public void LexerTest()
{
var lexer = new Lexer ();
lexer.AddDefinition(new TokenDefinition(ETokenType.Invalid,
new Regex(@"\*|\/|\+|\-")));
lexer.AddDefinition(new TokenDefinition(
ETokenType.Invalid,
new Regex(@"\d+")));
lexer.AddDefinition(new TokenDefinition(
ETokenType.WhiteSpace,
new Regex(@"\s+"),
true));
var tokens = lexer.Tokenize("1 * 2 / 3 + 4 - 5");
Assert.AreEqual (17,17);
}
/// Author: Max Hamulyak
/// Date: 20-06-2015
/// <summary>
/// Build the lexer needed to parse a string, based on a programming language language rule.
/// </summary>
/// <returns>The language lexer.</returns>
protected override ILexer BuildLanguageLexer()
{
ILexer lexer = new Lexer ();
Regex endOfLineRegex = new Regex(@"\r\n|\r|\n", RegexOptions.Compiled);
lexer.AddDefinition(new TokenDefinition(ETokenType.EOL,endOfLineRegex));
lexer.AddDefinition(new TokenDefinition(
ETokenType.WhiteSpace,
new Regex(@"[ \t]")));
lexer.AddDefinition(new TokenDefinition(ETokenType.FUNCTIONDeclaration, new Regex(@"def[ \s](?<functionName>[a-z]+[A-Za-z0-9]*)\(\)", RegexOptions.Multiline)));
lexer.AddDefinition (new TokenDefinition (ETokenType.Command,
new Regex (@"(moveForward\(\)|rotateRight\(\)|pickUp\('([^)]+)\))")));
lexer.AddDefinition(new TokenDefinition(ETokenType.FUNCTIONCall, new Regex(@"(?<functionName>[a-z]+[A-Za-z0-9]*)\(\)", RegexOptions.Multiline)));
lexer.AddDefinition (new TokenDefinition(ETokenType.PythonRange, new Regex(@"range\(((\d+),(\d+))\)")));
lexer.AddDefinition (new TokenDefinition (ETokenType.KeyWord, new Regex (@"(in)")));
Regex commentRegex = new Regex (@"\#(.*)", RegexOptions.Multiline);
lexer.AddDefinition(new TokenDefinition(ETokenType.CommentLine,commentRegex));
lexer.AddDefinition (new TokenDefinition (ETokenType.Invalid, new Regex (@"(moveBackward\(\)|rotateLeft\(\))")));
#region PythonBlockSyntax
forLoopRegex = new Regex(@"for[ \t](?<loopIndex>[a-z]{1}[a-zA-Z]*)[ \t]*[ \t]*in[ \t]*range\((?<startingLow>[0-9]+),(?<endingHigh>[0-9]+)\)", RegexOptions.Multiline);
lexer.AddDefinition(new TokenDefinition(ETokenType.FOR,forLoopRegex));
lexer.AddDefinition(new TokenDefinition (ETokenType.WHILE,
new Regex (@"\bwhile\b", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition (ETokenType.FOR,
new Regex (@"\bfor\b", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition (ETokenType.IF,new Regex (@"\bif\b", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition (ETokenType.ELSEIF,
new Regex (@"\belif\b", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition (ETokenType.ELSE,
new Regex (@"\belse\b", RegexOptions.Multiline)));
// lexer.AddDefinition(new TokenDefinition(ETokenType.FUNCTIONDeclaration, new Regex(@"(?:def\s)[a-z]{3,10}\(\)", RegexOptions.Multiline)));
// lexer.AddDefinition(new TokenDefinition(ETokenType.FUNCTIONCall, new Regex(@"[a-z]{3,10}\(\)", RegexOptions.Compiled)));
#endregion
lexer.AddDefinition (new TokenDefinition (ETokenType.LogicalOperator,
new Regex (@"(or|and|not)")));
lexer.AddDefinition(new TokenDefinition (ETokenType.startBlock,
new Regex (@"(:)")));
lexer.AddDefinition (new TokenDefinition (ETokenType.logicInstruction,
new Regex (@"(left|right|forward|backward|at\('([^)]+)\))")));
lexer.AddDefinition (new TokenDefinition (ETokenType.leftParentheses,
new Regex (@"[\(]")));
lexer.AddDefinition (new TokenDefinition (ETokenType.rightParentheses,
new Regex (@"[\)]")));
lexer.AddDefinition (new TokenDefinition (ETokenType.Operator, new Regex (@"(\+|\-|\*|\/)")));
lexer.AddDefinition (new TokenDefinition (ETokenType.ComparisonOperator, new Regex (@"(\<=|/>=|\<|\>|==|\!=)")));
lexer.AddDefinition( new TokenDefinition (ETokenType.Literal, new Regex(@"(True|False)")));
lexer.AddDefinition(new TokenDefinition(ETokenType.VARIABLE,new Regex(@"([A-Za-z]+)")));
lexer.AddDefinition (new TokenDefinition (ETokenType.AssignmentOperator, new Regex ("(=)")));
lexer.AddDefinition( new TokenDefinition (ETokenType.Literal, new Regex(@"(\d+)")));
//"[^\"]*\"
lexer.AddDefinition( new TokenDefinition (ETokenType.Literal, new Regex(@"\""[^\""]*\""")));
lexer.AddDefinition (new TokenDefinition (ETokenType.KeyWord,
new Regex (@"\b(" + string.Join ("|", languageKeywords.Select (Regex.Escape).ToArray ()) + @"\b)", RegexOptions.Multiline)));
return lexer;
}
/// Author: Max Hamulyak
/// Date: 20-06-2015
/// <summary>
/// Build the lexer needed to parse a string, based on a programming language language rule.
/// </summary>
/// <returns>The language lexer.</returns>
protected override ILexer BuildLanguageLexer()
{
ILexer lexer = new Lexer ();
forLoopRegex = new Regex (@"for[ \t](?<loopIndex>[a-z]{1}[a-zA-Z]+)[ \t]*\:\=[ \t]*(?<startingLow>[0-9])+[ \t](?<direction>to|downto)[ \t](?<endingHigh>[0-9])+[ \t]do", RegexOptions.Multiline);
varDeclarationRegex = new Regex (@"^var[ \s](?<variableName>[A-Za-z]+)\:[ \t](?<variableType>integer|string|boolean);$", RegexOptions.Multiline);
lexer.AddDefinition (new TokenDefinition (ETokenType.AssignmentOperator, new Regex ("(:=)")));
lexer.AddDefinition(new TokenDefinition(ETokenType.startBlock,new Regex(";|:")));
Regex commentRegex = new Regex (@"\{(.*)\}", RegexOptions.Multiline);
lexer.AddDefinition(new TokenDefinition(ETokenType.CommentLine,commentRegex));
lexer.AddDefinition(new TokenDefinition (ETokenType.ELSEIF,
new Regex (@"(\belse[ \t]if\b)", RegexOptions.Multiline)));
Regex endOfLineRegex = new Regex(@"\r\n|\r|\n", RegexOptions.Compiled);
lexer.AddDefinition(new TokenDefinition(ETokenType.EOL,endOfLineRegex));
lexer.AddDefinition(new TokenDefinition(
ETokenType.WhiteSpace,
new Regex(@"[ \t]")));
lexer.AddDefinition (new TokenDefinition (ETokenType.Invalid, new Regex (@"([#{}])", RegexOptions.Multiline)));
lexer.AddDefinition (new TokenDefinition (ETokenType.endBlock, new Regex (@"(\bend\b;|\bend\b.|\bend\b)", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition (ETokenType.WHILE,
new Regex (@"\bwhile\b", RegexOptions.Multiline)));
//lexer.AddDefinition(new TokenDefinition (ETokenType.FOR,
// new Regex (@"\bfor\b", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition(ETokenType.FOR,forLoopRegex));
lexer.AddDefinition (new TokenDefinition (ETokenType.FOR, new Regex (@"\bfor\b", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition (ETokenType.IF,
new Regex (@"\bif\b", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition (ETokenType.ELSE,
new Regex (@"\belse\b", RegexOptions.Multiline)));
lexer.AddDefinition (new TokenDefinition (ETokenType.VARIABLE, varDeclarationRegex));
lexer.AddDefinition (new TokenDefinition (ETokenType.Command,
new Regex (@"(moveForward\(\)|rotateRight\(\)|pickUp\('([^)]+)\))")));
lexer.AddDefinition(new TokenDefinition(ETokenType.FUNCTIONDeclaration, new Regex(@"procedure[ \s](?<functionName>[a-z]+[A-Za-z0-9]*)\(\)", RegexOptions.Multiline)));
lexer.AddDefinition(new TokenDefinition(ETokenType.FUNCTIONCall, new Regex(@"(?<functionName>[a-z]+[A-Za-z0-9]*)\(\)", RegexOptions.Multiline)));
lexer.AddDefinition (new TokenDefinition (ETokenType.logicInstruction,
new Regex (@"(left|right|forward|backward|at\('([^)]+)\))")));
lexer.AddDefinition(new TokenDefinition (ETokenType.startBlock,
new Regex (@"\bbegin\b", RegexOptions.Multiline)));
lexer.AddDefinition (new TokenDefinition (ETokenType.LogicalOperator,
new Regex (@"(or|and|not)")));
lexer.AddDefinition (new TokenDefinition (ETokenType.KeyWord,
new Regex (@"\b(" + string.Join ("|", languageKeywords.Select (Regex.Escape).ToArray ()) + @"\b)", RegexOptions.Multiline)));
lexer.AddDefinition (new TokenDefinition (ETokenType.leftParentheses,
new Regex (@"\(", RegexOptions.Multiline)));
lexer.AddDefinition (new TokenDefinition (ETokenType.rightParentheses,
new Regex (@"[\)]")));
lexer.AddDefinition (new TokenDefinition (ETokenType.Operator, new Regex (@"(\+|\-|\*|\/)")));
lexer.AddDefinition (new TokenDefinition (ETokenType.ComparisonOperator, new Regex (@"(\<=|/>=|\<|\>|==|\!=)")));
lexer.AddDefinition(new TokenDefinition(ETokenType.VARIABLE,new Regex(@"([A-Za-z]+)")));
//lexer.AddDefinition(new TokenDefinition(ETokenType.VARIABLE,new Regex(@"(^[a-zA-Z]+$)", RegexOptions.Multiline)));
lexer.AddDefinition( new TokenDefinition (ETokenType.Literal, new Regex(@"(\d+)")));
lexer.AddDefinition( new TokenDefinition (ETokenType.Literal, new Regex(@"(True|False)")));
//"[^\"]*\"
lexer.AddDefinition( new TokenDefinition (ETokenType.Literal, new Regex(@"\""[^\""]*\""")));
lexer.AddDefinition (new TokenDefinition (ETokenType.DataType, new Regex (@"(integer|string|boolean)")));
return lexer;
}
