parser[I, T]) = new {
  def || (q : => Parser[I, T]) = new AltParser[I, T](p, q)
  def ==>[S] (f: => T => S) = new FunParser[I, T, S](p, f)
  def ~[S] (q : => Parser[I, S]) = new SeqParser[I, T, S](p, q)
}

implicit def TokenOps(t: Token) = new {
    def || (q : => Parser[List[Token], List[Token]]) = new AltParser[List[Token], List[Token]](List(t), q)
    def || (qs : List[Token]) = new AltParser[List[Token], List[Token]](List(t), qs)
    def ==>[S] (f: => List[Token] => S) = new FunParser[List[Token], List[Token], S](List(t), f)
    def ~[S](q : => Parser[List[Token], S]) =
        new SeqParser[List[Token], List[Token], S](List(t), q)
    def ~ (qs : List[Token]) =
        new SeqParser[List[Token], List[Token], List[Token]](List(t), qs)
}

implicit def TokenListOps(ts: List[Token]) = new {
    def || (q : => Parser[List[Token], List[Token]]) = new AltParser[List[Token], List[Token]](ts, q)
    def || (qs : List[Token]) = new AltParser[List[Token], List[Token]](ts, qs)
    def ==>[S] (f: => List[Token] => S) = new FunParser[List[Token], List[Token], S](ts, f)
    def ~[S](q : => Parser[List[Token], S]) =
        new SeqParser[List[Token], List[Token], S](ts, q)
    def ~ (qs : List[Token]) =
        new SeqParser[List[Token], List[Token], List[Token]](ts, qs)
}

// Abstract Syntax Trees
abstract class Stmt
abstract class AExp
abstract class BExp

type Block = List[Stmt]

case object Skip extends Stmt
case object Break extends Stmt
case class If(a: BExp, bl1: Block, bl2: Block) extends Stmt
case class While(b: BExp, bl: Block) extends Stmt
case class Assign(s: String, a: AExp) extends Stmt
case class Read(s: String) extends Stmt
case class WriteId(s: String) extends Stmt  // for printing values of variables
case class WriteString(s: String) extends Stmt  // for printing words
case class For(counter: String, lower: AExp, upper: AExp, code: Block) extends Stmt


case class Var(s: String) extends AExp
case class Num(i: Int) extends AExp
case class Aop(o: String, a1: AExp, a2: AExp) extends AExp

case object True extends BExp
case object False extends BExp
case class Bop(o: String, a1: AExp, a2: AExp) extends BExp
case class And(b1: BExp, b2: BExp) extends BExp
case class Or(b1: BExp, b2: BExp) extends BExp

case class IdParser() extends Parser[List[Token], String] {
    def parse(tsb: List[Token]) = tsb match {
        case T_ID(id) :: rest => Set((id, rest))
        case _ => Set()
    }
}

case class NumParser() extends Parser[List[Token], Int] {
    def parse(tsb: List[Token]) = tsb match {
        case T_NUM(n) :: rest => Set((n, rest))
        case _ => Set()
    }
}

case class StringParser() extends Parser[List[Token], String] {
    def parse(tsb: List[Token]) = tsb match {
        case T_STRING(s) :: rest => Set((s, rest))
        case _ => Set()
    }
}

// WHILE Language Parsing
lazy val AExp: Parser[List[Token], AExp] = 
  (Te ~ T_OP("+") ~ AExp) ==> { case x ~ _ ~ z => Aop("+", x, z): AExp } ||
  (Te ~ T_OP("-") ~ AExp) ==> { case x ~ _ ~ z => Aop("-", x, z): AExp } || Te
lazy val Te: Parser[List[Token], AExp] = 
  (Fa ~ T_OP("*") ~ Te) ==> { case x ~ _ ~ z => Aop("*", x, z): AExp } || 
  (Fa ~ T_OP("/") ~ Te) ==> { case x ~ _ ~ z => Aop("/", x, z): AExp } || 
  (Fa ~ T_OP("%") ~ Te) ==> { case x ~ _ ~ z => Aop("%", x, z): AExp } || Fa  
lazy val Fa: Parser[List[Token], AExp] = 
   (T_PAREN("(") ~ AExp ~ T_PAREN(")")) ==> { case _ ~ y ~ _ => y } || 
   IdParser() ==> Var  || 
   NumParser() ==> Num

lazy val BExp: Parser[List[Token], BExp] = 
   (AExp ~ T_OP("==") ~ AExp) ==> { case x ~ _ ~ z => Bop("==", x, z): BExp } || 
   (AExp ~ T_OP("!=") ~ AExp) ==> { case x ~ _ ~ z => Bop("!=", x, z): BExp } || 
   (AExp ~ T_OP("<") ~ AExp) ==> { case x ~ _ ~ z => Bop("<", x, z): BExp } || 
   (AExp ~ T_OP(">") ~ AExp) ==> { case x ~ _ ~ z => Bop(">", x, z): BExp } ||
   (T_PAREN("(") ~ BExp ~ List(T_PAREN(")"), T_OP("&&")) ~ BExp) ==> { case _ ~ y ~ _ ~ v => And(y, v): BExp } ||
   (T_PAREN("(") ~ BExp ~ List(T_PAREN(")"), T_OP("||")) ~ BExp) ==> { case _ ~ y ~ _ ~ v => Or(y, v): BExp } ||
   (T_KEYWORD("true") ==> (_ => True: BExp )) || 
   (T_KEYWORD("false") ==> (_ => False: BExp )) ||
   (T_PAREN("(") ~ BExp ~ T_PAREN(")")) ==> { case _ ~ x ~ _ => x }

lazy val Stmt: Parser[List[Token], Stmt] =
    T_KEYWORD("skip") ==> (_ => Skip: Stmt) ||
    T_KEYWORD("break") ==> (_ => Break: Stmt) ||
    (IdParser() ~ T_OP(":=") ~ AExp) ==> { case id ~ _ ~ z => Assign(id, z): Stmt } ||
    (T_KEYWORD("if") ~ BExp ~ T_KEYWORD("then") ~ Block ~ T_KEYWORD("else") ~ Block) ==> { case _ ~ y ~ _ ~ u ~ _ ~ w => If(y, u, w): Stmt } ||
    (T_KEYWORD("while") ~ BExp ~ T_KEYWORD("do") ~ Block) ==> { case _ ~ y ~ _ ~ w => While(y, w) : Stmt } ||
    (T_KEYWORD("read") ~ IdParser()) ==> { case _ ~ id => Read(id): Stmt} ||
    (T_KEYWORD("write") ~ IdParser()) ==> { case _ ~ id => WriteId(id): Stmt} ||
    (T_KEYWORD("write") ~ StringParser()) ==> { case _ ~ s => WriteString(s): Stmt} ||
    (T_KEYWORD("for") ~ IdParser() ~ T_OP(":=") ~ AExp ~ T_KEYWORD("upto") ~ AExp ~ T_KEYWORD("do") ~ Block) ==> {
      case _ ~ id ~ _ ~ lower ~ _ ~ upper ~ _ ~ blck => For(id, lower, upper, blck): Stmt
    }

lazy val Stmts: Parser[List[Token], Block] =
    (Stmt ~ T_SEMI ~ Stmts) ==> { case x ~ _ ~ z => x :: z : Block } ||
    (Stmt ==> (s => List(s) : Block))

lazy val Block: Parser[List[Token], Block] =
    (T_PAREN("{") ~ Stmts ~ T_PAREN("}")) ==> { case x ~ y ~ z => y} ||
    (Stmt ==> (s => List(s)))