258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
1 |
abstract class Rexp
|
49
|
2 |
case object NULL extends Rexp
|
|
3 |
case object EMPTY extends Rexp
|
|
4 |
case class CHAR(c: Char) extends Rexp
|
258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
5 |
case class ALT(r1: Rexp, r2: Rexp) extends Rexp
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
6 |
case class SEQ(r1: Rexp, r2: Rexp) extends Rexp
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
7 |
case class STAR(r: Rexp) extends Rexp
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
8 |
case class NTIMES(r: Rexp, n: Int) extends Rexp
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
9 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
10 |
def simp(r: Rexp): Rexp = r match {
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
11 |
case ALT(r1, r2) => {
|
341
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
12 |
(simp(r1), simp(r2)) match {
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
13 |
case (NULL, r2s) => r2s
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
14 |
case (r1s, NULL) => r1s
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
15 |
case (r1s, r2s) => if (r1s == r2s) r1s else ALT(r1s, r2s)
|
258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
16 |
}
|
49
|
17 |
}
|
258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
18 |
case SEQ(r1, r2) => {
|
341
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
19 |
(simp(r1), simp(r2)) match {
|
258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
20 |
case (NULL, _) => NULL
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
21 |
case (_, NULL) => NULL
|
341
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
22 |
case (EMPTY, r2s) => r2s
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
23 |
case (r1s, EMPTY) => r1s
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
24 |
case (r1s, r2s) => SEQ(r1s, r2s)
|
258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
25 |
}
|
49
|
26 |
}
|
258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
27 |
case NTIMES(r, n) => NTIMES(simp(r), n)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
28 |
case r => r
|
49
|
29 |
}
|
|
30 |
|
|
31 |
def nullable (r: Rexp) : Boolean = r match {
|
|
32 |
case NULL => false
|
|
33 |
case EMPTY => true
|
|
34 |
case CHAR(_) => false
|
|
35 |
case ALT(r1, r2) => nullable(r1) || nullable(r2)
|
|
36 |
case SEQ(r1, r2) => nullable(r1) && nullable(r2)
|
|
37 |
case STAR(_) => true
|
59
|
38 |
case NTIMES(r, i) => if (i == 0) true else nullable(r)
|
49
|
39 |
}
|
|
40 |
|
|
41 |
def der (c: Char, r: Rexp) : Rexp = r match {
|
|
42 |
case NULL => NULL
|
|
43 |
case EMPTY => NULL
|
|
44 |
case CHAR(d) => if (c == d) EMPTY else NULL
|
|
45 |
case ALT(r1, r2) => ALT(der(c, r1), der(c, r2))
|
|
46 |
case SEQ(r1, r2) =>
|
|
47 |
if (nullable(r1)) ALT(SEQ(der(c, r1), r2), der(c, r2))
|
|
48 |
else SEQ(der(c, r1), r2)
|
|
49 |
case STAR(r) => SEQ(der(c, r), STAR(r))
|
|
50 |
case NTIMES(r, i) =>
|
|
51 |
if (i == 0) NULL else SEQ(der(c, r), NTIMES(r, i - 1))
|
|
52 |
}
|
|
53 |
|
|
54 |
def ders (s: List[Char], r: Rexp) : Rexp = s match {
|
|
55 |
case Nil => r
|
258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
56 |
case c::s => ders(s, simp(der(c, r)))
|
49
|
57 |
}
|
|
58 |
|
261
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
59 |
def matches(r: Rexp, s: String) : Boolean = nullable(ders(s.toList, r))
|
49
|
60 |
|
|
61 |
|
|
62 |
//one or zero
|
|
63 |
def OPT(r: Rexp) = ALT(r, EMPTY)
|
|
64 |
|
258
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
65 |
def EVIL(n: Int) = SEQ(NTIMES(OPT(CHAR('a')), n), NTIMES(CHAR('a'), n))
|
49
|
66 |
|
|
67 |
def time_needed[T](i: Int, code: => T) = {
|
|
68 |
val start = System.nanoTime()
|
|
69 |
for (j <- 1 to i) code
|
|
70 |
val end = System.nanoTime()
|
|
71 |
(end - start)/(i * 1.0e9)
|
|
72 |
}
|
|
73 |
|
|
74 |
|
121
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
75 |
for (i <- 1 to 12001 by 500) {
|
261
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
76 |
println(i + " " + "%.5f".format(time_needed(1, matches(EVIL(i), "a" * i))))
|
49
|
77 |
}
|
|
78 |
|
|
79 |
|
363
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
80 |
// some convenience for typing in regular expressions
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
81 |
import scala.language.implicitConversions
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
82 |
import scala.language.reflectiveCalls
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
83 |
def charlist2rexp(s : List[Char]) : Rexp = s match {
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
84 |
case Nil => EMPTY
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
85 |
case c::Nil => CHAR(c)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
86 |
case c::s => SEQ(CHAR(c), charlist2rexp(s))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
87 |
}
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
88 |
implicit def string2rexp(s : String) : Rexp = charlist2rexp(s.toList)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
89 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
90 |
implicit def RexpOps (r: Rexp) = new {
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
91 |
def | (s: Rexp) = ALT(r, s)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
92 |
def % = STAR(r)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
93 |
def ~ (s: Rexp) = SEQ(r, s)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
94 |
}
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
95 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
96 |
implicit def stringOps (s: String) = new {
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
97 |
def | (r: Rexp) = ALT(s, r)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
98 |
def | (r: String) = ALT(s, r)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
99 |
def % = STAR(s)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
100 |
def ~ (r: Rexp) = SEQ(s, r)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
101 |
def ~ (r: String) = SEQ(s, r)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
102 |
}
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
103 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
104 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
105 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
106 |
def PLUS(r: Rexp) = SEQ(r, STAR(r))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
107 |
def RANGE(s: List[Char]) : Rexp = s match {
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
108 |
case Nil => NULL
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
109 |
case c::s => ALT(CHAR(c), RANGE(s))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
110 |
}
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
111 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
112 |
println("EVILS:")
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
113 |
val EVIL1 = PLUS(PLUS("a" ~ "a" ~ "a"))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
114 |
val EVIL2 = PLUS(PLUS("aaaaaaaaaaaaaaaaaaa" ~ OPT("a"))) // 19 as ~ a?
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
115 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
116 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
117 |
//40 * aaa matches
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
118 |
//43 * aaa + aa does not
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
119 |
//45 * aaa + a
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
120 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
121 |
println("EVIL1:")
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
122 |
println(matches(EVIL1, "aaa" * 40))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
123 |
println(matches(EVIL1, "aaa" * 43 + "aa"))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
124 |
println(matches(EVIL1, "aaa" * 45 + "a"))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
125 |
println("EVIL2:")
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
126 |
println(matches(EVIL2, "aaa" * 40))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
127 |
println(matches(EVIL2, "aaa" * 43 + "aa"))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
128 |
println(matches(EVIL2, "aaa" * 45 + "a"))
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
129 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
130 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
131 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
132 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
133 |
println("EMAIL:")
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
134 |
val LOWERCASE = "abcdefghijklmnopqrstuvwxyz"
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
135 |
val DIGITS = "0123456789"
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
136 |
val EMAIL = PLUS(RANGE((LOWERCASE + DIGITS + "_" + "." + "-").toList)) ~ "@" ~
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
137 |
PLUS(RANGE((LOWERCASE + DIGITS + "." + "-").toList)) ~ "." ~
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
138 |
NMTIMES(RANGE((LOWERCASE + ".").toList), 2, 6)
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
139 |
|
Christian Urban <christian dot urban at kcl dot ac dot uk>
diff
changeset
|
140 |
val my_email = "christian.urban@kcl.ac.uk"
|