guillm/42sh
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

feat(lexer): operators done

Browse source
This commit is contained in:
matteo 2026-01-23 19:34:47 +01:00
parent 3cdba0ad9c
commit 1e5593fc8e
3 changed files with 111 additions and 43 deletions
Download patch file Download diff file Expand all files Collapse all files

102
src/lexer/lexer_utils.c
View file

@ -24,24 +24,12 @@ static void set_token_spechar(struct token *tok, char *begin, ssize_t size)
case '\n':
tok->type = TOKEN_NEWLINE;
break;
case '\'':
tok->type = TOKEN_QUOTE;
break;
case '"':
tok->type = TOKEN_DOUBLE_QUOTE;
break;
case '`':
tok->type = TOKEN_GRAVE;
break;
case '#':
tok->type = TOKEN_COMMENT;
break;
case '|':
tok->type = TOKEN_PIPE;
break;
case '&':
tok->type = TOKEN_AMPERSAND;
break;
case '\\':
tok->type = TOKEN_BACKSLASH;
break;
@ -57,12 +45,6 @@ static void set_token_spechar(struct token *tok, char *begin, ssize_t size)
case '}':
tok->type = TOKEN_RIGHT_BRACKET;
break;
case '<':
tok->type = TOKEN_LESS;
break;
case '>':
tok->type = TOKEN_GREATER;
break;
case '*':
tok->type = TOKEN_STAR;
break;
@ -117,6 +99,47 @@ static void set_token_keyword(struct token *tok, char *begin, ssize_t size)
strncpy(tok->data, begin, size);
}
/* @brief: if an operator is found at [begin],
* [tok->token_type] is set accordingly
*/
static void set_token_operator(struct token *tok, char *begin, ssize_t size)
{
if (tok->type != TOKEN_NULL)
return;
if (strncmp(begin, ">", size) == 0)
{
tok->type = TOKEN_REDIR_RIGHT;
}
else if (strncmp(begin, "<", size) == 0)
{
tok->type = TOKEN_REDIR_LEFT;
}
else if (strncmp(begin, ">>", size) == 0)
{
tok->type = TOKEN_REDIR_DOUBLE_RIGHT;
}
else if (strncmp(begin, ">&", size) == 0)
{
tok->type = TOKEN_REDIR_RIGHT_AMP;
}
else if (strncmp(begin, ">|", size) == 0)
{
tok->type = TOKEN_REDIR_RIGHT_PIPE;
}
else if (strncmp(begin, "<&", size) == 0)
{
tok->type = TOKEN_REDIR_LEFT_AMP;
}
else if (strncmp(begin, "<>", size) == 0)
{
tok->type = TOKEN_REDIR_LEFT_RIGHT;
}
else if (strncmp(begin, "|", size) == 0)
{
tok->type = TOKEN_PIPE;
}
}
/* @brief: if token_type has not yet been set, then it is a TOKEN_WORD
* Also allocates the data and fills it.
*/
@ -147,6 +170,23 @@ static void set_token_ION(struct token *tok, char *begin, ssize_t size)
}
}
/* @brief: check if [c] is a delimiter for end of line.
* @return: true if [c] == '\n' or EOF. false otherwise.
*/
static bool is_end_of_line(char c)
{
return c == EOF || c == '\n';
}
bool is_special_char(char c)
{
if (c == EOF)
return true;
char special_chars[] = "\n'\"`;#|&\\(){}<>*";
return strchr(special_chars, c) != NULL;
}
struct token *new_token(char *begin, ssize_t size, bool only_digits)
{
struct token *tok = calloc(1, sizeof(struct token));
@ -156,6 +196,7 @@ struct token *new_token(char *begin, ssize_t size, bool only_digits)
if (only_digits)
set_token_ION(tok, begin, size);
set_token_operator(tok, begin, size);
set_token_spechar(tok, begin, size);
set_token_keyword(tok, begin, size);
set_token_word(tok, begin, size);
@ -204,14 +245,35 @@ void stream_init(struct lexer_context *ctx)
ctx->end_previous_token = trimed_stream;
}
ssize_t len_op_sepchar(char *stream, ssize_t i)
{
if (!is_special_char(stream[i]))
return -1; // should never happen
if (stream[i] != '>' && stream[i] != '<')
return 1; // special character (cannot be operator)
// operator
if (stream[i] == '<')
{
if (stream[i + 1] == '&' || stream[i + 1] == '>')
return 2; // <&, <>
}
else if (stream[i + 1] == '>' || stream[i + 1] == '|'
|| stream[i + 1] == '&')
return 2; // >>, >&, >|
return 1; // >, <
}
void go_end_of_line(struct lexer_context *ctx)
{
if (ctx == NULL || ctx->end_previous_token == NULL)
return;
ssize_t i = 0;
while (ctx->end_previous_token[i] != '\n'
&& ctx->end_previous_token[i] != EOF)
while (!is_end_of_line(ctx->end_previous_token[i]))
{
i++;
}