public int Lex(IEnumerator <int> input, StringBuilder capture, out bool more)
{
var states = FillEpsilonClosure();
while (input.MoveNext())
{
more = true;
var next = FA.FillMove(states, input.Current);
if (0 == next.Count)
{
foreach (var state in states)
{
if (state.IsAccepting)
{
return(state.AcceptSymbol);
}
}
return(-1);
}
capture.Append(char.ConvertFromUtf32(input.Current));
states = next;
}
more = false;
foreach (var state in states)
{
if (state.IsAccepting)
{
return(state.AcceptSymbol);
}
}
return(-1);
}
/// <summary>
/// This is where the work happens
/// </summary>
/// <returns>The symbol that was matched. members _state _line,_column,_position,_buffer and _input are also modified.</returns>
int _Lex()
{
int acc;
var states = _initialStates;
_buffer.Clear();
switch (_state)
{
case -1: // initial
if (!_MoveNextInput())
{
_state = -2;
acc = _GetAcceptingSymbol(states);
if (-1 < acc)
{
return(acc);
}
else
{
return(-1); // "#ERROR";
}
}
_state = 0; // running
break;
case -2: // end of stream
return(-2); // "#EOS";
}
// Here's where we run most of the match. FillMove runs one interation of the NFA state machine.
// We match until we can't match anymore (greedy matching) and then report the symbol of the last
// match we found, or an error ("#ERROR") if we couldn't find one.
while (true)
{
var next = FA.FillMove(states, _input.Current);
if (0 == next.Count) // couldn't find any states
{
break;
}
_buffer.Append(_input.Current);
states = next;
if (!_MoveNextInput())
{
// end of stream
_state = -2;
acc = _GetAcceptingSymbol(states);
if (-1 < acc) // do we accept?
{
return(acc);
}
else
{
return(-1); // "#ERROR";
}
}
}
acc = _GetAcceptingSymbol(states);
if (-1 < acc) // do we accept?
{
return(acc);
}
else
{
// handle the error condition
_buffer.Append(_input.Current);
if (!_MoveNextInput())
{
_state = -2;
}
return(-1); // "#ERROR";
}
}
