X-Git-Url: http://nsz.repo.hu/git/?a=blobdiff_plain;f=lexer.c;h=7c610825bc24344e083cf5d71f97f1379ccc3278;hb=00fad4c6c28b600fd17a6c3e8320dd6227c28e30;hp=66401d5408fac3c072431ec21a517f0f4e2e717d;hpb=6fb5b6813ea936aee285ede3b7b2b096303a3a4c;p=cparser

diff --git a/lexer.c b/lexer.c
index 66401d5..7c61082 100644
--- a/lexer.c
+++ b/lexer.c
@@ -1,5 +1,6 @@
 #include <config.h>
 
+#include "diagnostic.h"
 #include "lexer.h"
 #include "token_t.h"
 #include "symbol_table_t.h"
@@ -10,6 +11,7 @@
 #include "type_t.h"
 #include "target_architecture.h"
 #include "parser.h"
+#include "warning.h"
 
 #include <assert.h>
 #include <errno.h>
@@ -61,8 +63,8 @@ static void parse_error(const char *msg)
 
 static inline void next_real_char(void)
 {
-	bufpos++;
-	if(bufpos >= bufend) {
+	assert(bufpos <= bufend);
+	if (bufpos >= bufend) {
 		size_t s = fread(buf + MAX_PUTBACK, 1, sizeof(buf) - MAX_PUTBACK,
 		                 input);
 		if(s == 0) {
@@ -72,20 +74,13 @@ static inline void next_real_char(void)
 		bufpos = buf + MAX_PUTBACK;
 		bufend = buf + MAX_PUTBACK + s;
 	}
-	c = *(bufpos);
+	c = *bufpos++;
 }
 
 static inline void put_back(int pc)
 {
-	assert(bufpos >= buf);
-	//assert(bufpos < buf+MAX_PUTBACK || *bufpos == pc);
-
-	char *p = buf + (bufpos - buf);
-	*p = (char) pc;
-
-	/* going backwards in the buffer is legal as long as it's not more often
-	 * than MAX_PUTBACK */
-	bufpos--;
+	assert(bufpos > buf);
+	*(--bufpos - buf + buf) = (char) pc;
 
 #ifdef DEBUG_CHARS
 	printf("putback '%c'\n", pc);
@@ -101,11 +96,11 @@ static inline void next_char(void);
 			next_char();                      \
 		}                                     \
 		lexer_token.source_position.linenr++; \
-		code;                                 \
+		code                                  \
 	case '\n':                                \
 		next_char();                          \
 		lexer_token.source_position.linenr++; \
-		code;
+		code
 
 #define eat(c_type)  do { assert(c == c_type); next_char(); } while(0)
 
@@ -156,8 +151,8 @@ static inline void next_char(void)
 	case '>': c = '}'; break;
 	case '-': c = '~'; break;
 	default:
-		put_back('?');
 		put_back(c);
+		put_back('?');
 		c = '?';
 		break;
 	}
@@ -676,28 +671,30 @@ static int parse_escape_sequence(void)
 	}
 }
 
-const char *concat_strings(const char *s1, const char *s2)
+string_t concat_strings(const string_t *const s1, const string_t *const s2)
 {
-	size_t  len1   = strlen(s1);
-	size_t  len2   = strlen(s2);
+	const size_t len1 = s1->size - 1;
+	const size_t len2 = s2->size - 1;
 
-	char   *concat = obstack_alloc(&symbol_obstack, len1 + len2 + 1);
-	memcpy(concat, s1, len1);
-	memcpy(concat + len1, s2, len2 + 1);
+	char *const concat = obstack_alloc(&symbol_obstack, len1 + len2 + 1);
+	memcpy(concat, s1->begin, len1);
+	memcpy(concat + len1, s2->begin, len2 + 1);
 
+#if 0 /* TODO hash */
 	const char *result = strset_insert(&stringset, concat);
 	if(result != concat) {
 		obstack_free(&symbol_obstack, concat);
 	}
 
 	return result;
+#else
+	return (string_t){ concat, len1 + len2 + 1 };
+#endif
 }
 
 static void parse_string_literal(void)
 {
-	unsigned    start_linenr = lexer_token.source_position.linenr;
-	char       *string;
-	const char *result;
+	const unsigned start_linenr = lexer_token.source_position.linenr;
 
 	assert(c == '"');
 	next_char();
@@ -734,16 +731,22 @@ end_of_string:
 
 	/* add finishing 0 to the string */
 	obstack_1grow(&symbol_obstack, '\0');
-	string = obstack_finish(&symbol_obstack);
+	const size_t      size   = (size_t)obstack_object_size(&symbol_obstack);
+	const char *const string = obstack_finish(&symbol_obstack);
 
+#if 0 /* TODO hash */
 	/* check if there is already a copy of the string */
 	result = strset_insert(&stringset, string);
 	if(result != string) {
 		obstack_free(&symbol_obstack, string);
 	}
+#else
+	const char *const result = string;
+#endif
 
-	lexer_token.type     = T_STRING_LITERAL;
-	lexer_token.v.string = result;
+	lexer_token.type           = T_STRING_LITERAL;
+	lexer_token.v.string.begin = result;
+	lexer_token.v.string.size  = size;
 }
 
 static void parse_wide_character_constant(void)
@@ -995,7 +998,7 @@ static void parse_line_directive(void)
 		next_pp_token();
 	}
 	if(pp_token.type == T_STRING_LITERAL) {
-		lexer_token.source_position.input_name = pp_token.v.string;
+		lexer_token.source_position.input_name = pp_token.v.string.begin;
 		next_pp_token();
 	}
 
@@ -1035,6 +1038,10 @@ static void parse_preprocessor_identifier(void)
 		error_directive();
 		break;
 	case TP_pragma:
+		if (warning.unknown_pragmas) {
+			warningf(lexer_token.source_position, "encountered unknown #pragma");
+		}
+		eat_until_newline();
 		break;
 	}
 }
@@ -1122,6 +1129,21 @@ 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':
+					put_back(c);
+					c = '.';
+					parse_number_dec();
+					return;
+
 				case '.':
 					MAYBE_PROLOG
 					MAYBE('.', T_DOTDOTDOT)
@@ -1285,6 +1307,8 @@ void lexer_open_stream(FILE *stream, const char *input_name)
 	lexer_token.source_position.input_name = input_name;
 
 	symbol_L = symbol_table_insert("L");
+	bufpos = NULL;
+	bufend = NULL;
 
 	/* place a virtual \n at the beginning so the lexer knows that we're
 	 * at the beginning of a line */