diff -r c989f50da5de -r e461b5325b5e testing4/re.scala --- a/testing4/re.scala Thu Dec 06 18:56:26 2018 +0000 +++ b/testing4/re.scala Thu Dec 06 21:49:43 2018 +0000 @@ -1,23 +1,21 @@ // Part 1 about Regular Expression Matching //========================================== -//object CW9a { - // Regular Expressions abstract class Rexp case object ZERO extends Rexp case object ONE 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 ALT(r1: Rexp, r2: Rexp) extends Rexp // alternative +case class SEQ(r1: Rexp, r2: Rexp) extends Rexp // sequence +case class STAR(r: Rexp) extends Rexp // star -// some convenience for typing in regular expressions + +// some convenience for typing regular expressions import scala.language.implicitConversions import scala.language.reflectiveCalls - def charlist2rexp(s: List[Char]): Rexp = s match { case Nil => ONE case c::Nil => CHAR(c) @@ -46,30 +44,38 @@ // accordingly. def nullable (r: Rexp) : Boolean = r match { - case ZERO => false - case ONE => true - case CHAR(_) => false - case ALT(r1, r2) => nullable(r1) || nullable(r2) - case SEQ(r1, r2) => nullable(r1) && nullable(r2) - case STAR(_) => true + case ZERO => false + case ONE => true + case CHAR(_) => false + case ALT(r1, r2) => nullable(r1) | nullable(r2) + case SEQ(r1, r2) => nullable(r1) & nullable(r2) + case STAR(_) => true } + + // (2) Complete the function der according to // the definition given in the coursework; this // function calculates the derivative of a // regular expression w.r.t. a character. +//TODO: debug +//TODO: understand this more. +// first test runs +// test 2 fails +// test 3 runs +// test 4 runs def der (c: Char, r: Rexp) : Rexp = r match { - case ZERO => ZERO - case ONE => ZERO - case CHAR(d) => if (c == d) ONE else ZERO - 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(r1) => SEQ(der(c, r1), STAR(r1)) + //TODO: debug + case ZERO => ZERO + case ONE => ZERO + case CHAR(r1) => if (c == r1) ONE else ZERO + 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(r1) => SEQ(der(c, r1), STAR(r1)) } + // (3) Complete the simp function according to // the specification given in the coursework; this // function simplifies a regular expression from @@ -78,19 +84,21 @@ // STAR-regular expressions. def simp(r: Rexp) : Rexp = r match { - case ALT(r1, r2) => (simp(r1), simp(r2)) match { - case (ZERO, r2s) => r2s - case (r1s, ZERO) => r1s - case (r1s, r2s) => if (r1s == r2s) r1s else ALT (r1s, r2s) - } - case SEQ(r1, r2) => (simp(r1), simp(r2)) match { - case (ZERO, _) => ZERO - case (_, ZERO) => ZERO - case (ONE, r2s) => r2s - case (r1s, ONE) => r1s - case (r1s, r2s) => SEQ(r1s, r2s) - } - case r => r + case STAR(_) => r + case SEQ(r1, r2) => (simp(r1), simp(r2)) match { // potential failure + case (_, ZERO) => ZERO + case (ZERO, _) => ZERO + case (r1, ONE) => simp(r1) + case (ONE, r2) => simp(r2) + case (r1, r2) => SEQ(r1, r2) + } + case ALT(r1, r2) => (simp(r1), simp(r2)) match { + case (r1, ZERO) => simp(r1) + case (ZERO, r1) => simp(r1) + case (r1, r2) if r1 == r2 => simp(r1) + case (r1, r2) => ALT(r1, r2) + } + case r => r } @@ -98,58 +106,60 @@ // calculates the derivative w.r.t. a string; the second // is the regular expression matcher taking a regular // expression and a string and checks whether the -// string matches the regular expression. +// string matches the regular expression def ders (s: List[Char], r: Rexp) : Rexp = s match { - case Nil => r - case c::s => ders(s, simp(der(c, r))) + case Nil => r + case c::cs => ders(cs, simp(der(c, r))) } -// main matcher function -def matcher(r: Rexp, s: String) = nullable(ders(s.toList, r)) +def matcher(r: Rexp, s: String): Boolean = { + nullable(ders(s.toList, r)) +} + // (5) Complete the size function for regular // expressions according to the specification // given in the coursework. - def size(r: Rexp): Int = r match { - case ZERO => 1 - case ONE => 1 - case CHAR(_) => 1 - case ALT(r1, r2) => 1 + size(r1) + size (r2) - case SEQ(r1, r2) => 1 + size(r1) + size (r2) - case STAR(r1) => 1 + size(r1) + case ZERO => 1 + case ONE => 1 + case CHAR(_) => 1 + case SEQ(r1, r2) => 1 + size(r1) + size(r2) + case ALT(r1, r2) => 1 + size(r1) + size(r2) + case STAR(r1) => 1 + size(r1) } +// some testing data -// some testing data -/* -matcher(("a" ~ "b") ~ "c", "abc") // => true -matcher(("a" ~ "b") ~ "c", "ab") // => false +//matcher(("a" ~ "b") ~ "c", "abc") // => true +//matcher(("a" ~ "b") ~ "c", "ab") // => false // the supposedly 'evil' regular expression (a*)* b val EVIL = SEQ(STAR(STAR(CHAR('a'))), CHAR('b')) -matcher(EVIL, "a" * 1000 ++ "b") // => true -matcher(EVIL, "a" * 1000) // => false +//matcher(EVIL, "a" * 1000 ++ "b") // => true +//matcher(EVIL, "a" * 1000) // => false // size without simplifications -size(der('a', der('a', EVIL))) // => 28 -size(der('a', der('a', der('a', EVIL)))) // => 58 +//size(der('a', der('a', EVIL))) // => 28 +//size(der('a', der('a', der('a', EVIL)))) // => 58 + + // size with simplification -size(simp(der('a', der('a', EVIL)))) // => 8 -size(simp(der('a', der('a', der('a', EVIL))))) // => 8 +//size(simp(der('a', der('a', EVIL)))) // => 8 +//size(simp(der('a', der('a', der('a', EVIL))))) // => 8 // Python needs around 30 seconds for matching 28 a's with EVIL. // Java 9 and later increase this to an "astonishing" 40000 a's in -// around 30 seconds. +// 30 seconds. // -// Lets see how long it takes to match strings with -// 5 Million a's...it should be in the range of a -// couple of seconds. +// Lets see how long it really takes to match strings with +// 5 Million a's...it should be in the range of a couple +// of seconds. def time_needed[T](i: Int, code: => T) = { val start = System.nanoTime() @@ -158,19 +168,17 @@ (end - start)/(i * 1.0e9) } -for (i <- 0 to 5000000 by 500000) { - println(i + " " + "%.5f".format(time_needed(2, matcher(EVIL, "a" * i)))) -} +//for (i <- 0 to 5000000 by 500000) { +// println(i + " " + "%.5f".format(time_needed(2, matcher(EVIL, "a" * i)))) +//} // another "power" test case -simp(Iterator.iterate(ONE:Rexp)(r => SEQ(r, ONE | ONE)).drop(100).next) == ONE +println(simp(Iterator.iterate(ONE:Rexp)(r => ALT(r, r)).drop(40).next)) // the Iterator produces the rexp // // SEQ(SEQ(SEQ(..., ONE | ONE) , ONE | ONE), ONE | ONE) // -// where SEQ is nested 100 times. - -*/ +// where SEQ is nested 50 times. -//} +