public Token(Token next)
{
this.next = next;
}
// make sure that peeking starts at the current scan position
public void ResetPeek () { pt = tokens; }
// peek for the next token, ignore pragmas
public Token Peek () {
if (pt.next == null) {
do {
pt = pt.next = NextToken();
} while (pt.kind > maxT); // skip pragmas
} else {
do {
pt = pt.next;
} while (pt.kind > maxT);
}
return pt;
}
// get the next token (possibly a token already seen during peeking)
public Token Scan () {
if (tokens.next == null) {
return NextToken();
} else {
pt = tokens = tokens.next;
return tokens;
}
}
Token NextToken() {
while (false || ch >= 9 && ch <= 10 || ch == 13) NextCh();
int apx = 0;
t = new Token();
t.pos = pos; t.col = col; t.line = line;
int state;
try { state = start[ch]; } catch (KeyNotFoundException) { state = 0; }
tlen = 0; AddCh();
switch (state) {
case -1: { t.kind = eofSym; break; } // NextCh already done
case 0: { t.kind = noSym; break; } // NextCh already done
case 1:
if (ch == '$' || ch >= '0' && ch <= '9' || ch == 92 || ch == '_' || ch >= 'a' && ch <= 'z') {AddCh(); goto case 1;}
else {t.kind = 1; t.val = new String(tval, 0, tlen); CheckLiteral(); return t;}
case 2:
{t.kind = 2; break;}
case 3:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 4;}
else {t.kind = noSym; break;}
case 4:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 4;}
else {t.kind = 5; break;}
case 5:
{
tlen -= apx;
buffer.Pos = t.pos; NextCh(); line = t.line; col = t.col;
for (int i = 0; i < tlen; i++) NextCh();
t.kind = 5; break;}
case 6:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 7;}
else {t.kind = noSym; break;}
case 7:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 7;}
else if (ch == 'e') {AddCh(); goto case 8;}
else {t.kind = 6; break;}
case 8:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 10;}
else if (ch == '+' || ch == '-') {AddCh(); goto case 9;}
else {t.kind = noSym; break;}
case 9:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 10;}
else {t.kind = noSym; break;}
case 10:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 10;}
else {t.kind = 6; break;}
case 11:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 11;}
else if (ch == 'e') {AddCh(); goto case 12;}
else {t.kind = 6; break;}
case 12:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 14;}
else if (ch == '+' || ch == '-') {AddCh(); goto case 13;}
else {t.kind = noSym; break;}
case 13:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 14;}
else {t.kind = noSym; break;}
case 14:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 14;}
else {t.kind = 6; break;}
case 15:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 17;}
else if (ch == '+' || ch == '-') {AddCh(); goto case 16;}
else {t.kind = noSym; break;}
case 16:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 17;}
else {t.kind = noSym; break;}
case 17:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 17;}
else {t.kind = 6; break;}
case 18:
if (ch <= 9 || ch >= 11 && ch <= '!' || ch >= '#' && ch <= '[' || ch >= ']' && ch <= 65535) {AddCh(); goto case 18;}
else if (ch == '"') {AddCh(); goto case 34;}
else if (ch == 92) {AddCh(); goto case 37;}
else {t.kind = noSym; break;}
case 19:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 20;}
else {t.kind = noSym; break;}
case 20:
if (ch <= 9 || ch >= 11 && ch <= '!' || ch >= '#' && ch <= '/' || ch >= ':' && ch <= '[' || ch >= ']' && ch <= '`' || ch >= 'g' && ch <= 65535) {AddCh(); goto case 18;}
else if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 38;}
else if (ch == '"') {AddCh(); goto case 34;}
else if (ch == 92) {AddCh(); goto case 37;}
else {t.kind = noSym; break;}
case 21:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 22;}
else {t.kind = noSym; break;}
case 22:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 23;}
else {t.kind = noSym; break;}
case 23:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 18;}
else {t.kind = noSym; break;}
case 24:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 25;}
else {t.kind = noSym; break;}
case 25:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 26;}
else {t.kind = noSym; break;}
case 26:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 27;}
else {t.kind = noSym; break;}
case 27:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 28;}
else {t.kind = noSym; break;}
case 28:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 29;}
else {t.kind = noSym; break;}
case 29:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 30;}
else {t.kind = noSym; break;}
case 30:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 31;}
else {t.kind = noSym; break;}
case 31:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 18;}
else {t.kind = noSym; break;}
case 32:
if (ch == '"') {AddCh(); goto case 33;}
else {t.kind = noSym; break;}
case 33:
if (ch <= '!' || ch >= '#' && ch <= 65535) {AddCh(); goto case 33;}
else if (ch == '"') {AddCh(); goto case 40;}
else {t.kind = noSym; break;}
case 34:
{t.kind = 7; break;}
case 35:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 35;}
else if (ch == '.') {apx++; AddCh(); goto case 41;}
else if (ch == 'e') {AddCh(); goto case 15;}
else {t.kind = 5; break;}
case 36:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 35;}
else if (ch == '.') {apx++; AddCh(); goto case 41;}
else if (ch == 'x') {AddCh(); goto case 3;}
else if (ch == 'e') {AddCh(); goto case 15;}
else {t.kind = 5; break;}
case 37:
if (ch == '"' || ch == '$' || ch == 39 || ch == '0' || ch == 92 || ch >= 'a' && ch <= 'b' || ch == 'f' || ch == 'n' || ch == 'r' || ch == 't' || ch == 'v') {AddCh(); goto case 18;}
else if (ch == 'x') {AddCh(); goto case 19;}
else if (ch == 'u') {AddCh(); goto case 42;}
else {t.kind = noSym; break;}
case 38:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 39;}
else if (ch <= 9 || ch >= 11 && ch <= '!' || ch >= '#' && ch <= '/' || ch >= ':' && ch <= '[' || ch >= ']' && ch <= '`' || ch >= 'g' && ch <= 65535) {AddCh(); goto case 18;}
else if (ch == '"') {AddCh(); goto case 34;}
else if (ch == 92) {AddCh(); goto case 37;}
else {t.kind = noSym; break;}
case 39:
if (ch <= 9 || ch >= 11 && ch <= '!' || ch >= '#' && ch <= '[' || ch >= ']' && ch <= 65535) {AddCh(); goto case 18;}
else if (ch == '"') {AddCh(); goto case 34;}
else if (ch == 92) {AddCh(); goto case 37;}
else {t.kind = noSym; break;}
case 40:
if (ch == '"') {AddCh(); goto case 33;}
else {t.kind = 7; break;}
case 41:
if (ch <= '/' || ch >= ':' && ch <= 65535) {apx++; AddCh(); goto case 5;}
else if (ch >= '0' && ch <= '9') {apx = 0; AddCh(); goto case 11;}
else {t.kind = noSym; break;}
case 42:
if (ch >= '0' && ch <= '9' || ch >= 'a' && ch <= 'f') {AddCh(); goto case 21;}
else if (ch == 'l') {AddCh(); goto case 24;}
else {t.kind = noSym; break;}
case 43:
{t.kind = 8; break;}
case 44:
{t.kind = 9; break;}
case 45:
{t.kind = 10; break;}
}
t.val = new String(tval, 0, tlen);
return t;
}
void Init() {
pos = -1; line = 1; col = 0;
oldEols = 0;
NextCh();
if (ch == 0xEF) { // check optional byte order mark for UTF-8
NextCh(); int ch1 = ch;
NextCh(); int ch2 = ch;
if (ch1 != 0xBB || ch2 != 0xBF) {
throw new FatalError(String.Format("illegal byte order mark: EF {0,2:X} {1,2:X}", ch1, ch2));
}
buffer = new UTF8Buffer(buffer); col = 0;
NextCh();
}
start = new Dictionary<int, int>(128);
for (int i = 36; i <= 36; ++i) start[i] = 1;
for (int i = 92; i <= 92; ++i) start[i] = 1;
for (int i = 95; i <= 95; ++i) start[i] = 1;
for (int i = 97; i <= 122; ++i) start[i] = 1;
for (int i = 49; i <= 57; ++i) start[i] = 35;
start[126] = 2;
start[48] = 36;
start[46] = 6;
start[34] = 18;
start[64] = 32;
start[40] = 43;
start[44] = 44;
start[41] = 45;
start[Buffer.EOF] = -1;
pt = tokens = new Token(); // first token is a dummy
}
