X-Git-Url: http://nsz.repo.hu/git/?a=blobdiff_plain;f=lexer.c;h=5ca734a9256a171d0ac75c1898de04d0b8dc96c2;hb=43298905d2d0a8eabe545794d056bbd7a104b9bf;hp=cba071b07289f0ebe3f86dac35d846c7ed4411e4;hpb=70abad8ade78c811d76cc79741fccfc6bd7d58ab;p=cparser

diff --git a/lexer.c b/lexer.c
index cba071b..5ca734a 100644
--- a/lexer.c
+++ b/lexer.c
@@ -56,6 +56,7 @@ static char        buf[1024 + MAX_PUTBACK];
 static const char *bufend;
 static const char *bufpos;
 static strset_t    stringset;
+bool               allow_dollar_in_symbol = true;
 
 /**
  * Prints a parse error message at the current token.
@@ -81,6 +82,11 @@ static inline void next_real_char(void)
 {
 	assert(bufpos <= bufend);
 	if (bufpos >= bufend) {
+		if (input == NULL) {
+			c = EOF;
+			return;
+		}
+
 		size_t s = fread(buf + MAX_PUTBACK, 1, sizeof(buf) - MAX_PUTBACK,
 		                 input);
 		if(s == 0) {
@@ -189,6 +195,7 @@ end_of_next_char:;
 }
 
 #define SYMBOL_CHARS  \
+	case '$': if (!allow_dollar_in_symbol) goto dollar_sign; \
 	case 'a':         \
 	case 'b':         \
 	case 'c':         \
@@ -276,6 +283,7 @@ static void parse_symbol(void)
 			break;
 
 		default:
+dollar_sign:
 			goto end_symbol;
 		}
 	}
@@ -343,7 +351,7 @@ static void parse_integer_suffix(bool is_oct_hex)
 			if(v >= TARGET_LONG_MIN && v <= TARGET_LONG_MAX) {
 				lexer_token.datatype = type_long;
 				return;
-			} else if(is_oct_hex && v >= 0 && v <= TARGET_ULONG_MAX) {
+			} else if(is_oct_hex && v >= 0 && (unsigned long long)v <= (unsigned long long)TARGET_ULONG_MAX) {
 				lexer_token.datatype = type_unsigned_long;
 				return;
 			}
@@ -479,34 +487,74 @@ end:
  */
 static void parse_number_hex(void)
 {
+	bool is_float = false;
 	assert(c == 'x' || c == 'X');
 	next_char();
 
+	obstack_1grow(&symbol_obstack, '0');
+	obstack_1grow(&symbol_obstack, 'x');
+
 	while(isxdigit(c)) {
 		obstack_1grow(&symbol_obstack, (char) c);
 		next_char();
 	}
-	obstack_1grow(&symbol_obstack, '\0');
-	char *string = obstack_finish(&symbol_obstack);
 
-	if(c == '.' || c == 'p' || c == 'P') {
+	if (c == '.') {
+		obstack_1grow(&symbol_obstack, (char) c);
+		next_char();
+
+		while (isxdigit(c)) {
+			obstack_1grow(&symbol_obstack, (char) c);
+			next_char();
+		}
+		is_float = true;
+	}
+	if (c == 'p' || c == 'P') {
+		obstack_1grow(&symbol_obstack, (char) c);
 		next_char();
-		internal_error("Hex floating point numbers not implemented yet");
+
+		if (c == '-' || c == '+') {
+			obstack_1grow(&symbol_obstack, (char) c);
+			next_char();
+		}
+
+		while (isxdigit(c)) {
+			obstack_1grow(&symbol_obstack, (char) c);
+			next_char();
+		}
+		is_float = true;
 	}
+
+	obstack_1grow(&symbol_obstack, '\0');
+	char *string = obstack_finish(&symbol_obstack);
 	if(*string == '\0') {
 		parse_error("invalid hex number");
 		lexer_token.type = T_ERROR;
+		obstack_free(&symbol_obstack, string);
+		return;
 	}
 
-	const char *endptr;
-	lexer_token.type       = T_INTEGER;
-	lexer_token.v.intvalue = parse_int_string(string, &endptr, 16);
-	if(*endptr != '\0') {
-		parse_error("hex number literal too long");
+	if (is_float) {
+		char *endptr;
+		lexer_token.type         = T_FLOATINGPOINT;
+		lexer_token.v.floatvalue = strtold(string, &endptr);
+
+		if(*endptr != '\0') {
+			parse_error("invalid hex float literal");
+		}
+
+		parse_floating_suffix();
+	} else {
+		const char *endptr;
+		lexer_token.type       = T_INTEGER;
+		lexer_token.v.intvalue = parse_int_string(string + 2, &endptr, 16);
+		if(*endptr != '\0') {
+			parse_error("hex number literal too long");
+		}
+		parse_integer_suffix(true);
 	}
 
 	obstack_free(&symbol_obstack, string);
-	parse_integer_suffix(true);
 }
 
 /**
@@ -561,23 +609,23 @@ static void parse_number_oct(void)
 static void parse_number_dec(void)
 {
 	bool is_float = false;
-	while(isdigit(c)) {
+	while (isdigit(c)) {
 		obstack_1grow(&symbol_obstack, (char) c);
 		next_char();
 	}
 
-	if(c == '.') {
+	if (c == '.') {
 		obstack_1grow(&symbol_obstack, '.');
 		next_char();
 
-		while(isdigit(c)) {
+		while (isdigit(c)) {
 			obstack_1grow(&symbol_obstack, (char) c);
 			next_char();
 		}
 		is_float = true;
 	}
 	if(c == 'e' || c == 'E') {
-		obstack_1grow(&symbol_obstack, 'e');
+		obstack_1grow(&symbol_obstack, (char) c);
 		next_char();
 
 		if(c == '-' || c == '+') {
@@ -744,7 +792,7 @@ static int parse_escape_sequence(void)
 	int ec = c;
 	next_char();
 
-	switch(ec) {
+	switch (ec) {
 	case '"':  return '"';
 	case '\'': return '\'';
 	case '\\': return '\\';
@@ -770,6 +818,10 @@ static int parse_escape_sequence(void)
 	case EOF:
 		parse_error("reached end of file while parsing escape sequence");
 		return EOF;
+	case 'e':
+		if (c_mode & _GNUC)
+			return 27;   /* hopefully 27 is ALWAYS the code for ESACAPE */
+		/*fallthrough*/
 	default:
 		parse_error("unknown escape sequence");
 		return EOF;
@@ -1098,7 +1150,10 @@ static void skip_multiline_comment(void)
 		case '/':
 			next_char();
 			if (c == '*') {
-				/* TODO: nested comment, warn here */
+				/* nested comment, warn here */
+				if (warning.comment) {
+					warningf(&lexer_token.source_position, "'/*' within comment");
+				}
 			}
 			break;
 		case '*':
@@ -1140,6 +1195,15 @@ static void skip_line_comment(void)
 		case '\r':
 			return;
 
+		case '\\':
+			next_char();
+			if (c == '\n' || c == '\r') {
+				if (warning.comment)
+					warningf(&lexer_token.source_position, "multi-line comment");
+				return;
+			}
+			break;
+
 		default:
 			next_char();
 			break;
@@ -1222,7 +1286,7 @@ static void parse_line_directive(void)
 /**
  * STDC pragmas.
  */
-typedef enum {
+typedef enum stdc_pragma_kind_t {
 	STDC_UNKNOWN,
 	STDC_FP_CONTRACT,
 	STDC_FENV_ACCESS,
@@ -1232,7 +1296,7 @@ typedef enum {
 /**
  * STDC pragma values.
  */
-typedef enum {
+typedef enum stdc_pragma_value_kind_t {
 	STDC_VALUE_UNKNOWN,
 	STDC_VALUE_ON,
 	STDC_VALUE_OFF,
@@ -1428,16 +1492,7 @@ void lexer_next_preprocessing_token(void)
 
 		case '.':
 			MAYBE_PROLOG
-				case '0':
-				case '1':
-				case '2':
-				case '3':
-				case '4':
-				case '5':
-				case '6':
-				case '7':
-				case '8':
-				case '9':
+				DIGITS
 					put_back(c);
 					c = '.';
 					parse_number_dec();
@@ -1568,8 +1623,9 @@ void lexer_next_preprocessing_token(void)
 			return;
 
 		default:
+dollar_sign:
+			errorf(&lexer_token.source_position, "unknown character '%c' found", c);
 			next_char();
-			errorf(&lexer_token.source_position, "unknown character '%c' found\n", c);
 			lexer_token.type = T_ERROR;
 			return;
 		}
@@ -1579,15 +1635,13 @@ void lexer_next_preprocessing_token(void)
 void lexer_next_token(void)
 {
 	lexer_next_preprocessing_token();
-	if(lexer_token.type != '\n')
-		return;
 
+	while (lexer_token.type == '\n') {
 newline_found:
-	do {
 		lexer_next_preprocessing_token();
-	} while(lexer_token.type == '\n');
+	}
 
-	if(lexer_token.type == '#') {
+	if (lexer_token.type == '#') {
 		parse_preprocessor_directive();
 		goto newline_found;
 	}
@@ -1596,6 +1650,7 @@ newline_found:
 void init_lexer(void)
 {
 	strset_init(&stringset);
+	symbol_L = symbol_table_insert("L");
 }
 
 void lexer_open_stream(FILE *stream, const char *input_name)
@@ -1604,7 +1659,6 @@ void lexer_open_stream(FILE *stream, const char *input_name)
 	lexer_token.source_position.linenr     = 0;
 	lexer_token.source_position.input_name = input_name;
 
-	symbol_L = symbol_table_insert("L");
 	bufpos = NULL;
 	bufend = NULL;
 
@@ -1613,6 +1667,20 @@ void lexer_open_stream(FILE *stream, const char *input_name)
 	c = '\n';
 }
 
+void lexer_open_buffer(const char *buffer, size_t len, const char *input_name)
+{
+	input                                  = NULL;
+	lexer_token.source_position.linenr     = 0;
+	lexer_token.source_position.input_name = input_name;
+
+	bufpos = buffer;
+	bufend = buffer + len;
+
+	/* place a virtual \n at the beginning so the lexer knows that we're
+	 * at the beginning of a line */
+	c = '\n';
+}
+
 void exit_lexer(void)
 {
 	strset_destroy(&stringset);