Minishell/lexer.c

/* ************************************************************************** */
/*                                                                            */
/*                                                        :::      ::::::::   */
/*   lexer.c                                            :+:      :+:    :+:   */
/*                                                    +:+ +:+         +:+     */
/*   By: mea <marvin@42.fr>                         +#+  +:+       +#+        */
/*                                                +#+#+#+#+#+   +#+           */
/*   Created: 2022/05/02 13:44:57 by narnaud           #+#    #+#             */
/*   Updated: 2022/05/06 11:19:26 by narnaud          ###   ########.fr       */
/*                                                                            */
/* ************************************************************************** */

#include "minishell.h"

int	create_token(t_lexer *lex, char str[])
{
	t_token	*tok;
	t_type	type;

	type = lex->type;
	if (lex->next_type)
		lex->type = lex->next_type;
	else
		lex->type = WORD;
	lex->next_type = WORD;
	if (!str[0])
		return (0);
	tok = lex->tokens;	
	if (!lex->tokens)
		tok = ft_calloc(1, sizeof *tok);
	else
	{
		while (tok->next)
			tok = tok->next;
		tok->next = ft_calloc(1, sizeof *tok);
		tok = tok->next;
	}
	tok->type = type;
	tok->value = strdup(str);
	if (DEBUG)
		printf("token value : %s - token type : %d\n", tok->value, type);
	if (!lex->tokens)
		lex->tokens = tok;
	return (1);
}

int	check_register(t_lexer *lex, char **line, char *tmp)
{
	int	spaces;

	if (lex->state != ROOT_ST)
		return (0);
	spaces = 0;
	if (ft_isspace(**line))
	{
		while (ft_isspace(**line))
			(*line)++;
		spaces = 1;
	}
	if (**line == '>' && (*line)++)
	{
		if (**line == '>' && (*line)++)
			lex->next_type = ADD;
		else
			lex->next_type = OUT;
		return (create_token(lex, tmp));
	}
	else if (**line == '<' && (*line)++)
	{
		if (**line == '<' && (*line)++)
			lex->next_type = HD;
		else
			lex->next_type = IN;
		return (create_token(lex, tmp));
	}
	else if (**line == '|')
	{
		(*line)++;
		lex->next_type = PIPE;
		create_token(lex, tmp);
		return (create_token(lex, "|"));
	}
	if (!spaces || !*tmp)
		return (0);
	return (create_token(lex, tmp));
}

int	replace_var(t_datas *datas, char **line, char *tmp, int tmp_i)
{
	int		name_len;
	int		i;
	char	*var_name;
	char	*value;
	char 	**env;

	i = 1;
	if ((*line)[1] == '?')
	{
		value = ft_itoa(datas->exit_code);
		name_len = 2;
	}
	else
	{
		while (ft_isalpha((*line)[i]) || ft_isdigit((*line)[i]) || (*line)[i] == '_')
			i++;
		var_name = ft_substr(*line, 1, i - 1);
		i = 0;
		while (datas->envp[i])
		{
			env = ft_split(datas->envp[i], '=');
			if (!ft_strncmp(var_name, env[0], ft_strlen(var_name) + 1))
				break ;
			i++;
		}
		if (!datas->envp[i])
			value = strdup("");
		else
			value = env[1];
		name_len = ft_strlen(var_name);
	}
	i = 0;
	while (value[i])
		tmp[tmp_i++] = value[i++];
	tmp[tmp_i] = 0;
	*line += name_len + 1;
	return (tmp_i);
}

int	check_state(t_lexer *lex, char **line)
{
	t_state	new;

	new = OLD_ST;
	if (**line == '"')
	{
		if (lex->state == D_QUOTE_ST)
			new = ROOT_ST;
		else if (lex->state == ROOT_ST)
			new = D_QUOTE_ST;
	}
	else if (**line == '\'')
	{	
		if (lex->state == S_QUOTE_ST)
			new = ROOT_ST;
		else if (lex->state == ROOT_ST)
			new = S_QUOTE_ST;
	}
	if (new)
	{
		(*line)++;
		lex->state = new;
		return (1);
	}
	return (0);
}

t_token	*lexer(t_datas *datas, char *line)
{
	t_lexer	*lex;
	char	*tmp;
	int		tmp_i;

	lex = ft_calloc(1, sizeof *lex);
	lex->state = ROOT_ST;
	tmp = ft_calloc(1024, sizeof *tmp);
	tmp_i = 0;
	while (*line)
	{
		if (check_state(lex, &line))
			continue;
		if (lex->state != S_QUOTE_ST && *line == '$')
			tmp_i = replace_var(datas, &line, tmp, tmp_i);
		if (check_register(lex, &line, tmp))
		{
			tmp_i = 0;
			ft_bzero(tmp, 1024);
			continue ;
		}
		tmp[tmp_i] = *line;
		line++;
		tmp_i++;
	}
	create_token(lex, tmp);
	return (lex->tokens);
}