ADVANCED-anchored-classes.sleela
lib/examples/regex/ADVANCED-anchored-classes.sleela · 60 line(s) · sleela
#sleela 1.7
// ADVANCED: a tiny anchored matcher with character classes and a one-or-more
// quantifier, enough to validate a simple token like an identifier or a number.
// Implemented on arrays of character codes, no regex intrinsic involved.
//
// Rules demonstrated (all anchored to the whole input):
// isDigits(text) -> text is one or more ASCII digits (like ^[0-9]+$)
// isIdentifier(text) -> starts with a letter/underscore,
// then letters/digits/underscores (like ^[A-Za-z_]\w*$)
class ADVAnchoredClasses {
boolean isDigit(int c) { return c >= 48 && c <= 57; }
boolean isLower(int c) { return c >= 97 && c <= 122; }
boolean isUpper(int c) { return c >= 65 && c <= 90; }
boolean isAlpha(int c) { return isLower(c) || isUpper(c); }
boolean isUnder(int c) { return c == 95; }
// ^[0-9]+$
boolean isDigits(int[] t) {
if (arrayLength(t) == 0) return false;
int i = 0;
while (i < arrayLength(t)) {
if (!isDigit(t[i])) return false;
i = i + 1;
}
return true;
}
// ^[A-Za-z_][A-Za-z0-9_]*$
boolean isIdentifier(int[] t) {
if (arrayLength(t) == 0) return false;
int first = t[0];
if (!isAlpha(first) && !isUnder(first)) return false;
int i = 1;
while (i < arrayLength(t)) {
int c = t[i];
if (!isAlpha(c) && !isDigit(c) && !isUnder(c)) return false;
i = i + 1;
}
return true;
}
void main() {
// "2048"
int[] num = new int[4];
num[0]=50; num[1]=48; num[2]=52; num[3]=56;
// "sleela_v2"
int[] id = new int[9];
id[0]=115; id[1]=108; id[2]=101; id[3]=101; id[4]=108;
id[5]=97; id[6]=95; id[7]=118; id[8]=50;
// "2bad" -- starts with a digit, not a valid identifier
int[] bad = new int[4];
bad[0]=50; bad[1]=98; bad[2]=97; bad[3]=100;
print("isDigits('2048') = " + isDigits(num));
print("isDigits('sleela_v2') = " + isDigits(id));
print("isIdentifier('sleela_v2')= " + isIdentifier(id));
print("isIdentifier('2bad') = " + isIdentifier(bad));
}
}