Mercurial Mercurial > hg > lexing / diff
summary | shortlog | changelog | graph | tags | bookmarks | branches | files | changeset | file | latest | revisions | annotate | diff | comparison | raw | help
Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
exps/both.scala
changeset 305 6e2cef17a9b3
parent 300 b7987014fed8
child 306 6756b026c5fe 
--- a/exps/both.scala	Mon Feb 04 13:10:26 2019 +0000
+++ b/exps/both.scala	Thu Feb 07 10:52:41 2019 +0000
@@ -31,7 +31,7 @@
 abstract class Rexp 
 case object ZERO extends Rexp
 case object ONE extends Rexp
-case class PRED(f: Char => Boolean) extends Rexp
+case class PRED(f: Char => Boolean, s: String = "_") extends Rexp 
 case class ALTS(rs: List[Rexp]) extends Rexp 
 case class SEQ(r1: Rexp, r2: Rexp) extends Rexp 
 case class STAR(r: Rexp) extends Rexp 
@@ -39,7 +39,7 @@
 
 
 // abbreviations
-def CHAR(c: Char) = PRED(_ == c)
+def CHAR(c: Char) = PRED(_ == c, c.toString)
 def ALT(r1: Rexp, r2: Rexp) = ALTS(List(r1, r2))
 def PLUS(r: Rexp) = SEQ(r, STAR(r))
 
@@ -47,7 +47,7 @@
 abstract class ARexp 
 case object AZERO extends ARexp
 case class AONE(bs: Bits) extends ARexp
-case class APRED(bs: Bits, f: Char => Boolean) extends ARexp
+case class APRED(bs: Bits, f: Char => Boolean, s: String = "_") extends ARexp
 case class AALTS(bs: Bits, rs: List[ARexp]) extends ARexp 
 case class ASEQ(bs: Bits, r1: ARexp, r2: ARexp) extends ARexp 
 case class ASTAR(bs: Bits, r: ARexp) extends ARexp 
@@ -92,15 +92,27 @@
 
 
 // string of a regular expressions - for testing purposes
-  def string(r: Rexp): String = r match {
-    case ZERO => "0"
-    case ONE => "1"
-    case PRED(_) => "_"
-    case ALTS(rs) => rs.map(string).mkString("[", "|", "]")
-    case SEQ(r1, r2) => s"(${string(r1)} ~ ${string(r2)})"
-    case STAR(r) => s"{${string(r)}}*"
-    case RECD(x, r) => s"(${x}! ${string(r)})"
-  }
+def string(r: Rexp): String = r match {
+  case ZERO => "0"
+  case ONE => "1"
+  case PRED(_, s) => s
+  case ALTS(rs) => rs.map(string).mkString("[", "|", "]")
+  case SEQ(r1, r2) => s"(${string(r1)} ~ ${string(r2)})"
+  case STAR(r) => s"{${string(r)}}*"
+  case RECD(x, r) => s"(${x}! ${string(r)})"
+}
+
+// string of an annotated regular expressions - for testing purposes
+
+def astring(a: ARexp): String = a match {
+  case AZERO => "0"
+  case AONE(_) => "1"
+  case APRED(_, _, s) => s
+  case AALTS(_, rs) => rs.map(astring).mkString("[", "|", "]")
+  case ASEQ(_, r1, r2) => s"(${astring(r1)} ~ ${astring(r2)})"
+  case ASTAR(_, r) => s"{${astring(r)}}*"
+}
+ 
 
 //--------------------------------------------------------------------------------------------------------
 // START OF NON-BITCODE PART
@@ -111,7 +123,7 @@
 def nullable (r: Rexp) : Boolean = r match {
   case ZERO => false
   case ONE => true
-  case PRED(_) => false
+  case PRED(_, _) => false
   case ALTS(rs) => rs.exists(nullable)
   case SEQ(r1, r2) => nullable(r1) && nullable(r2)
   case STAR(_) => true
@@ -122,7 +134,7 @@
 def der (c: Char, r: Rexp) : Rexp = r match {
   case ZERO => ZERO
   case ONE => ZERO
-  case PRED(f) => if (f(c)) ONE else ZERO
+  case PRED(f, _) => if (f(c)) ONE else ZERO
   case ALTS(List(r1, r2)) => ALTS(List(der(c, r1), der(c, r2)))
   case SEQ(r1, r2) => 
     if (nullable(r1)) ALTS(List(SEQ(der(c, r1), r2), der(c, r2)))
@@ -171,7 +183,7 @@
   case (SEQ(r1, r2), Right(v2)) => Sequ(mkeps(r1), inj(r2, c, v2))
   case (ALTS(List(r1, r2)), Left(v1)) => Left(inj(r1, c, v1))
   case (ALTS(List(r1, r2)), Right(v2)) => Right(inj(r2, c, v2))
-  case (PRED(_), Empty) => Chr(c) 
+  case (PRED(_, _), Empty) => Chr(c) 
   case (RECD(x, r1), _) => Rec(x, inj(r1, c, v))
 }
 
@@ -264,7 +276,7 @@
 def fuse(bs: Bits, r: ARexp) : ARexp = r match {
   case AZERO => AZERO
   case AONE(cs) => AONE(bs ++ cs)
-  case APRED(cs, f) => APRED(bs ++ cs, f)
+  case APRED(cs, f, s) => APRED(bs ++ cs, f, s)
   case AALTS(cs, rs) => AALTS(bs ++ cs, rs)
   case ASEQ(cs, r1, r2) => ASEQ(bs ++ cs, r1, r2)
   case ASTAR(cs, r) => ASTAR(bs ++ cs, r)
@@ -274,7 +286,7 @@
 def internalise(r: Rexp) : ARexp = r match {
   case ZERO => AZERO
   case ONE => AONE(Nil)
-  case PRED(f) => APRED(Nil, f)
+  case PRED(f, s) => APRED(Nil, f, s)
   case ALTS(List(r1, r2)) => 
     AALTS(Nil, List(fuse(List(Z), internalise(r1)), fuse(List(S), internalise(r2))))
   case ALTS(r1::rs) => {
@@ -291,7 +303,7 @@
 // decoding of values from bit sequences
 def decode_aux(r: Rexp, bs: Bits) : (Val, Bits) = (r, bs) match {
   case (ONE, bs) => (Empty, bs)
-  case (PRED(f), C(c)::bs) => (Chr(c), bs)
+  case (PRED(f, _), C(c)::bs) => (Chr(c), bs)
   case (ALTS(r::Nil), bs) => decode_aux(r, bs)
   case (ALTS(rs), bs) => bs match {
     case Z::bs1 => {
@@ -330,7 +342,7 @@
 def erase(r: ARexp) : Rexp = r match{
   case AZERO => ZERO
   case AONE(_) => ONE
-  case APRED(bs, f) => PRED(f)
+  case APRED(bs, f, s) => PRED(f, s)
   case AALTS(bs, rs) => ALTS(rs.map(erase(_)))
   case ASEQ(bs, r1, r2) => SEQ (erase(r1), erase(r2))
   case ASTAR(cs, r)=> STAR(erase(r))
@@ -342,7 +354,7 @@
 def bnullable (r: ARexp) : Boolean = r match {
   case AZERO => false
   case AONE(_) => true
-  case APRED(_,_) => false
+  case APRED(_,_,_) => false
   case AALTS(_, rs) => rs.exists(bnullable)
   case ASEQ(_, r1, r2) => bnullable(r1) && bnullable(r2)
   case ASTAR(_, _) => true
@@ -362,7 +374,7 @@
 def bder(c: Char, r: ARexp) : ARexp = r match {
   case AZERO => AZERO
   case AONE(_) => AZERO
-  case APRED(bs, f) => if (f(c)) AONE(bs:::List(C(c))) else AZERO
+  case APRED(bs, f, _) => if (f(c)) AONE(bs:::List(C(c))) else AZERO
   case AALTS(bs, rs) => AALTS(bs, rs.map(bder(c, _)))
   case ASEQ(bs, r1, r2) => 
     if (bnullable(r1)) AALT(bs, ASEQ(Nil, bder(c, r1), r2), fuse(bmkeps(r1), bder(c, r2)))
@@ -378,7 +390,6 @@
   case c::s => bders(s, bder(c, r))
 }
 
-
 def flats(rs: List[ARexp]): List[ARexp] = rs match {
     case Nil => Nil
     case AZERO :: rs1 => flats(rs1)
@@ -404,6 +415,7 @@
 
 
 
+
 def bders_simp (s: List[Char], r: ARexp) : ARexp = s match {
   case Nil => r
   case c::s => bders_simp(s, bsimp(bder(c, r)))
@@ -485,7 +497,7 @@
 def size(r: Rexp) : Int = r match {
   case ZERO => 1
   case ONE => 1
-  case PRED(_) => 1
+  case PRED(_,_) => 1
   case SEQ(r1, r2) => 1 + size(r1) + size(r2)
   case ALTS(rs) => 1 + rs.map(size).sum
   case STAR(r) => 1 + size(r)
@@ -498,9 +510,9 @@
 // Lexing Rules for a Small While Language
 
 //symbols
-val SYM = PRED("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ".contains(_))
+val SYM = PRED("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ".contains(_), "SYM")
 //digits
-val DIGIT = PRED("0123456789".contains(_))
+val DIGIT = PRED("0123456789".contains(_), "NUM")
 //identifiers
 val ID = SYM ~ (SYM | DIGIT).% 
 //numbers
@@ -578,13 +590,14 @@
   println(" Bit full simp: " + time(btokenise_simp_full(WHILE_REGS, fib_prog * i)))
 }
 
+
 println("Original " + size(WHILE_REGS))
-println("Size Bit  " + asize(bders_simp((fib_prog * 10).toList, internalise(WHILE_REGS))))
-println("Size Bitf " + asize(bders_simp_full((fib_prog * 10).toList, internalise(WHILE_REGS))))
-println("Size Old  " + size(ders_simp((fib_prog * 10).toList, WHILE_REGS)))
+println("Size Bit  " + asize(bders_simp((fib_prog * 1).toList, internalise(WHILE_REGS))))
+println("Size Bitf " + asize(bders_simp_full((fib_prog * 1).toList, internalise(WHILE_REGS))))
+println("Size Old  " + size(ders_simp((fib_prog * 1).toList, WHILE_REGS)))
 
 
-//System.exit(0)
+System.exit(0)
 
 println("Internal sizes test OK or strange")
 
@@ -662,3 +675,4 @@
 
 
 
+
lexing