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();
}
pt = tokens = new Token(); // first token is a dummy
}
public void Parse()
{
la = new Token();
la.val = "";
Get();
KLAIM();
Expect(0);
}
// make sure that peeking starts at the current scan position
public void ResetPeek()
{
pt = tokens;
}
// 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;
}
}
private void SetStartPos(Node n, Token t)
{
n.LexicalInfo.StartLine = t.line;
n.LexicalInfo.StartColumn = t.col;
}
// peek for the next token, ignore pragmas
public Token Peek()
{
do {
if (pt.next == null) {
pt.next = NextToken();
}
pt = pt.next;
} while (pt.kind > maxT); // skip pragmas
return pt;
}
private void SetPos(Node n, Token start, Token end)
{
n.LexicalInfo.StartLine = start.line;
n.LexicalInfo.StartColumn = start.col;
n.LexicalInfo.EndLine = end.line;
n.LexicalInfo.EndColumn = end.col + end.val.Length;
}
private void SetPos(Node n, Token t)
{
n.LexicalInfo.StartLine = t.line;
n.LexicalInfo.StartColumn = t.col;
n.LexicalInfo.EndLine = t.line;
n.LexicalInfo.EndColumn = t.col + t.val.Length;
}
private void SetEndPos(Node n, Token t)
{
n.LexicalInfo.EndLine = t.line;
n.LexicalInfo.EndColumn = t.col;
}
private void CopyPos(Node n, Node source, Token end)
{
n.LexicalInfo.StartLine = source.LexicalInfo.StartLine;
n.LexicalInfo.StartColumn = source.LexicalInfo.StartColumn;
n.LexicalInfo.EndLine = t.line;
n.LexicalInfo.EndColumn = t.col;
}
Token NextToken()
{
while (ch == ' ' ||
ch >= 9 && ch <= 10 || ch == 13
) NextCh();
if (ch == '#' && Comment0()) return NextToken();
t = new Token();
t.pos = pos; t.col = col; t.line = line;
int state;
if (start.ContainsKey(ch)) { state = (int) start[ch]; }
else { 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 >= '0' && ch <= '9' || ch >= 'A' && ch <= 'Z' || ch >= 'a' && ch <= 'z') {AddCh(); goto case 1;}
else {t.kind = 1; break;}
case 2:
if (ch >= '0' && ch <= '9' || ch >= 'A' && ch <= 'Z' || ch >= 'a' && ch <= 'z') {AddCh(); goto case 2;}
else {t.kind = 2; t.val = new String(tval, 0, tlen); CheckLiteral(); return t;}
case 3:
if (ch >= 'a' && ch <= 'z') {AddCh(); goto case 4;}
else {t.kind = noSym; break;}
case 4:
if (ch >= '0' && ch <= '9' || ch >= 'A' && ch <= 'Z' || ch >= 'a' && ch <= 'z') {AddCh(); goto case 4;}
else {t.kind = 3; break;}
case 5:
if (ch >= '0' && ch <= '9') {AddCh(); goto case 5;}
else {t.kind = 4; break;}
case 6:
{t.kind = 5; break;}
case 7:
if (ch == ':') {AddCh(); goto case 8;}
else {t.kind = noSym; break;}
case 8:
{t.kind = 6; break;}
case 9:
{t.kind = 7; break;}
case 10:
{t.kind = 8; break;}
case 11:
{t.kind = 9; break;}
case 12:
{t.kind = 10; break;}
case 13:
{t.kind = 11; break;}
case 14:
{t.kind = 13; break;}
case 15:
{t.kind = 14; break;}
case 16:
{t.kind = 15; break;}
case 17:
{t.kind = 18; break;}
case 18:
{t.kind = 21; break;}
case 19:
{t.kind = 22; break;}
case 20:
{t.kind = 23; break;}
case 21:
if (ch == '|') {AddCh(); goto case 6;}
else {t.kind = 12; break;}
}
t.val = new String(tval, 0, tlen);
return t;
}
