vm/src/lexer.c

#include <stdlib.h>
#include <stdio.h>
#include <regex.h>

#include "lexer.h"

static regex_t regex(const char* str)
{
	regex_t reg;
	int err = regcomp(&reg, str, REG_EXTENDED);
	if (err)
	{
		fprintf(stderr, "Invalid regex: '%s'\n", str);
		fprintf(stderr, "Error code: %i\n", err);
		exit(1);
	}

	return reg;
}

static int regmatch(regex_t reg, char* str, regmatch_t* pmatch)
{
	return (regexec(&reg, str, 1, pmatch, 0) != REG_NOMATCH);
}

static int tokens_append(
		lexer_tokens* tokens,
		lexer_token_class tokenClass,
		char* str,
		size_t len)
{
	tokens->length += 1;
	if (tokens->length > tokens->allocd)
	{
		tokens->allocd *= 2;
		tokens->pairs = realloc(tokens->pairs, tokens->allocd);
		if (!tokens->pairs)
			return 1;
	}

	lexer_token pair = tokens->pairs[tokens->length - 1];
	pair.tokenClass = tokenClass;
	pair.str = str;
	pair.len = len;

	return 0;
}

lexer_tokens* lexer_analyze(char* str)
{
	regex_t whitespace = regex("\\s+");
	regex_t identifier = regex("[a-zA-Z][a-zA-Z0-9]*");
	regex_t keyword = regex("if|ret|func|str|arr|err|null");
	regex_t operator = regex("=|==");
	regex_t separator = regex("\\,");

	regex_t integer = regex("[0-9]+");
	regex_t string = regex("\\\"[^\\\"]*\\\"");

	regex_t function_start = regex("\\{");
	regex_t function_end = regex("\\}");
	regex_t expression_start = regex("\\(");
	regex_t expression_end = regex("\\)");

	lexer_tokens* tokens = malloc(sizeof(lexer_tokens));
	if (!tokens)
		return NULL;

	tokens->pairs = NULL;
	tokens->length = 0;
	tokens->allocd = 0;

	regmatch_t pmatch;
	size_t offset = 0;

	#define APPEND(tclass) tokens_append(tokens, tclass, str + offset + pmatch.rm_so, pmatch.rm_eo - pmatch.rm_so)

	while (1)
	{
		if (regmatch(whitespace, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_WHITESPACE))
				return NULL;
		}
		else if (regmatch(identifier, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_IDENTIFIER))
				return NULL;
		}
		else if (regmatch(keyword, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_KEYWORD))
				return NULL;
		} else if (regmatch(operator, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_OPERATOR))
				return NULL;
		}
		else if (regmatch(separator, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_SEPARATOR))
				return NULL;
		}
		else if (regmatch(integer, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_INTEGER))
				return NULL;
		}
		else if (regmatch(string, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_STRING))
				return NULL;
		}
		else if (regmatch(function_start, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_FUNCTION_START))
				return NULL;
		}
		else if (regmatch(function_end, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_FUNCTION_END))
				return NULL;
		}
		else if (regmatch(expression_start, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_EXPRESSION_START))
				return NULL;
		}
		else if (regmatch(expression_end, str + offset, &pmatch))
		{
			if (APPEND(LEXER_TOKEN_EXPRESSION_END))
				return NULL;
		}
		else
		{
			return tokens;
		}

		offset += pmatch.rm_eo;
	}
}