public void LexerStateToDFATest()
{
Lexicon lexicon = new Lexicon();
LexerState global = lexicon.DefaultLexer;
LexerState keywords = global.CreateSubState();
LexerState xml = keywords.CreateSubState();
var ID = global.DefineToken(RE.Range('a', 'z').Concat(
(RE.Range('a', 'z') | RE.Range('0', '9')).Many()));
var NUM = global.DefineToken(RE.Range('0', '9').Many1());
var ERROR = global.DefineToken(RE.Range(Char.MinValue, (char)255));
var IF = keywords.DefineToken(RE.Literal("if"));
var ELSE = keywords.DefineToken(RE.Literal("else"));
var XMLNS = xml.DefineToken(RE.Literal("xmlns"));
DFAModel dfa = DFAModel.Create(lexicon);
CompressedTransitionTable tc = CompressedTransitionTable.Compress(dfa);
ScannerInfo si = lexicon.CreateScannerInfo();
FiniteAutomationEngine engine = new FiniteAutomationEngine(si.TransitionTable, si.CharClassTable);
engine.InputString("if");
Assert.AreEqual(ID.Index, si.GetTokenIndex(engine.CurrentState));
engine.Reset();
engine.InputString("12345");
Assert.AreEqual(NUM.Index, si.GetTokenIndex(engine.CurrentState));
engine.Reset();
engine.InputString("asdf12dd");
Assert.AreEqual(ID.Index, si.GetTokenIndex(engine.CurrentState));
engine.Reset();
engine.InputString("A");
Assert.AreEqual(ERROR.Index, si.GetTokenIndex(engine.CurrentState));
engine.Reset();
engine.InputString("AAA");
Assert.IsTrue(engine.IsAtStoppedState);
engine.Reset();
engine.InputString("if ");
Assert.IsTrue(engine.IsAtStoppedState);
engine.Reset();
si.LexerStateIndex = keywords.Index;
engine.InputString("if");
Assert.AreEqual(IF.Index, si.GetTokenIndex(engine.CurrentState));
engine.Reset();
engine.InputString("else");
Assert.AreEqual(ELSE.Index, si.GetTokenIndex(engine.CurrentState));
engine.Reset();
engine.InputString("xmlns");
Assert.AreEqual(ID.Index, si.GetTokenIndex(engine.CurrentState));
engine.Reset();
si.LexerStateIndex = xml.Index;
engine.InputString("if");
Assert.AreEqual(IF.Index, si.GetTokenIndex(engine.CurrentState));
engine.Reset();
engine.InputString("xml");
Assert.IsFalse(engine.IsAtStoppedState);
engine.Reset();
engine.InputString("xmlns");
Assert.AreEqual(XMLNS.Index, si.GetTokenIndex(engine.CurrentState));
;
}
public void CompactCharSetTest()
{
Lexicon lexicon = new Lexicon();
LexerState global = lexicon.DefaultLexer;
LexerState keywords = global.CreateSubState();
LexerState xml = keywords.CreateSubState();
var lettersCategories = new[] { UnicodeCategory.LetterNumber,
UnicodeCategory.LowercaseLetter,
UnicodeCategory.ModifierLetter,
UnicodeCategory.OtherLetter,
UnicodeCategory.TitlecaseLetter,
UnicodeCategory.UppercaseLetter };
var RE_IDCHAR = RE.CharsOf(c => lettersCategories.Contains(Char.GetUnicodeCategory(c)));
var ID = global.DefineToken(RE_IDCHAR.Concat(
(RE_IDCHAR | RE.Range('0', '9')).Many()));
var NUM = global.DefineToken(RE.Range('0', '9').Many1());
var WHITESPACE = global.DefineToken(RE.Symbol(' ').Many());
var IF = keywords.DefineToken(RE.Literal("if"));
var ELSE = keywords.DefineToken(RE.Literal("else"));
var XMLNS = xml.DefineToken(RE.Literal("xmlns"));
var scannerInfo = lexicon.CreateScannerInfo();
scannerInfo.LexerStateIndex = xml.Index;
Scanner s = new Scanner(scannerInfo);
string source = "xmlns 你好吗1 123 蘏臦囧綗 AＢＣＤ if";
SourceReader sr = new SourceReader(new StringReader(source));
s.SetSource(sr);
s.SetTriviaTokens(WHITESPACE.Index);
var l1 = s.Read();
Assert.AreEqual(XMLNS.Index, l1.TokenIndex);
var l2 = s.Read();
Assert.AreEqual(ID.Index, l2.TokenIndex);
var l3 = s.Read();
Assert.AreEqual(NUM.Index, l3.TokenIndex);
var l4 = s.Read();
Assert.AreEqual(ID.Index, l4.TokenIndex);
var l5 = s.Read();
Assert.AreEqual(ID.Index, l5.TokenIndex);
var l6 = s.Read();
Assert.AreEqual(IF.Index, l6.TokenIndex);
}
public void SkipTokenTest()
{
Lexicon lexicon = new Lexicon();
LexerState global = lexicon.DefaultLexer;
LexerState keywords = global.CreateSubState();
LexerState xml = keywords.CreateSubState();
var ID = global.DefineToken(RE.Range('a', 'z').Concat(
(RE.Range('a', 'z') | RE.Range('0', '9')).Many()));
var NUM = global.DefineToken(RE.Range('0', '9').Many1());
var WHITESPACE = global.DefineToken(RE.Symbol(' ').Many());
var ERROR = global.DefineToken(RE.Range(Char.MinValue, (char)255));
var IF = keywords.DefineToken(RE.Literal("if"));
var ELSE = keywords.DefineToken(RE.Literal("else"));
var XMLNS = xml.DefineToken(RE.Literal("xmlns"));
ScannerInfo info = lexicon.CreateScannerInfo();
PeekableScanner scanner = new PeekableScanner(info);
string source = "asdf04a 1107 else Z if vvv xmlns 772737";
StringReader sr = new StringReader(source);
scanner.SetSource(new SourceReader(sr));
scanner.SetTriviaTokens(WHITESPACE.Index, ERROR.Index);
info.LexerStateIndex = xml.Index;
Lexeme l1 = scanner.Read();
Assert.AreEqual(ID.Index, l1.TokenIndex);
Assert.AreEqual("asdf04a", l1.Value);
Assert.AreEqual(0, l1.PrefixTrivia.Count);
Lexeme l2 = scanner.Read();
Assert.AreEqual(NUM.Index, l2.TokenIndex);
Assert.AreEqual("1107", l2.Value);
Assert.AreEqual(1, l2.PrefixTrivia.Count);
Lexeme l3 = scanner.Read();
Assert.AreEqual(ELSE.Index, l3.TokenIndex);
Assert.AreEqual("else", l3.Value);
Assert.AreEqual(1, l2.PrefixTrivia.Count);
Lexeme l4 = scanner.Read();
Assert.AreEqual(IF.Index, l4.TokenIndex);
Assert.AreEqual("if", l4.Value);
Assert.AreEqual(3, l4.PrefixTrivia.Count);
int p1 = scanner.Peek();
Assert.AreEqual(ID.Index, p1);
int p2 = scanner.Peek2();
int p3 = scanner.Peek(3);
int peof = scanner.Peek(4);
Assert.AreEqual(info.EndOfStreamTokenIndex, peof);
Lexeme l6 = scanner.Read();
Lexeme l7 = scanner.Read();
Assert.AreEqual(XMLNS.Index, l7.TokenIndex);
Lexeme l8 = scanner.Read();
Assert.AreEqual(NUM.Index, l8.TokenIndex);
Lexeme leof = scanner.Read();
Assert.AreEqual(info.EndOfStreamTokenIndex, leof.TokenIndex);
Assert.AreEqual(leof.Span.StartLocation.CharIndex, leof.Span.EndLocation.CharIndex);
Assert.AreEqual(source.Length, leof.Span.StartLocation.CharIndex);
}
public void ScannerTest()
{
Lexicon lexicon = new Lexicon();
LexerState global = lexicon.DefaultLexer;
LexerState keywords = global.CreateSubState();
LexerState xml = keywords.CreateSubState();
var ID = global.DefineToken(RE.Range('a', 'z').Concat(
(RE.Range('a', 'z') | RE.Range('0', '9')).Many()));
var NUM = global.DefineToken(RE.Range('0', '9').Many1());
var WHITESPACE = global.DefineToken(RE.Symbol(' ').Many());
var ERROR = global.DefineToken(RE.Range(Char.MinValue, (char)255));
var IF = keywords.DefineToken(RE.Literal("if"));
var ELSE = keywords.DefineToken(RE.Literal("else"));
var XMLNS = xml.DefineToken(RE.Literal("xmlns"));
ScannerInfo info = lexicon.CreateScannerInfo();
PeekableScanner scanner = new PeekableScanner(info);
string source = "asdf04a 1107 else Z if vvv xmlns 772737";
StringReader sr = new StringReader(source);
scanner.SetSource(new SourceReader(sr));
Lexeme l1 = scanner.Read();
Assert.AreEqual(ID.Index, l1.TokenIndex);
Assert.AreEqual("asdf04a", l1.Value);
Assert.AreEqual(0, l1.Span.StartLocation.Column);
Assert.AreEqual(6, l1.Span.EndLocation.Column);
Lexeme l2 = scanner.Read();
Assert.AreEqual(WHITESPACE.Index, l2.TokenIndex);
Assert.AreEqual(" ", l2.Value);
Lexeme l3 = scanner.Read();
Assert.AreEqual(NUM.Index, l3.TokenIndex);
Assert.AreEqual("1107", l3.Value);
Lexeme l4 = scanner.Read();
Assert.AreEqual(WHITESPACE.Index, l4.TokenIndex);
Lexeme l5 = scanner.Read();
Assert.AreEqual(ID.Index, l5.TokenIndex);
int p1 = scanner.Peek();
Assert.AreEqual(WHITESPACE.Index, p1);
int p2 = scanner.Peek2();
Assert.AreEqual(ERROR.Index, p2);
int p3 = scanner.Peek(3);
Assert.AreEqual(WHITESPACE.Index, p3);
int p4 = scanner.Peek(4);
Assert.AreEqual(ID.Index, p4);
int p5 = scanner.Peek(5);
Assert.AreEqual(WHITESPACE.Index, p5);
Lexeme l6 = scanner.Read();
Lexeme l7 = scanner.Read();
Assert.AreEqual(ERROR.Index, l7.TokenIndex);
int p3_2 = scanner.Peek();
Assert.AreEqual(p3, p3_2);
Lexeme l8 = scanner.Read(); // whitespace
Lexeme l9 = scanner.Read(); // ID:if
Lexeme l10 = scanner.Read(); // whitespace
Lexeme l11 = scanner.Read(); // ID:vvv
Lexeme l12 = scanner.Read(); // whitespace
Lexeme l13 = scanner.Read(); // ID:xmlns
Lexeme l14 = scanner.Read(); // whitespace
Lexeme l15 = scanner.Read(); // NUM:772737
Lexeme leof = scanner.Read(); // eof
Assert.AreEqual(info.EndOfStreamTokenIndex, leof.TokenIndex);
Assert.AreEqual(leof.Span.StartLocation.CharIndex, leof.Span.EndLocation.CharIndex);
Assert.AreEqual(source.Length, leof.Span.StartLocation.CharIndex);
Lexeme leof2 = scanner.Read(); //after eof, should return eof again
Assert.AreEqual(info.EndOfStreamTokenIndex, leof2.TokenIndex);
Assert.AreEqual(leof.Span.StartLocation.CharIndex, leof2.Span.StartLocation.CharIndex);
}
protected override void OnDefineLexer(Compilers.Scanners.Lexicon lexicon, ICollection <Token> skippedTokens)
{
var lettersCategories = new HashSet <UnicodeCategory>()
{
UnicodeCategory.LetterNumber,
UnicodeCategory.LowercaseLetter,
UnicodeCategory.ModifierLetter,
UnicodeCategory.OtherLetter,
UnicodeCategory.TitlecaseLetter,
UnicodeCategory.UppercaseLetter
};
RE RE_IdChar = null;
RE RE_SpaceChar = null;
RE RE_InputChar = null;
RE RE_NotSlashOrAsterisk = null;
CharSetExpressionBuilder charSetBuilder = new CharSetExpressionBuilder();
charSetBuilder.DefineCharSet(c => lettersCategories.Contains(Char.GetUnicodeCategory(c)), re => RE_IdChar = re | RE.Symbol('_'));
charSetBuilder.DefineCharSet(c => Char.GetUnicodeCategory(c) == UnicodeCategory.SpaceSeparator, re => RE_SpaceChar = re);
charSetBuilder.DefineCharSet(c => "\u000D\u000A\u0085\u2028\u2029".IndexOf(c) < 0, re => RE_InputChar = re);
charSetBuilder.DefineCharSet(c => "/*".IndexOf(c) < 0, re => RE_NotSlashOrAsterisk = re);
charSetBuilder.Build();
var lex = lexicon.Lexer;
//keywords
K_CLASS = lex.DefineToken(RE.Literal("class"));
K_PUBLIC = lex.DefineToken(RE.Literal("public"));
K_STATIC = lex.DefineToken(RE.Literal("static"));
K_VOID = lex.DefineToken(RE.Literal("void"));
K_MAIN = lex.DefineToken(RE.Literal("Main"));
K_STRING = lex.DefineToken(RE.Literal("string"));
K_RETURN = lex.DefineToken(RE.Literal("return"));
K_INT = lex.DefineToken(RE.Literal("int"));
K_BOOL = lex.DefineToken(RE.Literal("bool"));
K_IF = lex.DefineToken(RE.Literal("if"));
K_ELSE = lex.DefineToken(RE.Literal("else"));
K_WHILE = lex.DefineToken(RE.Literal("while"));
K_SYSTEM = lex.DefineToken(RE.Literal("System"));
K_CONSOLE = lex.DefineToken(RE.Literal("Console"));
K_WRITELINE = lex.DefineToken(RE.Literal("WriteLine"));
K_LENGTH = lex.DefineToken(RE.Literal("Length"));
K_TRUE = lex.DefineToken(RE.Literal("true"));
K_FALSE = lex.DefineToken(RE.Literal("false"));
K_THIS = lex.DefineToken(RE.Literal("this"));
K_NEW = lex.DefineToken(RE.Literal("new"));
//id & literals
ID = lex.DefineToken(RE_IdChar >>
(RE_IdChar | RE.Range('0', '9')).Many(), "identifier");
INTEGER_LITERAL = lex.DefineToken(RE.Range('0', '9').Many1(), "integer literal");
//symbols
LOGICAL_AND = lex.DefineToken(RE.Literal("&&"));
LOGICAL_OR = lex.DefineToken(RE.Literal("||"));
LOGICAL_NOT = lex.DefineToken(RE.Symbol('!'));
LESS = lex.DefineToken(RE.Symbol('<'));
GREATER = lex.DefineToken(RE.Symbol('>'));
EQUAL = lex.DefineToken(RE.Literal("=="));
ASSIGN = lex.DefineToken(RE.Symbol('='));
PLUS = lex.DefineToken(RE.Symbol('+'));
MINUS = lex.DefineToken(RE.Symbol('-'));
ASTERISK = lex.DefineToken(RE.Symbol('*'));
SLASH = lex.DefineToken(RE.Symbol('/'));
LEFT_PH = lex.DefineToken(RE.Symbol('('));
RIGHT_PH = lex.DefineToken(RE.Symbol(')'));
LEFT_BK = lex.DefineToken(RE.Symbol('['));
RIGHT_BK = lex.DefineToken(RE.Symbol(']'));
LEFT_BR = lex.DefineToken(RE.Symbol('{'));
RIGHT_BR = lex.DefineToken(RE.Symbol('}'));
COMMA = lex.DefineToken(RE.Symbol(','));
COLON = lex.DefineToken(RE.Symbol(':'));
SEMICOLON = lex.DefineToken(RE.Symbol(';'));
DOT = lex.DefineToken(RE.Symbol('.'));
//skips
WHITESPACE = lex.DefineToken(RE_SpaceChar | RE.CharSet("\u0009\u000B\u000C"));
LINE_BREAKER = lex.DefineToken(
RE.CharSet("\u000D\u000A\u0085\u2028\u2029") |
RE.Literal("\r\n")
);
var RE_DelimitedCommentSection = RE.Symbol('/') | (RE.Symbol('*').Many() >> RE_NotSlashOrAsterisk);
COMMENT = lex.DefineToken(
(RE.Literal("//") >> RE_InputChar.Many()) |
(RE.Literal("/*") >> RE_DelimitedCommentSection.Many() >> RE.Symbol('*').Many1() >> RE.Symbol('/')),
"comment");
skippedTokens.Add(WHITESPACE);
skippedTokens.Add(LINE_BREAKER);
skippedTokens.Add(COMMENT);
}
public void MultipleLexerParsingTest()
{
Lexicon lexicon = new Lexicon();
Lexer global = lexicon.Lexer;
Lexer keywords = global.CreateSubLexer();
var PROPERTY = global.DefineToken(RE.Literal("property"));
var ID = global.DefineToken(RE.Range('a', 'z').Concat(
(RE.Range('a', 'z') | RE.Range('0', '9')).Many()), "ID");
var NUM = global.DefineToken(RE.Range('0', '9').Many1(), "NUM");
var EQ = global.DefineToken(RE.Symbol('='));
var SEMICOLON = global.DefineToken(RE.Symbol(';'));
var LB = global.DefineToken(RE.Symbol('{'));
var RB = global.DefineToken(RE.Symbol('}'));
var WHITESPACE = global.DefineToken(RE.Symbol(' ').Union(RE.Symbol('\t')), "[ ]");
var GET = keywords.DefineToken(RE.Literal("get"));
var assignStatement =
from id in ID
from eq in EQ
from value in NUM
from st in SEMICOLON
select id.Value + "=" + value.Value;
var getDef =
from _get in GET
from lb in LB
from statements in assignStatement.Many()
from rb in RB
select new GetDef {
Statements = statements
};
var propDef =
from _prop in PROPERTY
from id in ID
from lb in LB
from getdef in getDef
from rb in RB
select new PropDef {
PropName = id.Value.Content, GetDef = getdef
};
string source = "property get { get { get = 1; } }";
SourceReader sr = new SourceReader(
new StringReader(source));
var info = lexicon.CreateScannerInfo();
Scanner scanner = new Scanner(info);
scanner.SetTriviaTokens(WHITESPACE.Index);
scanner.SetSource(sr);
CompilationErrorManager errorManager = new CompilationErrorManager();
errorManager.DefineError(1, 0, CompilationStage.Parsing, "Unexpected token '{0}'");
errorManager.DefineError(2, 0, CompilationStage.Parsing, "Missing token '{0}'");
errorManager.DefineError(3, 0, CompilationStage.Parsing, "Syntax error");
ProductionInfoManager pim = new ProductionInfoManager(propDef.SuffixedBy(Grammar.Eos()));
LR0Model lr0 = new LR0Model(pim);
lr0.BuildModel();
string dot = lr0.ToString();
TransitionTable tt = TransitionTable.Create(lr0, info);
SyntaxErrors errDef = new SyntaxErrors()
{
TokenUnexpectedId = 1, TokenMissingId = 2, OtherErrorId = 3
};
ParserEngine driver = new ParserEngine(tt, errDef);
Lexeme r;
do
{
r = scanner.Read();
driver.Input(r);
} while (!r.IsEndOfStream);
var el = errorManager.CreateErrorList();
var result = (PropDef)driver.GetResult(0, el);
Assert.AreEqual(0, el.Count);
Assert.AreEqual("get", result.PropName);
Assert.AreEqual("get=1", result.GetDef.Statements.First());
}
