VMBL/src/sylt/lexer.c

#include "lexer.h"
#include "exception.h"
#include <string.h>
#include <ctype.h>
#include <stdlib.h>
#include <stdio.h>

Lexer initLexer(char *source) {
	Lexer newLexer = {
		.source = source,
		.position = -1,
		.readPosition = 0,
		.lineNumber = 1,
		.currentChar = '\0'
	};
	readChar(&newLexer);
	return newLexer;
}

void readChar(Lexer *lexer) {
	if (lexer->readPosition >= strlen(lexer->source))
		lexer->currentChar = '\0';
	else
		lexer->currentChar = lexer->source[lexer->readPosition];

	lexer->position = lexer->readPosition;
	lexer->readPosition++;
}

void skipWhitespace(Lexer *lexer) {
	bool whitespace = true;

	while (whitespace)
	{
		switch (lexer->currentChar)
		{
		case ' ':
		case '\t':
		case '\n':
		case '\r':
			if (lexer->currentChar == '\n')
				lexer->lineNumber++;		

			readChar(lexer);
			break;
		
		default:
			whitespace = false;
			break;
		}
	}
	
	
	
}

Token readNumber(Lexer *lexer) {
	int startPos = lexer->position;
	int dotCount = 0;
	int strLength = 0;

	size_t outputCap = 8;
	char *output = malloc(outputCap);

	if (!output) {
		exceptionMessage(EXCEPTION_MEMORY_ALLOCATION_FAILURE, "Lexer failed to allocate memory.", lexer->lineNumber, lexer->position);
		exit(1);
	}

	while (isdigit(lexer->currentChar) || lexer->currentChar == '.')
	{
		if (lexer->currentChar == '.') {
			dotCount++;

			if (dotCount > 1) {
				exceptionMessage(EXCEPTION_MALFORMED_NUMBER, "Too many dots in decimal.", lexer->lineNumber, lexer->position);

				char buffer[strLength+1];
				strncpy(buffer, lexer->source + startPos, strLength+1);
				buffer[strLength+1] = 0;

				return NEW_TOKEN(lexer, TOKEN_ILLEGAL, buffer);
			}
		}
		
		// allocate more memory if we reach the end of our buffer
		if (strLength + 1 >= outputCap) {
			char *temp = realloc(output, outputCap*=2);

			if (!temp) {
				exceptionMessage(EXCEPTION_MEMORY_ALLOCATION_FAILURE, "Lexer failed to allocate memory.", lexer->lineNumber, lexer->position);
				exit(1);
			}

			output = temp;
		}

		output[strLength] = lexer->source[lexer->position];
		strLength++;
		readChar(lexer);		

		if (lexer->currentChar == '\0')
			break;
	}

	output[strLength] = 0;
	
	if (dotCount == 0)
		return NEW_TOKEN(lexer, TOKEN_INT, output);
	else
		return NEW_TOKEN(lexer, TOKEN_FLOAT, output);
}

Token nextToken(Lexer *lexer) {
	Token tok;
	skipWhitespace(lexer);

	switch (lexer->currentChar)
	{
	case '+':
		tok = NEW_TOKEN(lexer, TOKEN_PLUS, "+");
		break;
	case '-':
		tok =  NEW_TOKEN(lexer, TOKEN_MINUS, "-");
		break;
	case '*':
		tok =  NEW_TOKEN(lexer, TOKEN_ASTERISK, "*");
		break;
	case '/':
		tok =  NEW_TOKEN(lexer, TOKEN_SLASH, "/");
		break;
	case '%':
		tok =  NEW_TOKEN(lexer, TOKEN_MODULUS, "%");
		break;
	case '^':
		tok =  NEW_TOKEN(lexer, TOKEN_POW, "^");
		break;
	case '(':
		tok =  NEW_TOKEN(lexer, TOKEN_LPAREN, "(");
		break;
	case ')':
		tok =  NEW_TOKEN(lexer, TOKEN_RPAREN, ")");
		break;

	case '\0': // EOF
		tok = NEW_TOKEN(lexer, TOKEN_EOF, "EOF");
		break;
	
	default:
		if (isdigit(lexer->currentChar)) {
			tok = readNumber(lexer);
		} else {
			tok = NEW_TOKEN(lexer, TOKEN_ILLEGAL, &lexer->currentChar);
		}

		break;
	}

	readChar(lexer);
	return tok;
}