floats

3 files changed, 53 insertions, 25 deletions

diff --git a/lexer.c b/lexer.c
index a4ffd89..f472acc 100644
--- a/lexer.c
+++ b/lexer.c
@@ -54,6 +54,8 @@ bool lexer_peek(struct token* p_token) {
 #define is_alpha(c) (is_lower_alpha(c) || is_upper_alpha(c))
 #define is_numeric(c) ('0' <= c && c <= '9')
 #define is_alphanumeric(c) (is_alpha(c) || is_numeric(c))
+#define is_hexadecimal(c) \
+    (is_numeric(c) || ('a' <= c && c <= 'f') || ('A' <= c && c <= 'F'))
 #define is_ident_legal(c) (is_alphanumeric(c) || c == '_' || c == '$')
 
 static int consume_char() {
@@ -82,39 +84,55 @@ static void lex_ident(struct token* p_token, char ic) {
     };
 }
 
+static unsigned char digit_val(int c, unsigned char base) {
+    unsigned char c_val;
+    if (is_numeric(c)) c_val = c - '0';
+    else if (is_lower_alpha(c)) c_val = c - 'a' + 10;
+    else c_val = c - 'A' + 10;
+
+    if (c_val >= base)
+        LEXER_PANIC("invalid digit in base %hhu: %c", base, c);
+    return c_val;
+}
+
 static void lex_float_lit(
     struct token* p_token,
     unsigned char base,
-    double iv
+    float_lit_t iv
 ) {
-    LEXER_PANIC("floating point literals are not implemented");
+    if (consume_char() != '.')
+        LEXER_PANIC("sanity error, float literal without decimal point");
+    float_lit_t exp = 1.0;
+    while (is_hexadecimal(lookahead)) {
+        int c = consume_char();
+        exp /= base;
+        iv += digit_val(c, base) * exp;
+    }
+
+    *p_token = (struct token) {
+        .type = FLOAT_LIT,
+        .data.float_lit = iv,
+    };
 }
 
 static void lex_int_lit(struct token* p_token, int_lit_t iv) {
     unsigned char base = 10;
 
-    /* TODO: exponentiation, 2e10 f.e. */
     if (iv == 0) {
         if (lookahead == 'x' || lookahead == 'X'
                 || lookahead == 'b' || lookahead == 'B') {
             base = (lookahead == 'x' || lookahead == 'X') ? 16 : 2;
             int suffix = consume_char();
-            if (!is_alphanumeric(lookahead))
+            if (!is_hexadecimal(lookahead))
                 LEXER_PANIC("invalid suffix on integer constant: %c", suffix);
-        } else base = 8;
+        } else if (is_hexadecimal(lookahead)) base = 8;
     }
 
-    while (is_alphanumeric(lookahead)) {
+    while (is_hexadecimal(lookahead)) {
         int c = consume_char();
-        int_lit_t c_val;
-
-        if (is_numeric(c)) c_val = c - '0';
-        else if (is_lower_alpha(c)) c_val = c - 'a' + 10;
-        else c_val = c - 'A' + 10;
-
-        if (c_val >= base)
-            LEXER_PANIC("invalid digit in base %hhu: %c", base, c);
-
+        unsigned char c_val = digit_val(c, base);
+        if (base < 15 && (c == 'e' || c == 'E'))
+            LEXER_PANIC("exponentiation is not implemented");
         if (ckd_mul(&iv, iv, base))
             LEXER_PANIC("integer literal will overflow");
         if (ckd_add(&iv, iv, c_val))
@@ -122,8 +140,7 @@ static void lex_int_lit(struct token* p_token, int_lit_t iv) {
     }
 
     if (lookahead == '.') {
-        consume_char();
-        lex_float_lit(p_token, base, iv);
+        lex_float_lit(p_token, base, (float_lit_t) iv);
         return;
     }
 
@@ -134,13 +151,21 @@ static void lex_int_lit(struct token* p_token, int_lit_t iv) {
 }
 
 static char replace_escape_sequence(char c) {
-    if (c == '\'') return '\'';
-    else if (c == '\"') return '\"';
-    else if (c == '\\') return '\\';
-    else if (c == 'r') return '\r';
-    else if (c == 'n') return '\n';
-    else if (c == 't') return '\t';
-    else LEXER_PANIC("escape sequence not implemented");
+    switch (c) {
+        case '\'': return '\'';
+        case '\"': return '\"';
+        case '\\': return '\\';
+        case '?': return '?';
+        case 'r': return '\r';
+        case 'n': return '\n';
+        case 't': return '\t';
+        case 'v': return '\v';
+        case 'a': return '\a';
+        case 'b': return '\b';
+        case 'f': return '\f';
+        /* TODO: numeric escape sequences, e.g. \xff */
+        default: LEXER_PANIC("escape sequence not implemented");
+    }
 }
 
 static void lex_char_lit(struct token* p_token) {
diff --git a/main.c b/main.c
index d2a6ef5..6fc75e9 100644
--- a/main.c
+++ b/main.c
@@ -4,7 +4,7 @@
 
 int main(int argc, char** argv) {
     if (argc < 2) {
-        fprintf(stderr, "ccc: no input files");
+        fprintf(stderr, "ccc: no input files\n");
         return 1;
     }
 
@@ -12,6 +12,7 @@ int main(int argc, char** argv) {
     for (int i = 1; i < argc; i++) {
         lexer_load(argv[i]);
         while (lexer_pop(&token)) {
+            printf("[%s]: ", argv[i]);
             switch (token.type) {
                 case IDENTIFIER:
                     printf("got identifier: %s\n", token.data.identifier);
diff --git a/makefile b/makefile
new file mode 100644
index 0000000..9cab3aa
--- /dev/null
+++ b/makefile
@@ -0,0 +1,2 @@
+all:
+	gcc -std=c23 -g -O0 -Wall -Werror *.c -o ccc