# HG changeset patch # User Christian Urban # Date 1385956668 0 # Node ID 141041fc76b53c793241de2da434a53a82257a66 # Parent 599e6a6ab6c6ff4f24a64e067199655cf4f15710 added diff -r 599e6a6ab6c6 -r 141041fc76b5 progs/fun.scala --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/progs/fun.scala Mon Dec 02 03:57:48 2013 +0000 @@ -0,0 +1,530 @@ +import scala.language.implicitConversions +import scala.language.reflectiveCalls +import scala.util._ +import scala.annotation.tailrec +import scala.sys.process._ + +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 RANGE(cs: List[Char]) extends Rexp +case class SEQ(r1: Rexp, r2: Rexp) extends Rexp +case class PLUS(r: Rexp) extends Rexp +case class STAR(r: Rexp) extends Rexp +case class NTIMES(r: Rexp, n: Int) extends Rexp +case class NUPTOM(r: Rexp, n: Int, m: Int) extends Rexp + +object RANGE { + def apply(s: String) : RANGE = RANGE(s.toList) +} +def NMTIMES(r: Rexp, n: Int, m: Int) = { + if(m < n) throw new IllegalArgumentException("the number m cannot be smaller than n.") + else NUPTOM(r, n, m - n) +} + +case class NOT(r: Rexp) extends Rexp +case class OPT(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) + +implicit def RexpOps (r: Rexp) = new { + def | (s: Rexp) = ALT(r, s) + def % = STAR(r) + def ~ (s: Rexp) = SEQ(r, s) +} + +implicit def stringOps (s: String) = new { + def | (r: Rexp) = ALT(s, r) + def | (r: String) = ALT(s, r) + def % = STAR(s) + def ~ (r: Rexp) = SEQ(s, r) + def ~ (r: String) = SEQ(s, r) +} + + +// 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 PLUS(r) => nullable(r) + case NTIMES(r, i) => if (i == 0) true else nullable(r) + case NUPTOM(r, i, j) => if (i == 0) true else nullable(r) + case RANGE(_) => false + case NOT(r) => !(nullable(r)) + case OPT(_) => true +} + +// 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 PLUS(r) => SEQ(der(c, r), STAR(r)) + case NTIMES(r, i) => + if (i == 0) NULL else der(c, SEQ(r, NTIMES(r, i - 1))) + case NUPTOM(r, i, j) => + if (i == 0 && j == 0) NULL else + if (i == 0) ALT(der(c, NTIMES(r, j)), der(c, NUPTOM(r, 0, j - 1))) + else der(c, SEQ(r, NUPTOM(r, i - 1, j))) + case RANGE(cs) => if (cs contains c) EMPTY else NULL + case NOT(r) => NOT(der (c, r)) + case OPT(r) => der(c, r) +} + +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 PLUS(r) => zeroable(r) + case NTIMES(r, i) => if (i == 0) false else zeroable(r) + case NUPTOM(r, i, j) => if (i == 0) false else zeroable(r) + case RANGE(_) => false + case NOT(r) => !(zeroable(r)) // bug: incorrect definition for NOT + case OPT(_) => false +} + +// 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)) +} + + +// regular expressions for the While language +val SYM = RANGE("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz_") +val DIGIT = RANGE("0123456789") +val ID = SYM ~ (SYM | DIGIT).% +val NUM = PLUS(DIGIT) +val KEYWORD : Rexp = "if" | "then" | "else" | "read" | "write" | "def" +val SEMI: Rexp = ";" +val COMMA: Rexp = "," +val OP: Rexp = ":=" | "==" | "-" | "+" | "*" | "!=" | "<" | ">" | "%" | "=" | "/" +val WHITESPACE = PLUS(" " | "\n" | "\t") +val RPAREN: Rexp = ")" +val LPAREN: Rexp = "(" +val BEGIN: Rexp = "{" +val END: Rexp = "}" +val ALL = SYM | DIGIT | OP | " " | ":" | ";" | "\"" | "=" | "," | "(" | ")" +val ALL2 = ALL | "\n" +val COMMENT2 = ("/*" ~ NOT(ALL.% ~ "*/" ~ ALL.%) ~ "*/") +val COMMENT = ("/*" ~ ALL2.% ~ "*/") | ("//" ~ ALL.% ~ "\n") +val STRING = "\"" ~ ALL.% ~ "\"" + + +// token for While languag +abstract class Token +case object T_WHITESPACE extends Token +case object T_SEMI extends Token +case object T_COMMA extends Token +case object T_LPAREN extends Token +case object T_RPAREN extends Token +case object T_BEGIN extends Token +case object T_END extends Token +case object T_COMMENT extends Token +case class T_ID(s: String) extends Token +case class T_OP(s: String) extends Token +case class T_NUM(s: String) extends Token +case class T_KWD(s: String) extends Token +case class T_STRING(s: String) extends Token +case class T_ERR(s: String) extends Token // special error token + + +type TokenFun = String => Token +type LexRules = List[(Rexp, TokenFun)] +val While_lexing_rules: LexRules = + List((KEYWORD, (s) => T_KWD(s)), + (ID, (s) => T_ID(s)), + (COMMENT, (s) => T_COMMENT), + (OP, (s) => T_OP(s)), + (NUM, (s) => T_NUM(s)), + (SEMI, (s) => T_SEMI), + (COMMA, (s) => T_COMMA), + (LPAREN, (s) => T_LPAREN), + (RPAREN, (s) => T_RPAREN), + (BEGIN, (s) => T_BEGIN), + (END, (s) => T_END), + (STRING, (s) => T_STRING(s.drop(1).dropRight(1))), + (WHITESPACE, (s) => T_WHITESPACE)) + + +// calculates derivatives until all of them are zeroable +@tailrec +def munch(s: List[Char], + pos: Int, + rs: LexRules, + last: Option[(Int, TokenFun)]): Option[(Int, TokenFun)] = { + rs match { + case Nil => last + case rs if (s.length <= pos) => last + case rs => { + val ders = rs.map({case (r, tf) => (der(s(pos), r), tf)}) + val rs_nzero = ders.filterNot({case (r, _) => zeroable(r)}) + val rs_nulls = ders.filter({case (r, _) => nullable(r)}) + val new_last = if (rs_nulls != Nil) Some((pos, rs_nulls.head._2)) else last + munch(s, 1 + pos, rs_nzero, new_last) + } +}} + +// iterates the munching function and returns a Token list +def tokenize(s: String, rs: LexRules) : List[Token] = munch(s.toList, 0, rs, None) match { + case None if (s == "") => Nil + case None => List(T_ERR(s"Lexing error: $s")) + case Some((n, tf)) => { + val (head, tail) = s.splitAt(n + 1) + tf(head)::tokenize(tail, rs) + } +} + +def tokenizer(s:String) : List[Token] = + tokenize(s, While_lexing_rules).filter { + case T_ERR(s) => { println(s); sys.exit(-1) } + case T_WHITESPACE => false + case T_COMMENT => false + case _ => true + } + +def fromFile(name: String) : String = + io.Source.fromFile(name).mkString + +// tokenizer tests +//println(tokenizer(fromFile("loops.while")).mkString("\n")) +//println(tokenizer(fromFile("fib.while")).mkString("\n")) +//println(tokenizer(fromFile("collatz.while")).mkString("\n")) +//println(tokenizer(fromFile("defs.rec")).mkString("\n")) + +// Parser - Abstract syntax trees +abstract class Exp +abstract class BExp +abstract class Decl + +case class Def(name: String, args: List[String], body: Exp) extends Decl +case class Main(e: Exp) extends Decl + +case class Call(name: String, args: List[Exp]) extends Exp +case class If(a: BExp, e1: Exp, e2: Exp) extends Exp +case class Read(s: String) extends Exp +case class Write(s: String) extends Exp +case class WriteS(s: String) extends Exp +case class Var(s: String) extends Exp +case class Num(i: Int) extends Exp +case class Aop(o: String, a1: Exp, a2: Exp) extends Exp + +case object True extends BExp +case object False extends BExp +case class Bop(o: String, a1: Exp, a2: Exp) extends BExp + +// Parser combinators +abstract class Parser[I <% Seq[_], 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 + + def parse_single(ts: I) : T = parse_all(ts).toList match { + case List(t) => t + case _ => { println ("Parse Error") ; sys.exit(-1) } + } +} + +class SeqParser[I <% Seq[_], 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 ((head1, head2), tail2) +} + +class AltParser[I <% Seq[_], 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 <% Seq[_], 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) +} + +case class TokParser(tok: Token) extends Parser[List[Token], Token] { + def parse(ts: List[Token]) = ts match { + case t::ts if (t == tok) => Set((t, ts)) + case _ => Set () + } +} + +implicit def token2tparser(t: Token) = TokParser(t) + +case object NumParser extends Parser[List[Token], Int] { + def parse(ts: List[Token]) = ts match { + case T_NUM(s)::ts => Set((s.toInt, ts)) + case _ => Set () + } +} + +case object IdParser extends Parser[List[Token], String] { + def parse(ts: List[Token]) = ts match { + case T_ID(s)::ts => Set((s, ts)) + case _ => Set () + } +} + +case object StringParser extends Parser[List[Token], String] { + def parse(ts: List[Token]) = ts match { + case T_STRING(s)::ts => Set((s, ts)) + case _ => Set () + } +} + +implicit def ParserOps[I<% Seq[_], 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 TokOps(t: Token) = new { + def || (q : => Parser[List[Token], Token]) = new AltParser[List[Token], Token](t, q) + def ==>[S] (f: => Token => S) = new FunParser[List[Token], Token, S](t, f) + def ~[S](q : => Parser[List[Token], S]) = new SeqParser[List[Token], Token, S](t, q) +} + +def ListParser[I <% Seq[_], T, S](p: => Parser[I, T], q: => Parser[I, S]): Parser[I, List[T]] = { + (p ~ q ~ ListParser(p, q)) ==> { case ((x, y), z) => x :: z : List[T] } || + (p ==> ((s) => List(s))) +} + + +// arithmetic expressions +lazy val Exp: Parser[List[Token], Exp] = + (IdParser ~ T_LPAREN ~ ListParser(Exp, T_COMMA) ~ T_RPAREN) ==> + { case (((x, y), z), w) => Call(x, z): Exp } || + (T_KWD("if") ~ BExp ~ T_KWD("then") ~ Exp ~ T_KWD("else") ~ Exp) ==> + { case (((((x, y), z), u), v), w) => If(y, u, w): Exp } || + (T ~ T_OP("+") ~ Exp) ==> { case ((x, y), z) => Aop("+", x, z): Exp } || + (T ~ T_OP("-") ~ Exp) ==> { case ((x, y), z) => Aop("-", x, z): Exp } || T +lazy val T: Parser[List[Token], Exp] = + (F ~ T_OP("*") ~ T) ==> { case ((x, y), z) => Aop("*", x, z): Exp } || + (F ~ T_OP("/") ~ T) ==> { case ((x, y), z) => Aop("/", x, z): Exp } || + (F ~ T_OP("%") ~ T) ==> { case ((x, y), z) => Aop("%", x, z): Exp } || F +lazy val F: Parser[List[Token], Exp] = + (T_LPAREN ~ Exp ~ T_RPAREN) ==> { case ((x, y), z) => y: Exp } || + IdParser ==> { case x => Var(x): Exp } || + NumParser ==> { case x => Num(x): Exp } + +// boolean expressions +lazy val BExp: Parser[List[Token], BExp] = + (Exp ~ T_OP("==") ~ Exp) ==> { case ((x, y), z) => Bop("==", x, z): BExp } || + (Exp ~ T_OP("!=") ~ Exp) ==> { case ((x, y), z) => Bop("!=", x, z): BExp } || + (Exp ~ T_OP("<") ~ Exp) ==> { case ((x, y), z) => Bop("<", x, z): BExp } || + (Exp ~ T_OP(">") ~ Exp) ==> { case ((x, y), z) => Bop("<", z, x): BExp } || + (T_KWD("true") ==> ((_) => True)) || + (T_KWD("false") ==> ((_) => False: BExp)) + +lazy val Defn: Parser[List[Token], Decl] = + (T_KWD("def") ~ IdParser ~ T_LPAREN ~ ListParser(IdParser, T_COMMA) ~ T_RPAREN ~ T_OP("=") ~ Exp) ==> + { case ((((((x, y), z), w), u), v), r) => Def(y, w, r): Decl } + +lazy val Prog: Parser[List[Token], List[Decl]] = + (Defn ~ T_SEMI ~ Prog) ==> { case ((x, y), z) => x :: z : List[Decl] } || + (Exp ==> ((s) => List(Main(s)) : List[Decl])) + +// parser examples + +val p11 = """def zero(x) = 0""" +val p11_toks = tokenizer(p11) +val p11_ast = Defn.parse_all(p11_toks) +//println(p11_toks) +//println(p11_ast) + + +val p12_toks = tokenizer(fromFile("defs.rec")) +val p12_ast = Prog.parse_all(p12_toks) +//println(p12_toks.mkString(",")) +//println(p12_ast) + + + +// compiler - built-in functions +// copied from http://www.ceng.metu.edu.tr/courses/ceng444/link/jvm-cpm.html +// + +val beginning = """ +.class public XXX.XXX +.super java/lang/Object + +.method public ()V + aload_0 + invokenonvirtual java/lang/Object/()V + return +.end method + +.method public static write(I)V + .limit locals 5 + .limit stack 5 + iload 0 + getstatic java/lang/System/out Ljava/io/PrintStream; + swap + invokevirtual java/io/PrintStream/println(I)V + return +.end method + +.method public static writes(Ljava/lang/String;)V + .limit stack 2 + .limit locals 2 + getstatic java/lang/System/out Ljava/io/PrintStream; + aload 0 + invokevirtual java/io/PrintStream/println(Ljava/lang/String;)V + return +.end method + +.method public static read()I + .limit locals 10 + .limit stack 10 + + ldc 0 + istore 1 ; this will hold our final integer +Label1: + getstatic java/lang/System/in Ljava/io/InputStream; + invokevirtual java/io/InputStream/read()I + istore 2 + iload 2 + ldc 10 ; the newline delimiter + isub + ifeq Label2 + iload 2 + ldc 32 ; the space delimiter + isub + ifeq Label2 + + iload 2 + ldc 48 ; we have our digit in ASCII, have to subtract it from 48 + isub + ldc 10 + iload 1 + imul + iadd + istore 1 + goto Label1 +Label2: + ;when we come here we have our integer computed in Local Variable 1 + iload 1 + ireturn +.end method + +.method public static main([Ljava/lang/String;)V + .limit locals 200 + .limit stack 200 + +""" + +val ending = """ + + return + +.end method +""" + +// for generating new labels +var counter = -1 + +def Fresh(x: String) = { + counter += 1 + x ++ "_" ++ counter.toString() +} + +type Mem = Map[String, String] +type Instrs = List[String] + +def compile_exp(a: Exp, env : Mem) : Instrs = a match { + case Num(i) => List("ldc " + i.toString + "\n") + case Var(s) => List("iload " + env(s) + "\n") + case Aop("+", a1, a2) => compile_exp(a1, env) ++ compile_exp(a2, env) ++ List("iadd\n") + case Aop("-", a1, a2) => compile_exp(a1, env) ++ compile_exp(a2, env) ++ List("isub\n") + case Aop("*", a1, a2) => compile_exp(a1, env) ++ compile_exp(a2, env) ++ List("imul\n") + case If(b, a1, a2) => { + val if_else = Fresh("If_else") + val if_end = Fresh("If_end") + compile_bexp(b, env, if_else) ++ + compile_exp(a1, env) ++ + List("goto " + if_end + "\n") ++ + List("\n" + if_else + ":\n\n") ++ + compile_exp(a2, env) ++ + List("\n" + if_end + ":\n\n") + } + case Call(n, args) => + args.flatMap(a => compile_exp(a, env)) ++ + List ("invokestatic XXX/XXX/" + n + "(I)I\n") + +} + +def compile_bexp(b: BExp, env : Mem, jmp: String) : Instrs = b match { + case True => Nil + case False => List("goto " + jmp + "\n") + case Bop("=", a1, a2) => + compile_exp(a1, env) ++ compile_exp(a2, env) ++ List("if_icmpne " + jmp + "\n") + case Bop("!=", a1, a2) => + compile_exp(a1, env) ++ compile_exp(a2, env) ++ List("if_icmpeq " + jmp + "\n") + case Bop("<", a1, a2) => + compile_exp(a1, env) ++ compile_exp(a2, env) ++ List("if_icmpge " + jmp + "\n") +} + + + +def compile_decl(d: Decl) : Instrs = d match { + case Def(name, args, a) => Nil + case Main(a) => compile_exp(a, Map()) +} + +def compile(class_name: String, input: String) : String = { + val tks = tokenizer(input) + val ast = Prog.parse_single(tks) + val instructions = ast.flatMap(compile_decl).mkString + (instructions).replaceAllLiterally("XXX", class_name) +} + + +def compile_file(file_name: String) = { + val class_name = file_name.split('.')(0) + val output = compile(class_name, fromFile(file_name)) + val fw = new java.io.FileWriter(class_name + ".j") + fw.write(output) + fw.close() +} + +def time_needed[T](i: Int, code: => T) = { + val start = System.nanoTime() + for (j <- 1 to i) code + val end = System.nanoTime() + (end - start)/(i * 1.0e9) +} + + +def compile_run(file_name: String) : Unit = { + val class_name = file_name.split('.')(0) + compile_file(file_name) + println(fromFile("defs.j")) + //val test = ("java -jar jvm/jasmin-2.4/jasmin.jar " + class_name + ".j").!! + //("java " + class_name + "/" + class_name).! +} + + +//examples +//println(compile("test", p9)) +//compile_run("loops.while") +compile_run("defs.rec") +//compile_run("test.while")