usefull commit message

1 files changed, 174 insertions, 0 deletions

diff --git a/src/lexer.c b/src/lexer.c
new file mode 100644
index 0000000..407be25
--- /dev/null
+++ b/src/lexer.c
@@ -0,0 +1,174 @@
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+
+#include "common.h"
+#include "lexer.h"
+
+#define CH(lexer) (lexer)->str[(lexer)->str_idx]
+#define TOKEN_SEPARATOR_TABLE(X, l)                 \
+    X(('(' == CH(l)), on_separator(l, TOKEN_LP))    \
+    X((')' == CH(l)), on_separator(l, TOKEN_RP))    \
+    X(isspace(CH(l)), on_separator(l, TOKEN_NONE))  \
+    X(TABLE_ELSE,     acc_add_char(l, CH(l)))
+
+#define SET_TYPE(lexer, ttype) (lexer)->token.type = (ttype)
+#define SET_VALUE(lexer, member, tvalue) (lexer)->token.value.member = (tvalue)
+#define TOKEN_VALUE_TABLE(X, l)              \
+    X(is_int(l->acc, &l->token.value.num),   \
+        SET_TYPE(l, TOKEN_INT))              \
+    X(is_special(l->acc, &l->token.type),    \
+        ;)                                   \
+    X(TABLE_ELSE,                            \
+        SET_TYPE(l, TOKEN_ID);               \
+        SET_VALUE(l, id, l->acc))
+
+#define TOKEN_SPECIALS_TABLE(X)      \
+    X(TOKEN_DOT,        ".")         \
+    X(TOKEN_QUOTE,      "'")         \
+    X(TOKEN_UNQUOTE,    ",")         \
+    X(TOKEN_LAMBDA,     "lambda")    \
+    X(TOKEN_DEFINE,     "define")    \
+    X(TOKEN_QUOTE_FORM, "quote")
+
+#define TOKEN_VALUE_STRING_TABLE(X, tvalue)  \
+    X(TOKEN_LP,      "(")                    \
+    X(TOKEN_RP,      ")")                    \
+    X(TOKEN_ID,      "%s", tvalue.id)        \
+    X(TOKEN_STR,     "%s", tvalue.str)       \
+    X(TOKEN_INT,     "%d", tvalue.num)       \
+    TOKEN_SPECIALS_TABLE(X)                  \
+    X(TOKEN_NONE,    "(none)")
+
+#define TABLE_ELSE 1
+#define TABLE_END {}
+
+#define CALLBACK(test, callback)                \
+    if(test) {                                  \
+        ERR_NZ(callback, r, return r);          \
+    } else
+#define CALLBACK_BLIND(test, callback)          \
+    if(test) {                                  \
+        callback;                               \
+    } else
+
+#define LEN(arr) (sizeof(arr)/sizeof(*(arr)))
+
+const char * const token_type_string[] = {
+    TOKEN_TYPES(TO_STRING)
+};
+
+static int on_separator(lexer_t lexer, enum token_type type);
+static int acc_add_char(lexer_t lexer, char ch);
+static int acc_empty(lexer_t lexer);
+
+static int is_int(char *str, int *num);
+static int is_special(char *str, enum token_type *type);
+
+lexer_t lexer_create(FILE *fp)
+{
+    lexer_t lexer = malloc(sizeof(*lexer));
+    lexer->fp = fp;
+    lexer->line = 0;
+    lexer->str_idx = 0;
+    lexer->acc_idx = 0;
+    memset(lexer->acc, 0, sizeof(lexer->acc));
+    memset(lexer->str, 0, sizeof(lexer->str));
+       
+    lexer->token.type = TOKEN_NONE;
+
+    return lexer;
+}
+
+void lexer_destroy(lexer_t lexer)
+{
+    if(!lexer) return;
+    free(lexer);
+}
+
+int lexer_token_next(lexer_t lexer, struct token *token)
+{
+    if(lexer->acc_idx == 0 && lexer->acc[0] != '\0') {
+        memset(lexer->acc, 0, sizeof(lexer->acc));
+    }
+            
+    while(lexer->token.type == TOKEN_NONE)
+    {
+        if(lexer->str[lexer->str_idx] == '\0') {
+            ERR_Z(fgets(lexer->str, LEN(lexer->str), lexer->fp), return -EIO);
+            lexer->str_idx = 0;
+            lexer->line++;
+        }
+        
+        TOKEN_SEPARATOR_TABLE(CALLBACK, lexer) TABLE_END;
+    }
+    
+    *token = lexer->token;
+    lexer->token.type = TOKEN_NONE;
+    return 0;
+}
+
+int token_value_string(struct token *token, size_t buf_sz, char *buf)
+{
+#define AS_STRING(ttype, ...) \
+    case ttype: return snprintf(buf, buf_sz, __VA_ARGS__);
+        
+    switch(token->type) {
+        TOKEN_VALUE_STRING_TABLE(AS_STRING, token->value);
+    }
+
+    return 0;
+}
+
+static int on_separator(lexer_t lexer, enum token_type type)
+{
+    if(lexer->acc_idx > 0) return acc_empty(lexer);
+
+    lexer->token.type = type;
+    lexer->str_idx++;
+    return 0;
+}
+
+static int acc_add_char(lexer_t lexer, char ch)
+{
+    if(lexer->acc_idx >= LEN(lexer->acc) - 1) {
+        return -ENAMETOOLONG;
+    }
+    
+    lexer->acc[lexer->acc_idx++] = ch;
+    lexer->str_idx++;
+
+    return 0;
+}
+
+static int acc_empty(lexer_t lexer)
+{    
+    TOKEN_VALUE_TABLE(CALLBACK_BLIND, lexer) TABLE_END;
+    lexer->acc_idx = 0;
+    
+    return 0;
+}
+
+static int is_int(char *str, int *num)
+{
+    char *endptr = str;
+    long _num = strtol(str, &endptr, 10);
+
+    if(*endptr != '\0') return 0;
+
+    *num = (int)_num;
+    return 1;
+}
+
+static int is_special(char *str, enum token_type *type)
+{
+#define IS_SPECIAL(ttype, sstr)                  \
+    if(strcmp(sstr, str) == 0) {                 \
+        *type = ttype;                           \
+        return 1;                                \
+    } else
+
+    TOKEN_SPECIALS_TABLE(IS_SPECIAL) TABLE_END;
+    
+    return 0;
+}