Latest stuff, rewriting lexer

code_compiler/source/lexer.hpp

@@ -1,111 +1,60 @@
+/// AUTOGENERATED COPYRIGHT HEADER START
+// Copyright (C) 2017-2024 Michael Fabian 'Xaymar' Dirks <info@xaymar.com>
+// AUTOGENERATED COPYRIGHT HEADER END
 #pragma once
+#include <filesystem>
+#include <fstream>
 #include <inttypes.h>
+#include <iostream>
 #include <istream>
 #include <memory>
 #include <string>
 #include <utility>
 
-namespace blitz {	
-	class Lexer {
-		public:
-		enum class Token : uint64_t {
-			TokenUnknown,
-			TokenEOF,
-			TokenNewLine,
+// ToDo:
+// - Figure out a way to let the lexer output line and character information?
 
-			// Symbols
-			TokenPlus,
-			TokenMinus,
-			TokenSlashForward,
-			TokenSlashBackward,
-			TokenMultiply,
-			TokenEqual,
-			TokenOctothorp,
-			TokenPercent,
-			TokenDollar,
-			TokenRoundBracketOpen,
-			TokenRoundBracketClose,
-			TokenSquareBracketOpen,
-			TokenSquareBracketClose,
-			TokenAngleBracketOpen,
-			TokenAngleBracketClose,
-			TokenDot,
-			TokenColon,
-			TokenComma,
-			TokenSemicolon,
-			TokenCaret,
-			TokenBitNot /*~*/,
+namespace blitz {
+	struct token {
+		uint64_t    line;
+		uint64_t    character;
 
-			// String Delimiter
-			TokenDoubleQuote,
-
-			// Types
-			TokenText,
-			TokenNumber,
-			TokenDecimal,
-			TokenQuotedText, // Text encapsulated by TokenDoubleQuote
-			TokenComment,
-
-			// Binary
-			TokenNot,
-			TokenAnd, TokenOr, TokenXor,
-			TokenShl, TokenShr,
-			TokenSal, TokenSar,
-			TokenFalse, TokenTrue,
-
-			// Conversion
-			TokenFloat,
-			TokenString, TokenHex,
-			TokenInt,
-
-			// Control
-			TokenIf, TokenThen, TokenElseIf, TokenElse, TokenEndIf,
-			TokenSelect, TokenCase, TokenDefault, // End Select = TokenEnd, TokenSelect.
-			TokenGoto, TokenGosub,
-			TokenReturn,
-			TokenFunction, // End Function = TokenEnd, TokenFunction.
-			TokenEnd,
-			TokenStop /* DEBUGGER! Ignore in Release mode. */,
-
-			// Loop
-			TokenFor, TokenTo, TokenNext,
-			TokenWhile, TokenWend,
-			TokenRepeat, TokenUntil, TokenForever,
-			TokenExit,
-
-			// Math
-			TokenAbs, TokenSign /*Sgn*/,
-			TokenCos, TokenSin, TokenTan,
-			TokenACos, TokenASin, TokenATan, TokenATan2,
-			TokenLog, TokenLog10,
-			TokenCeil, TokenFloor,
-			TokenMod,
-			TokenPi,
-			TokenExp, TokenSqr,
-
-			// Variables
-			TokenConst,
-			TokenGlobal,
-			TokenLocal,
-
-			// Including files.
-			TokenInclude,
-		};
-
-		public:
-		Lexer();
-		~Lexer();
-
-		std::pair<Token, std::string> GetCurrentToken();
-		std::pair<Token, std::string> GetNextToken(std::shared_ptr<std::istream> fs);
-		
-		private:
-		blitz::Lexer::Token ConvertTextToToken(Token in, std::string text);
-
-		private:
-		Token m_currentToken = Token::TokenUnknown;
-		std::string m_currentText = "";
-		Token m_overrideToken = Token::TokenUnknown;
-		std::string m_overrideText = "";
+		std::string text;
+		enum class variant : uint64_t {
+			UNKNOWN, // We have absolutely no fucking clue.
+			CONTROL, // All kinds of control signals, like NewLine, EndOfFile, ...
+			SYMBOL, // All kinds of symbols.
+			TEXT, // HelloWorld
+			NUMBER, // 1, 1%
+			DECIMAL, // 1.0, 1#
+			STRING, // "HelloWorld"
+			COMMENT, // ; Whatever
+		} type;
 	};
-}
+
+	class lexer {
+		std::filesystem::path _file;
+		std::ifstream         _stream;
+
+		// Current location in the file.
+		uint64_t _line;
+		uint64_t _character;
+
+		blitz::token _current;
+		blitz::token _override;
+
+		public:
+		~lexer();
+		lexer(std::filesystem::path file);
+
+		/** Retrieve the current token information.
+		 */
+		blitz::token current();
+
+		/** Retrieve the next token in the given stream.
+		 *
+		 * This will replace the current token.
+		 */
+		blitz::token next();
+	};
+} // namespace blitz