1 /*****************************************************************************/
5 /* Source file line info structure */
9 /* (C) 1998-2003 Ullrich von Bassewitz */
10 /* Römerstrasse 52 */
11 /* D-70794 Filderstadt */
12 /* EMail: uz@cc65.org */
15 /* This software is provided 'as-is', without any expressed or implied */
16 /* warranty. In no event will the authors be held liable for any damages */
17 /* arising from the use of this software. */
19 /* Permission is granted to anyone to use this software for any purpose, */
20 /* including commercial applications, and to alter it and redistribute it */
21 /* freely, subject to the following restrictions: */
23 /* 1. The origin of this software must not be misrepresented; you must not */
24 /* claim that you wrote the original software. If you use this software */
25 /* in a product, an acknowledgment in the product documentation would be */
26 /* appreciated but is not required. */
27 /* 2. Altered source versions must be plainly marked as such, and must not */
28 /* be misrepresented as being the original software. */
29 /* 3. This notice may not be removed or altered from any source */
32 /*****************************************************************************/
62 /*****************************************************************************/
64 /*****************************************************************************/
68 Token CurTok; /* The current token */
69 Token NextTok; /* The next token */
74 #define TT_C 0 /* ANSI C token */
75 #define TT_EXT 1 /* cc65 extension */
78 static const struct Keyword {
79 char* Key; /* Keyword name */
80 unsigned char Tok; /* The token */
81 unsigned char Type; /* Token type */
83 { "_Pragma", TOK_PRAGMA, TT_C },
84 { "__AX__", TOK_AX, TT_C },
85 { "__A__", TOK_A, TT_C },
86 { "__EAX__", TOK_EAX, TT_C },
87 { "__X__", TOK_X, TT_C },
88 { "__Y__", TOK_Y, TT_C },
89 { "__asm__", TOK_ASM, TT_C },
90 { "__attribute__", TOK_ATTRIBUTE, TT_C },
91 { "__far__", TOK_FAR, TT_C },
92 { "__fastcall__", TOK_FASTCALL, TT_C },
93 { "__near__", TOK_NEAR, TT_C },
94 { "asm", TOK_ASM, TT_EXT },
95 { "auto", TOK_AUTO, TT_C },
96 { "break", TOK_BREAK, TT_C },
97 { "case", TOK_CASE, TT_C },
98 { "char", TOK_CHAR, TT_C },
99 { "const", TOK_CONST, TT_C },
100 { "continue", TOK_CONTINUE, TT_C },
101 { "default", TOK_DEFAULT, TT_C },
102 { "do", TOK_DO, TT_C },
103 { "double", TOK_DOUBLE, TT_C },
104 { "else", TOK_ELSE, TT_C },
105 { "enum", TOK_ENUM, TT_C },
106 { "extern", TOK_EXTERN, TT_C },
107 { "far", TOK_FAR, TT_EXT },
108 { "fastcall", TOK_FASTCALL, TT_EXT },
109 { "float", TOK_FLOAT, TT_C },
110 { "for", TOK_FOR, TT_C },
111 { "goto", TOK_GOTO, TT_C },
112 { "if", TOK_IF, TT_C },
113 { "int", TOK_INT, TT_C },
114 { "long", TOK_LONG, TT_C },
115 { "near", TOK_NEAR, TT_EXT },
116 { "register", TOK_REGISTER, TT_C },
117 { "restrict", TOK_RESTRICT, TT_C },
118 { "return", TOK_RETURN, TT_C },
119 { "short", TOK_SHORT, TT_C },
120 { "signed", TOK_SIGNED, TT_C },
121 { "sizeof", TOK_SIZEOF, TT_C },
122 { "static", TOK_STATIC, TT_C },
123 { "struct", TOK_STRUCT, TT_C },
124 { "switch", TOK_SWITCH, TT_C },
125 { "typedef", TOK_TYPEDEF, TT_C },
126 { "union", TOK_UNION, TT_C },
127 { "unsigned", TOK_UNSIGNED, TT_C },
128 { "void", TOK_VOID, TT_C },
129 { "volatile", TOK_VOLATILE, TT_C },
130 { "while", TOK_WHILE, TT_C },
132 #define KEY_COUNT (sizeof (Keywords) / sizeof (Keywords [0]))
136 /* Stuff for determining the type of an integer constant */
140 #define IT_ULONG 0x08
144 /*****************************************************************************/
146 /*****************************************************************************/
150 static int CmpKey (const void* Key, const void* Elem)
151 /* Compare function for bsearch */
153 return strcmp ((const char*) Key, ((const struct Keyword*) Elem)->Key);
158 static int FindKey (const char* Key)
159 /* Find a keyword and return the token. Return IDENT if the token is not a
164 K = bsearch (Key, Keywords, KEY_COUNT, sizeof (Keywords [0]), CmpKey);
165 if (K && (K->Type != TT_EXT || ANSI == 0)) {
174 static int SkipWhite (void)
175 /* Skip white space in the input stream, reading and preprocessing new lines
176 * if necessary. Return 0 if end of file is reached, return 1 otherwise.
181 if (NextLine () == 0) {
186 if (IsSpace (CurC)) {
196 void SymName (char* s)
197 /* Get symbol from input stream */
201 if (k != MAX_IDENTLEN) {
206 } while (IsIdent (CurC) || IsDigit (CurC));
213 /* Get symbol from input stream or return 0 if not a symbol. */
215 if (IsIdent (CurC)) {
225 static void UnknownChar (char C)
226 /* Error message for unknown character */
228 Error ("Invalid input character with code %02X", C & 0xFF);
229 NextChar (); /* Skip */
234 static void SetTok (int tok)
235 /* Set NextTok.Tok and bump line ptr */
243 static int ParseChar (void)
244 /* Parse a character. Converts \n into EOL, etc. */
250 /* Check for escape chars */
280 /* Hex character constant */
282 val = HexVal (CurC) << 4;
284 C = val | HexVal (CurC); /* Do not translate */
293 while (NextC >= '0' && NextC <= '7' && i++ < 4) {
295 C = (C << 3) | (CurC - '0');
299 Error ("Illegal character constant");
307 /* Skip the character read */
310 /* Do correct sign extension */
311 return SignExtendChar (C);
316 static void CharConst (void)
317 /* Parse a character constant. */
327 /* Check for closing quote */
329 Error ("`\'' expected");
335 /* Setup values and attributes */
336 NextTok.Tok = TOK_CCONST;
338 /* Translate into target charset */
339 NextTok.IVal = SignExtendChar (TgtTranslateChar (C));
341 /* Character constants have type int */
342 NextTok.Type = type_int;
347 static void StringConst (void)
348 /* Parse a quoted string */
350 NextTok.IVal = GetLiteralPoolOffs ();
351 NextTok.Tok = TOK_SCONST;
353 /* Be sure to concatenate strings */
354 while (CurC == '\"') {
356 /* Skip the quote char */
359 while (CurC != '\"') {
361 Error ("Unexpected newline");
364 AddLiteralChar (ParseChar ());
367 /* Skip closing quote char if there was one */
370 /* Skip white space, read new input */
375 /* Terminate the string */
376 AddLiteralChar ('\0');
381 void NextToken (void)
382 /* Get next token from input stream */
386 /* We have to skip white space here before shifting tokens, since the
387 * tokens and the current line info is invalid at startup and will get
388 * initialized by reading the first time from the file. Remember if
389 * we were at end of input and handle that later.
391 int GotEOF = (SkipWhite() == 0);
393 /* Current token is the lookahead token */
395 ReleaseLineInfo (CurTok.LI);
399 /* When reading the first time from the file, the line info in NextTok,
400 * which was copied to CurTok is invalid. Since the information from
401 * the token is used for error messages, we must make it valid.
403 if (CurTok.LI == 0) {
404 CurTok.LI = UseLineInfo (GetCurLineInfo ());
407 /* Remember the starting position of the next token */
408 NextTok.LI = UseLineInfo (GetCurLineInfo ());
410 /* Now handle end of input. */
412 /* End of file reached */
413 NextTok.Tok = TOK_CEOF;
417 /* Determine the next token from the lookahead */
418 if (IsDigit (CurC)) {
421 int HaveSuffix; /* True if we have a type suffix */
422 unsigned types; /* Possible types */
425 unsigned long k; /* Value */
429 types = IT_INT | IT_LONG | IT_ULONG;
432 /* Octal or hex constants may also be of type unsigned int */
433 types = IT_INT | IT_UINT | IT_LONG | IT_ULONG;
434 /* gobble 0 and examin next char */
436 if (toupper (CurC) == 'X') {
438 NextTok.Type = type_uint;
439 NextChar (); /* gobble "x" */
444 while (IsXDigit (CurC) && (DigitVal = HexVal (CurC)) < Base) {
445 k = k * Base + DigitVal;
448 /* Check for errorneous digits */
449 if (Base == 8 && IsDigit (CurC)) {
450 Error ("Numeric constant contains digits beyond the radix");
451 /* Do error recovery */
454 } while (IsDigit (CurC));
455 } else if (Base != 16 && IsXDigit (CurC)) {
456 Error ("Nondigits in number and not hexadecimal");
459 } while (IsXDigit (CurC));
462 /* Check for a suffix */
464 if (CurC == 'u' || CurC == 'U') {
467 if (toupper (CurC) != 'L') {
468 types = IT_UINT | IT_ULONG;
473 } else if (CurC == 'l' || CurC == 'L') {
476 if (toupper (CurC) != 'U') {
477 types = IT_LONG | IT_ULONG;
486 /* Check the range to determine the type */
488 /* Out of range for int */
490 /* If the value is in the range 0x8000..0xFFFF, unsigned int is not
491 * allowed, and we don't have a type specifying suffix, emit a
494 if (k <= 0xFFFF && (types & IT_UINT) == 0 && !HaveSuffix) {
495 Warning ("Constant is long");
499 /* Out of range for unsigned int */
502 if (k > 0x7FFFFFFF) {
503 /* Out of range for long int */
507 /* Now set the type string to the smallest type in types */
508 if (types & IT_INT) {
509 NextTok.Type = type_int;
510 } else if (types & IT_UINT) {
511 NextTok.Type = type_uint;
512 } else if (types & IT_LONG) {
513 NextTok.Type = type_long;
515 NextTok.Type = type_ulong;
518 /* Set the value and the token */
520 NextTok.Tok = TOK_ICONST;
526 /* Check for a keyword */
527 if ((NextTok.Tok = FindKey (token)) != TOK_IDENT) {
528 /* Reserved word found */
531 /* No reserved word, check for special symbols */
532 if (token [0] == '_') {
533 /* Special symbols */
534 if (strcmp (token, "__FILE__") == 0) {
535 NextTok.IVal = AddLiteral (GetCurrentFile());
536 NextTok.Tok = TOK_SCONST;
538 } else if (strcmp (token, "__LINE__") == 0) {
539 NextTok.Tok = TOK_ICONST;
540 NextTok.IVal = GetCurrentLine();
541 NextTok.Type = type_int;
543 } else if (strcmp (token, "__func__") == 0) {
544 /* __func__ is only defined in functions */
546 NextTok.IVal = AddLiteral (F_GetFuncName (CurrentFunc));
547 NextTok.Tok = TOK_SCONST;
553 /* No reserved word but identifier */
554 strcpy (NextTok.Ident, token);
555 NextTok.Tok = TOK_IDENT;
559 /* Monstrous switch statement ahead... */
567 NextTok.Tok = TOK_BOOL_NOT;
578 SetTok (TOK_MOD_ASSIGN);
580 NextTok.Tok = TOK_MOD;
588 SetTok (TOK_BOOL_AND);
591 SetTok (TOK_AND_ASSIGN);
594 NextTok.Tok = TOK_AND;
613 SetTok (TOK_MUL_ASSIGN);
615 NextTok.Tok = TOK_STAR;
626 SetTok (TOK_PLUS_ASSIGN);
629 NextTok.Tok = TOK_PLUS;
644 SetTok (TOK_MINUS_ASSIGN);
647 SetTok (TOK_PTR_REF);
650 NextTok.Tok = TOK_MINUS;
659 SetTok (TOK_ELLIPSIS);
664 NextTok.Tok = TOK_DOT;
671 SetTok (TOK_DIV_ASSIGN);
673 NextTok.Tok = TOK_DIV;
694 SetTok (TOK_SHL_ASSIGN);
696 NextTok.Tok = TOK_SHL;
700 NextTok.Tok = TOK_LT;
709 NextTok.Tok = TOK_ASSIGN;
722 SetTok (TOK_SHR_ASSIGN);
724 NextTok.Tok = TOK_SHR;
728 NextTok.Tok = TOK_GT;
747 SetTok (TOK_XOR_ASSIGN);
749 NextTok.Tok = TOK_XOR;
761 SetTok (TOK_BOOL_OR);
764 SetTok (TOK_OR_ASSIGN);
767 NextTok.Tok = TOK_OR;
788 void SkipTokens (const token_t* TokenList, unsigned TokenCount)
789 /* Skip tokens until we reach TOK_CEOF or a token in the given token list.
790 * This routine is used for error recovery.
793 while (CurTok.Tok != TOK_CEOF) {
795 /* Check if the current token is in the token list */
797 for (I = 0; I < TokenCount; ++I) {
798 if (CurTok.Tok == TokenList[I]) {
799 /* Found a token in the list */
804 /* Not in the list: Skip it */
812 int Consume (token_t Token, const char* ErrorMsg)
813 /* Eat token if it is the next in the input stream, otherwise print an error
814 * message. Returns true if the token was found and false otherwise.
817 if (CurTok.Tok == Token) {
828 int ConsumeColon (void)
829 /* Check for a colon and skip it. */
831 return Consume (TOK_COLON, "`:' expected");
836 int ConsumeSemi (void)
837 /* Check for a semicolon and skip it. */
839 /* Try do be smart about typos... */
840 if (CurTok.Tok == TOK_SEMI) {
844 Error ("`;' expected");
845 if (CurTok.Tok == TOK_COLON || CurTok.Tok == TOK_COMMA) {
854 int ConsumeComma (void)
855 /* Check for a comma and skip it. */
857 /* Try do be smart about typos... */
858 if (CurTok.Tok == TOK_COMMA) {
862 Error ("`,' expected");
863 if (CurTok.Tok == TOK_SEMI) {
872 int ConsumeLParen (void)
873 /* Check for a left parenthesis and skip it */
875 return Consume (TOK_LPAREN, "`(' expected");
880 int ConsumeRParen (void)
881 /* Check for a right parenthesis and skip it */
883 return Consume (TOK_RPAREN, "`)' expected");
888 int ConsumeLBrack (void)
889 /* Check for a left bracket and skip it */
891 return Consume (TOK_LBRACK, "`[' expected");
896 int ConsumeRBrack (void)
897 /* Check for a right bracket and skip it */
899 return Consume (TOK_RBRACK, "`]' expected");
904 int ConsumeLCurly (void)
905 /* Check for a left curly brace and skip it */
907 return Consume (TOK_LCURLY, "`{' expected");
912 int ConsumeRCurly (void)
913 /* Check for a right curly brace and skip it */
915 return Consume (TOK_RCURLY, "`}' expected");