diff -r e85600529ca5 -r 4794759139ea progs/regexp4.scala --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/progs/regexp4.scala Sat Jun 15 09:23:18 2013 -0400 @@ -0,0 +1,168 @@ +// regular expressions +abstract class Rexp + +case object NULL extends Rexp +case object EMPTY extends Rexp +case class CHAR(c: Char) extends Rexp +case class ALT(r1: Rexp, r2: Rexp) extends Rexp +case class SEQ(r1: Rexp, r2: Rexp) extends Rexp +case class STAR(r: Rexp) extends Rexp +case class NOT(r: Rexp) extends Rexp + + +// some convenience for typing in regular expressions +def charlist2rexp(s : List[Char]) : Rexp = s match { + case Nil => EMPTY + case c::Nil => CHAR(c) + case c::s => SEQ(CHAR(c), charlist2rexp(s)) +} +implicit def string2rexp(s : String) : Rexp = charlist2rexp(s.toList) + + +// nullable function: tests whether the regular +// expression can recognise the empty string +def nullable (r: Rexp) : Boolean = r match { + case NULL => false + case EMPTY => true + case CHAR(_) => false + case ALT(r1, r2) => nullable(r1) || nullable(r2) + case SEQ(r1, r2) => nullable(r1) && nullable(r2) + case STAR(_) => true + case NOT(r) => !(nullable(r)) +} + +// tests whether a regular expression +// recognises nothing +def zeroable (r: Rexp) : Boolean = r match { + case NULL => true + case EMPTY => false + case CHAR(_) => false + case ALT(r1, r2) => zeroable(r1) && zeroable(r2) + case SEQ(r1, r2) => zeroable(r1) || zeroable(r2) + case STAR(_) => false + case NOT(r) => !(zeroable(r)) +} + +def starts_with (r: Rexp, c: Char) : Boolean = r match { + case NULL => false + case EMPTY => false + case CHAR(d) => (c == d) + case ALT(r1, r2) => starts_with(r1, c) || starts_with(r2, c) + case SEQ(r1, r2) => if (nullable(r1)) (starts_with(r1, c) || starts_with(r2, c)) + else starts_with(r1, c) + case STAR(r) => starts_with(r, c) + case NOT(r) => !(starts_with(r, c)) +} + +// derivative of a regular expression w.r.t. a character +def der (c: Char, r: Rexp) : Rexp = r match { + case NULL => NULL + case EMPTY => NULL + case CHAR(d) => if (c == d) EMPTY else NULL + case ALT(r1, r2) => ALT(der(c, r1), der(c, r2)) + case SEQ(r1, r2) => + if (nullable(r1)) ALT(SEQ(der(c, r1), r2), der(c, r2)) + else SEQ(der(c, r1), r2) + case STAR(r) => SEQ(der(c, r), STAR(r)) + case NOT(r) => NOT(der (c, r)) +} + +// derivative w.r.t. a string (iterates der) +def ders (s: List[Char], r: Rexp) : Rexp = s match { + case Nil => r + case c::s => ders(s, der(c, r)) +} + +// main matcher function +def matcher(r: Rexp, s: String) : Boolean = nullable(ders(s.toList, r)) + + +// regular expression for specifying +// ranges of characters +def RANGE(s : List[Char]) : Rexp = s match { + case Nil => NULL + case c::Nil => CHAR(c) + case c::s => ALT(CHAR(c), RANGE(s)) +} + +//one or more +def PLUS(r: Rexp) = SEQ(r, STAR(r)) + + +//some regular expressions +val LOWERCASE = RANGE("abcdefghijklmnopqrstuvwxyz".toList) +val UPPERCASE = RANGE("ABCDEFGHIJKLMNOPQRSTUVWXYZ".toList) +val LETTER = ALT(LOWERCASE, UPPERCASE) +val DIGITS = RANGE("0123456789".toList) +val NONZERODIGITS = RANGE("123456789".toList) + +val IDENT = SEQ(LETTER, STAR(ALT(LETTER,DIGITS))) +val NUMBER = ALT(SEQ(NONZERODIGITS, STAR(DIGITS)), "0") +val WHITESPACE = RANGE(" \n".toList) +val SYMBOLS = RANGE("/*".toList) + +val ALL = ALT(ALT(ALT(LETTER, DIGITS), WHITESPACE), SYMBOLS) + +val COMMENT = SEQ(SEQ("/*", NOT(SEQ(SEQ(STAR(ALL), "*/"), STAR(ALL)))), "*/") + +println(matcher(NUMBER, "0")) +println(matcher(NUMBER, "01")) +println(matcher(NUMBER, "123450")) + +println(matcher(SEQ(STAR("a"), STAR("b")), "bbaaa")) +println(matcher(ALT(STAR("a"), STAR("b")), "")) +println(matcher("abc", "")) +println(matcher(STAR(ALT(EMPTY, "a")), "")) +println(matcher(STAR(EMPTY), "a")) +println(matcher("cab","cab")) +println(matcher(STAR("a"),"aaa")) +println(matcher("cab" ,"cab")) +println(matcher(STAR("a"),"aaa")) + +println(matcher(COMMENT, "/* */")) +println(matcher(COMMENT, "/* foobar comment */")) +println(matcher(COMMENT, "/* test */ test */")) + +// an example list of regular expressions +val regs: List[Rexp]= List("if", "then", "else", "+", IDENT, NUMBER, COMMENT, WHITESPACE) + + +def error (s: String) = throw new IllegalArgumentException ("Could not lex " + s) + +def munch(r: Rexp, s: List[Char], t: List[Char]) : Option[(List[Char], List[Char])] = + if (zeroable(r)) None else s match { + case Nil => if (nullable(r)) Some(Nil, t) else None + case c::s if (zeroable(der (c, r)) && nullable(r)) => Some(c::s, t) + //case c::s if (zeroable(der (c, r))) => None + case c::s => munch(der (c, r), s, t ::: List(c)) +} + + +def lex_one (regs: List[Rexp], s: List[Char]) : (List[Char], List[Char]) = { + val somes = regs.map { munch(_, s, Nil) } .flatten + if (somes == Nil) error(s.mkString) else (somes sortBy (_._1.length) head) +} + +def lex_all (regs: List[Rexp], s: List[Char]) : List[String] = s match { + case Nil => Nil + case _ => lex_one(regs, s) match { + case (rest, s) => s.mkString :: lex_all(regs, rest) + } +} + + + +starts_with(der('/', COMMENT), '*') + +munch(COMMENT, "/*ifff if 34 */".toList, Nil) +val COMMENT2 = NOT(SEQ(SEQ(STAR(ALL), "*/"), STAR(ALL))) + +der('a', COMMENT2) +zeroable(der('a', COMMENT2)) + +matcher(COMMENT2, "ifff if 34") +munch(COMMENT2, "ifff if 34".toList, Nil) +starts_with(COMMENT2, 'i') +lex_all(regs, "ifff if 34".toList) +lex_all(regs, "ifff $ if 34".toList) +