X-Git-Url: http://nsz.repo.hu/git/?a=blobdiff_plain;f=lexer.c;h=f1b00e368942380411244d9d8822d65a4fe213cc;hb=79e35de99b8c98654757dcbf42070ab8e53bd856;hp=faecfe36fd75f32a12ceca0c883bdb597f1e97b1;hpb=506423738bdd438ce4609b37e9b3896189e8c483;p=cparser

diff --git a/lexer.c b/lexer.c
index faecfe3..f1b00e3 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;
 		}
 	}
@@ -744,7 +752,7 @@ static int parse_escape_sequence(void)
 	int ec = c;
 	next_char();
 
-	switch(ec) {
+	switch (ec) {
 	case '"':  return '"';
 	case '\'': return '\'';
 	case '\\': return '\\';
@@ -770,6 +778,10 @@ 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;
@@ -1222,7 +1234,7 @@ 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 +1244,7 @@ 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 +1440,7 @@ 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 +1571,9 @@ 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 +1583,13 @@ 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 +1598,7 @@ 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 +1607,6 @@ 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 +1615,20 @@ 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);