public void RegexToDfaShouldConvertCharacterRegexToDfa()
{
var pattern = "a";
var regex = new RegexParser().Parse(pattern);
var nfa = new ThompsonConstructionAlgorithm().Transform(regex);
var dfa = new SubsetConstructionAlgorithm().Transform(nfa);
Assert.IsNotNull(dfa);
var lexerRule = new DfaLexerRule(dfa, "a");
var lexeme = new DfaLexemeFactory().Create(lexerRule);
Assert.IsTrue(lexeme.Scan('a'));
}
public void DfaLexemeShouldMatchOneOrMoreRandomWhitespaceCharacters()
{
var randomWhitespace = "\t\f\v \r\n";
var dfa = new DfaState();
var final = new DfaState(true);
dfa.AddTransition(new DfaTransition(new WhitespaceTerminal(), final));
final.AddTransition(new DfaTransition(new WhitespaceTerminal(), final));
var dfaLexerRule = new DfaLexerRule(dfa, new TokenType("whitespace"));
var whitespaceLexeme = new DfaLexeme(dfaLexerRule);
for (int i = 0; i < randomWhitespace.Length; i++)
Assert.IsTrue(whitespaceLexeme.Scan(randomWhitespace[i]));
}
public void DfaLexemeGivenCharacerLexemeNumberShouldFail()
{
var numberInput = "0";
var dfa = new DfaState();
var final = new DfaState(true);
dfa.AddTransition(new DfaTransition(new RangeTerminal('a', 'z'), final));
final.AddTransition(new DfaTransition(new RangeTerminal('a', 'z'), final));
var dfaLexerRule = new DfaLexerRule(dfa, new TokenType("lowerCase"));
var letterLexeme = new DfaLexeme(dfaLexerRule);
Assert.IsFalse(letterLexeme.Scan(numberInput[0]));
Assert.AreEqual(string.Empty, letterLexeme.Capture);
}
public void RegexToDfaShouldConvertOptionalCharacterClassToDfa()
{
var pattern = @"[-+]?[0-9]";
var regex = new RegexParser().Parse(pattern);
var nfa = new ThompsonConstructionAlgorithm().Transform(regex);
var dfa = new SubsetConstructionAlgorithm().Transform(nfa);
Assert.IsNotNull(dfa);
Assert.AreEqual(3, dfa.Transitions.Count);
var lexerRule = new DfaLexerRule(dfa, pattern);
AssertLexerRuleMatches(lexerRule, "+0");
AssertLexerRuleMatches(lexerRule, "-1");
AssertLexerRuleMatches(lexerRule, "9");
}
public void DfaLexemeShouldMatchMixedCaseWord()
{
var wordInput = "t90vAriabl3";
var dfa = new DfaState();
var final = new DfaState(true);
dfa.AddTransition(new DfaTransition(new RangeTerminal('a', 'z'), final));
dfa.AddTransition(new DfaTransition(new RangeTerminal('A', 'Z'), final));
final.AddTransition(new DfaTransition(new RangeTerminal('a', 'z'), final));
final.AddTransition(new DfaTransition(new RangeTerminal('A', 'Z'), final));
final.AddTransition(new DfaTransition(new DigitTerminal(), final));
var dfaLexerRule = new DfaLexerRule(dfa, new TokenType("Identifier"));
var indentifierLexeme = new DfaLexeme(dfaLexerRule);
for (int i = 0; i < wordInput.Length; i++)
Assert.IsTrue(indentifierLexeme.Scan(wordInput[i]));
}
private static BaseLexerRule CreateWhitespaceLexerRule()
{
var whitespaceTerminal = new WhitespaceTerminal();
var startWhitespace = new DfaState();
var finalWhitespace = new DfaState(true);
startWhitespace.AddTransition(new DfaTransition(whitespaceTerminal, finalWhitespace));
finalWhitespace.AddTransition(new DfaTransition(whitespaceTerminal, finalWhitespace));
var whitespace = new DfaLexerRule(startWhitespace, "whitespace");
return whitespace;
}
private static BaseLexerRule CreateIdentifierLexerRule()
{
// /[a-zA-Z][a-zA-Z0-9-_]*/
var identifierState = new DfaState();
var zeroOrMoreLetterOrDigit = new DfaState(true);
identifierState.AddTransition(
new DfaTransition(
new CharacterClassTerminal(
new RangeTerminal('a', 'z'),
new RangeTerminal('A', 'Z')),
zeroOrMoreLetterOrDigit));
zeroOrMoreLetterOrDigit.AddTransition(
new DfaTransition(
new CharacterClassTerminal(
new RangeTerminal('a', 'z'),
new RangeTerminal('A', 'Z'),
new DigitTerminal(),
new SetTerminal('-', '_')),
zeroOrMoreLetterOrDigit));
var identifier = new DfaLexerRule(identifierState, "identifier");
return identifier;
}
public DfaLexeme(DfaLexerRule lexerRule, int position)
: base(lexerRule, position)
{
this.captureBuilder = ObjectPoolExtensions.Allocate(SharedPools.Default <StringBuilder>());
this.currentState = lexerRule.Start;
}
private static ILexerRule CreateRuleNameLexerRule()
{
var ruleNameState = new DfaState();
var zeroOrMoreLetterOrDigit = new DfaState(true);
ruleNameState.AddTransition(
new DfaTransition(
new CharacterClassTerminal(
new RangeTerminal('a', 'z'),
new RangeTerminal('A', 'Z')),
zeroOrMoreLetterOrDigit));
zeroOrMoreLetterOrDigit.AddTransition(
new DfaTransition(
new CharacterClassTerminal(
new RangeTerminal('a', 'z'),
new RangeTerminal('A', 'Z'),
new DigitTerminal(),
new SetTerminal('-', '_')),
zeroOrMoreLetterOrDigit));
var ruleName = new DfaLexerRule(ruleNameState, new TokenType("rule-name"));
return ruleName;
}
IEnumerable<ProductionModel> Factor(EbnfFactor factor, ProductionModel currentProduction)
{
switch (factor.NodeType)
{
case EbnfNodeType.EbnfFactorGrouping:
var grouping = factor as EbnfFactorGrouping;
foreach (var production in Grouping(grouping, currentProduction))
yield return production;
break;
case EbnfNodeType.EbnfFactorOptional:
var optional = factor as EbnfFactorOptional;
foreach (var production in Optional(optional, currentProduction))
yield return production;
break;
case EbnfNodeType.EbnfFactorRepetition:
var repetition = factor as EbnfFactorRepetition;
foreach (var production in Repetition(repetition, currentProduction))
yield return production;
break;
case EbnfNodeType.EbnfFactorIdentifier:
var identifier = factor as EbnfFactorIdentifier;
var nonTerminal = GetNonTerminalFromQualifiedIdentifier(identifier.QualifiedIdentifier);
currentProduction.AddWithAnd(new NonTerminalModel(nonTerminal));
break;
case EbnfNodeType.EbnfFactorLiteral:
var literal = factor as EbnfFactorLiteral;
var stringLiteralRule = new StringLiteralLexerRule(literal.Value);
currentProduction.AddWithAnd( new LexerRuleModel(stringLiteralRule));
break;
case EbnfNodeType.EbnfFactorRegex:
var regex = factor as EbnfFactorRegex;
var nfa = _thompsonConstructionAlgorithm.Transform(regex.Regex);
var dfa = _subsetConstructionAlgorithm.Transform(nfa);
var dfaLexerRule = new DfaLexerRule(dfa, regex.Regex.ToString());
currentProduction.AddWithAnd(new LexerRuleModel(dfaLexerRule));
break;
}
}
