public void MatchTest(string regex, bool caseSensitive, string input, bool match)
{
var expression = RegexParser.Parse(regex, 0);
var mapper = new UnicodeCodepointMapper(false, false);
var rxCodepoint = expression.ToInvariant(mapper, new UnicodeCharSetProvider(), caseSensitive);
var nfa = NfaBuilder <Codepoint> .Build(rxCodepoint.Optimize());
var dfa = DfaBuilder <Codepoint> .Build(nfa, Codepoints.EOF);
var state = dfa.StartState;
foreach (var codepoint in input.ToCodepoints().Append(Codepoints.EOF))
{
var nextStateId = state.GetTransition(codepoint);
if (nextStateId == Dfa <Codepoint> .Reject)
{
Assert.False(match, "Regex has not matched input, but expected a match");
break;
}
if (nextStateId == Dfa <Codepoint> .Accept)
{
Assert.True(match, "Regex has matched input, but expected a non-match");
break;
}
state = dfa.GetState(nextStateId);
Assert.NotNull(state);
}
}
public void MatchTestUtf8(string regex, bool caseSensitive, byte[] input, bool match)
{
var inputString = Encoding.UTF8.GetString(input);
this.output.WriteLine(inputString);
this.output.WriteLine(string.Join(" ", inputString.Select(c => ((int)c).ToString("x4"))));
var expression = RegexParser.Parse(regex, 0);
var mapper = new UnicodeUtf8Mapper(false, false);
var rxCodepoint = expression.ToInvariant(mapper, new UnicodeCharSetProvider(), caseSensitive);
var nfa = NfaBuilder <byte> .Build(rxCodepoint.Optimize());
var dfa = DfaBuilder <byte> .Build(nfa, Utf8Bytes.EOF);
var state = dfa.StartState;
foreach (var inputByte in input.Append(Utf8Bytes.EOF))
{
var nextStateId = state.GetTransition(inputByte);
if (nextStateId == Dfa <byte> .Reject)
{
Assert.False(match, "Regex has not matched input, but expected a match");
break;
}
if (nextStateId == Dfa <byte> .Accept)
{
Assert.True(match, "Regex has matched input, but expected a non-match");
break;
}
state = dfa.GetState(nextStateId);
Assert.NotNull(state);
}
}
public void CreateDiagram(string regex, bool caseSensitive)
{
var expression = RegexParser.Parse(regex, 0);
var mapper = new UnicodeCodepointMapper(false, false);
var rxCodepoint = expression.ToInvariant(mapper, new UnicodeCharSetProvider(), caseSensitive);
var rxAccept = new RxAccept <Codepoint>(rxCodepoint, default(SymbolId), null);
var nfa = NfaBuilder <Codepoint> .Build(rxAccept);
WriteDiagram(this.output, nfa);
}
public void CreateUtf8Diagram(string regex, bool caseSensitive)
{
var expression = RegexParser.Parse(regex, 0);
var mapper = new UnicodeUtf8Mapper(false, false);
var rxCodepoint = expression.ToInvariant(mapper, new UnicodeCharSetProvider(), caseSensitive);
var nfa = NfaBuilder <byte> .Build(rxCodepoint.Optimize());
var dfa = DfaBuilder <byte> .Build(nfa, 0xFF);
NfaFactoryTest.WriteDiagram(this.output, nfa);
WriteDiagram(this.output, dfa);
}
public void Actions()
{
var symbolA = 1;
var regexA = "a";
var r1 = RegexParser.Parse(regexA, symbolA);
var symbolB = 2;
var regexB = "b";
var r2 = RegexParser.Parse(regexB, symbolB);
RegexExpression r = new RegexAlternation(r1, r2);
var mapper = new UnicodeCodepointMapper(false, false);
var rxCodepoint = r.ToInvariant(mapper, new UnicodeCharSetProvider(), true);
var nfa = NfaBuilder <Codepoint> .Build(rxCodepoint.Optimize());
NfaFactoryTest.WriteDiagram(this.output, nfa);
var dfa = DfaBuilder <Codepoint> .Build(nfa, Codepoints.EOF);
WriteDiagram(this.output, dfa);
var state = dfa.StartState;
foreach (var codepoint in regexA.ToCodepoints().Append(Codepoints.EOF))
{
var nextStateId = state.GetTransition(codepoint);
Assert.NotEqual(Dfa <Codepoint> .Reject, nextStateId);
if (nextStateId == Dfa <Codepoint> .Accept)
{
Assert.Equal(dfa.SymbolStates[state.Id], symbolA);
break;
}
state = dfa.GetState(nextStateId);
Assert.NotNull(state);
}
state = dfa.StartState;
foreach (var codepoint in regexB.ToCodepoints().Append(Codepoints.EOF))
{
var nextStateId = state.GetTransition(codepoint);
Assert.NotEqual(Dfa <Codepoint> .Reject, nextStateId);
if (nextStateId == Dfa <Codepoint> .Accept)
{
Assert.Equal(dfa.SymbolStates[state.Id], symbolB);
break;
}
state = dfa.GetState(nextStateId);
Assert.NotNull(state);
}
}