-/*
- * scanner.c
- *
- * Ullrich von Bassewitz, 07.06.1998
- */
+/*****************************************************************************/
+/* */
+/* scanner.c */
+/* */
+/* Source file line info structure */
+/* */
+/* */
+/* */
+/* (C) 1998-2003 Ullrich von Bassewitz */
+/* Römerstrasse 52 */
+/* D-70794 Filderstadt */
+/* EMail: uz@cc65.org */
+/* */
+/* */
+/* This software is provided 'as-is', without any expressed or implied */
+/* warranty. In no event will the authors be held liable for any damages */
+/* arising from the use of this software. */
+/* */
+/* Permission is granted to anyone to use this software for any purpose, */
+/* including commercial applications, and to alter it and redistribute it */
+/* freely, subject to the following restrictions: */
+/* */
+/* 1. The origin of this software must not be misrepresented; you must not */
+/* claim that you wrote the original software. If you use this software */
+/* in a product, an acknowledgment in the product documentation would be */
+/* appreciated but is not required. */
+/* 2. Altered source versions must be plainly marked as such, and must not */
+/* be misrepresented as being the original software. */
+/* 3. This notice may not be removed or altered from any source */
+/* distribution. */
+/* */
+/*****************************************************************************/
#include <ctype.h>
/* common */
+#include "chartype.h"
#include "tgttrans.h"
/* cc65 */
#include "error.h"
#include "function.h"
#include "global.h"
+#include "hexval.h"
#include "ident.h"
#include "input.h"
#include "litpool.h"
unsigned char Tok; /* The token */
unsigned char Type; /* Token type */
} Keywords [] = {
+ { "_Pragma", TOK_PRAGMA, TT_C },
+ { "__AX__", TOK_AX, TT_C },
{ "__A__", TOK_A, TT_C },
- { "__AX__", TOK_AX, TT_C },
- { "__EAX__", TOK_EAX, TT_C },
- { "__X__", TOK_X, TT_C },
- { "__Y__", TOK_Y, TT_C },
- { "__asm__", TOK_ASM, TT_C },
- { "__attribute__", TOK_ATTRIBUTE, TT_C },
- { "__far__", TOK_FAR, TT_C },
- { "__fastcall__", TOK_FASTCALL, TT_C },
+ { "__EAX__", TOK_EAX, TT_C },
+ { "__X__", TOK_X, TT_C },
+ { "__Y__", TOK_Y, TT_C },
+ { "__asm__", TOK_ASM, TT_C },
+ { "__attribute__", TOK_ATTRIBUTE, TT_C },
+ { "__far__", TOK_FAR, TT_C },
+ { "__fastcall__", TOK_FASTCALL, TT_C },
+ { "__near__", TOK_NEAR, TT_C },
{ "asm", TOK_ASM, TT_EXT },
{ "auto", TOK_AUTO, TT_C },
{ "break", TOK_BREAK, TT_C },
{ "if", TOK_IF, TT_C },
{ "int", TOK_INT, TT_C },
{ "long", TOK_LONG, TT_C },
+ { "near", TOK_NEAR, TT_EXT },
{ "register", TOK_REGISTER, TT_C },
+ { "restrict", TOK_RESTRICT, TT_C },
{ "return", TOK_RETURN, TT_C },
{ "short", TOK_SHORT, TT_C },
{ "signed", TOK_SIGNED, TT_C },
}
Preprocess ();
}
- if (CurC == ' ' || CurC == '\r') {
+ if (IsSpace (CurC)) {
NextChar ();
} else {
return 1;
*s++ = CurC;
}
NextChar ();
- } while (IsIdent (CurC) || isdigit (CurC));
+ } while (IsIdent (CurC) || IsDigit (CurC));
*s = '\0';
}
-static void unknown (char C)
+static void UnknownChar (char C)
/* Error message for unknown character */
{
Error ("Invalid input character with code %02X", C & 0xFF);
-static unsigned hexval (int c)
-/* Convert a hex digit into a value */
-{
- if (!isxdigit (c)) {
- Error ("Invalid hexadecimal digit: `%c'", c);
- }
- if (isdigit (c)) {
- return c - '0';
- } else {
- return toupper (c) - 'A' + 10;
- }
-}
-
-
-
static void SetTok (int tok)
-/* set nxttok and bump line ptr */
+/* Set NextTok.Tok and bump line ptr */
{
- nxttok = tok;
+ NextTok.Tok = tok;
NextChar ();
}
-static int SignExtendChar (int C)
-/* Do correct sign extension of a character */
-{
- if (SignedChars && (C & 0x80) != 0) {
- return C | ~0xFF;
- } else {
- return C & 0xFF;
- }
-}
-
-
-
static int ParseChar (void)
/* Parse a character. Converts \n into EOL, etc. */
{
- int i;
- unsigned val;
+ int I;
+ unsigned Val;
int C;
/* Check for escape chars */
if (CurC == '\\') {
NextChar ();
switch (CurC) {
+ case '?':
+ C = '\?';
+ break;
+ case 'a':
+ C = '\a';
+ break;
case 'b':
C = '\b';
- break;
+ break;
case 'f':
- C = '\f';
- break;
+ C = '\f';
+ break;
case 'r':
- C = '\r';
- break;
+ C = '\r';
+ break;
case 'n':
- C = '\n';
- break;
+ C = '\n';
+ break;
case 't':
- C = '\t';
- break;
+ C = '\t';
+ break;
+ case 'v':
+ C = '\v';
+ break;
case '\"':
- C = '\"';
- break;
+ C = '\"';
+ break;
case '\'':
- C = '\'';
- break;
+ C = '\'';
+ break;
case '\\':
- C = '\\';
- break;
+ C = '\\';
+ break;
case 'x':
case 'X':
- /* Hex character constant */
- NextChar ();
- val = hexval (CurC) << 4;
- NextChar ();
- C = val | hexval (CurC); /* Do not translate */
- break;
+ /* Hex character constant */
+ NextChar ();
+ Val = HexVal (CurC) << 4;
+ NextChar ();
+ C = Val | HexVal (CurC); /* Do not translate */
+ break;
case '0':
case '1':
+ case '2':
+ case '3':
+ case '4':
+ case '5':
+ case '6':
+ case '7':
/* Octal constant */
- i = 0;
- C = CurC - '0';
- while (NextC >= '0' && NextC <= '7' && i++ < 4) {
- NextChar ();
- C = (C << 3) | (CurC - '0');
+ I = 0;
+ Val = CurC - '0';
+ while (NextC >= '0' && NextC <= '7' && ++I <= 3) {
+ NextChar ();
+ Val = (Val << 3) | (CurC - '0');
}
+ C = (int) Val;
+ if (Val >= 256) {
+ Error ("Character constant out of range");
+ C = ' ';
+ }
break;
default:
Error ("Illegal character constant");
C = ' ';
+ /* Try to do error recovery, otherwise the compiler will spit
+ * out thousands of errors in this place and abort.
+ */
+ if (CurC != '\'' && CurC != '\0') {
+ while (NextC != '\'' && NextC != '\"' && NextC != '\0') {
+ NextChar ();
+ }
+ }
break;
}
} else {
}
/* Setup values and attributes */
- nxttok = TOK_CCONST;
+ NextTok.Tok = TOK_CCONST;
/* Translate into target charset */
- nxtval = SignExtendChar (TgtTranslateChar (C));
+ NextTok.IVal = SignExtendChar (TgtTranslateChar (C));
/* Character constants have type int */
- nxttype = type_int;
+ NextTok.Type = type_int;
}
static void StringConst (void)
/* Parse a quoted string */
{
- nxtval = GetLiteralOffs ();
- nxttok = TOK_SCONST;
+ NextTok.IVal = GetLiteralPoolOffs ();
+ NextTok.Tok = TOK_SCONST;
/* Be sure to concatenate strings */
while (CurC == '\"') {
{
ident token;
+ /* We have to skip white space here before shifting tokens, since the
+ * tokens and the current line info is invalid at startup and will get
+ * initialized by reading the first time from the file. Remember if
+ * we were at end of input and handle that later.
+ */
+ int GotEOF = (SkipWhite() == 0);
+
/* Current token is the lookahead token */
+ if (CurTok.LI) {
+ ReleaseLineInfo (CurTok.LI);
+ }
CurTok = NextTok;
+ /* When reading the first time from the file, the line info in NextTok,
+ * which was copied to CurTok is invalid. Since the information from
+ * the token is used for error messages, we must make it valid.
+ */
+ if (CurTok.LI == 0) {
+ CurTok.LI = UseLineInfo (GetCurLineInfo ());
+ }
+
/* Remember the starting position of the next token */
- NextTok.Pos = GetCurrentLine();
+ NextTok.LI = UseLineInfo (GetCurLineInfo ());
- /* Skip spaces and read the next line if needed */
- if (SkipWhite () == 0) {
+ /* Now handle end of input. */
+ if (GotEOF) {
/* End of file reached */
- nxttok = TOK_CEOF;
+ NextTok.Tok = TOK_CEOF;
return;
}
/* Determine the next token from the lookahead */
- if (isdigit (CurC)) {
+ if (IsDigit (CurC)) {
/* A number */
int HaveSuffix; /* True if we have a type suffix */
unsigned types; /* Possible types */
- unsigned base;
- unsigned long k; /* Value */
+ unsigned Base;
+ unsigned DigitVal;
+ unsigned long k; /* Value */
k = 0;
- base = 10;
+ Base = 10;
types = IT_INT | IT_LONG | IT_ULONG;
if (CurC == '0') {
/* gobble 0 and examin next char */
NextChar ();
if (toupper (CurC) == 'X') {
- base = 16;
- nxttype = type_uint;
+ Base = 16;
+ NextTok.Type = type_uint;
NextChar (); /* gobble "x" */
} else {
- base = 8;
+ Base = 8;
}
}
- while (1) {
- if (isdigit (CurC)) {
- k = k * base + (CurC - '0');
- } else if (base == 16 && isxdigit (CurC)) {
- k = (k << 4) + hexval (CurC);
- } else {
- break; /* not digit */
- }
- NextChar (); /* gobble char */
- }
+ while (IsXDigit (CurC) && (DigitVal = HexVal (CurC)) < Base) {
+ k = k * Base + DigitVal;
+ NextChar ();
+ }
+ /* Check for errorneous digits */
+ if (Base == 8 && IsDigit (CurC)) {
+ Error ("Numeric constant contains digits beyond the radix");
+ /* Do error recovery */
+ do {
+ NextChar ();
+ } while (IsDigit (CurC));
+ } else if (Base != 16 && IsXDigit (CurC)) {
+ Error ("Nondigits in number and not hexadecimal");
+ do {
+ NextChar ();
+ } while (IsXDigit (CurC));
+ }
/* Check for a suffix */
HaveSuffix = 1;
/* Now set the type string to the smallest type in types */
if (types & IT_INT) {
- nxttype = type_int;
+ NextTok.Type = type_int;
} else if (types & IT_UINT) {
- nxttype = type_uint;
+ NextTok.Type = type_uint;
} else if (types & IT_LONG) {
- nxttype = type_long;
+ NextTok.Type = type_long;
} else {
- nxttype = type_ulong;
+ NextTok.Type = type_ulong;
}
/* Set the value and the token */
- nxtval = k;
- nxttok = TOK_ICONST;
+ NextTok.IVal = k;
+ NextTok.Tok = TOK_ICONST;
return;
}
if (IsSym (token)) {
/* Check for a keyword */
- if ((nxttok = FindKey (token)) != TOK_IDENT) {
+ if ((NextTok.Tok = FindKey (token)) != TOK_IDENT) {
/* Reserved word found */
return;
}
/* No reserved word, check for special symbols */
if (token [0] == '_') {
/* Special symbols */
- if (strcmp (token, "__FILE__") == 0) {
- nxtval = AddLiteral (GetCurrentFile());
- nxttok = TOK_SCONST;
+ if (strcmp (token, "__FILE__") == 0) {
+ NextTok.IVal = AddLiteral (GetCurrentFile());
+ NextTok.Tok = TOK_SCONST;
return;
} else if (strcmp (token, "__LINE__") == 0) {
- nxttok = TOK_ICONST;
- nxtval = GetCurrentLine();
- nxttype = type_int;
- return;
- } else if (strcmp (token, "__fixargs__") == 0) {
- nxttok = TOK_ICONST;
- nxtval = GetParamSize (CurrentFunc);
- nxttype = type_uint;
+ NextTok.Tok = TOK_ICONST;
+ NextTok.IVal = GetCurrentLine();
+ NextTok.Type = type_int;
return;
} else if (strcmp (token, "__func__") == 0) {
/* __func__ is only defined in functions */
if (CurrentFunc) {
- nxtval = AddLiteral (GetFuncName (CurrentFunc));
- nxttok = TOK_SCONST;
+ NextTok.IVal = AddLiteral (F_GetFuncName (CurrentFunc));
+ NextTok.Tok = TOK_SCONST;
return;
}
}
if (CurC == '=') {
SetTok (TOK_NE);
} else {
- nxttok = TOK_BOOL_NOT;
+ NextTok.Tok = TOK_BOOL_NOT;
}
break;
if (CurC == '=') {
SetTok (TOK_MOD_ASSIGN);
} else {
- nxttok = TOK_MOD;
+ NextTok.Tok = TOK_MOD;
}
break;
SetTok (TOK_AND_ASSIGN);
break;
default:
- nxttok = TOK_AND;
+ NextTok.Tok = TOK_AND;
}
break;
if (CurC == '=') {
SetTok (TOK_MUL_ASSIGN);
} else {
- nxttok = TOK_STAR;
+ NextTok.Tok = TOK_STAR;
}
break;
SetTok (TOK_PLUS_ASSIGN);
break;
default:
- nxttok = TOK_PLUS;
+ NextTok.Tok = TOK_PLUS;
}
break;
SetTok (TOK_PTR_REF);
break;
default:
- nxttok = TOK_MINUS;
+ NextTok.Tok = TOK_MINUS;
}
break;
if (CurC == '.') {
SetTok (TOK_ELLIPSIS);
} else {
- unknown (CurC);
+ UnknownChar (CurC);
}
} else {
- nxttok = TOK_DOT;
+ NextTok.Tok = TOK_DOT;
}
break;
if (CurC == '=') {
SetTok (TOK_DIV_ASSIGN);
} else {
- nxttok = TOK_DIV;
+ NextTok.Tok = TOK_DIV;
}
break;
if (CurC == '=') {
SetTok (TOK_SHL_ASSIGN);
} else {
- nxttok = TOK_SHL;
+ NextTok.Tok = TOK_SHL;
}
break;
default:
- nxttok = TOK_LT;
+ NextTok.Tok = TOK_LT;
}
break;
if (CurC == '=') {
SetTok (TOK_EQ);
} else {
- nxttok = TOK_ASSIGN;
+ NextTok.Tok = TOK_ASSIGN;
}
break;
if (CurC == '=') {
SetTok (TOK_SHR_ASSIGN);
} else {
- nxttok = TOK_SHR;
+ NextTok.Tok = TOK_SHR;
}
break;
default:
- nxttok = TOK_GT;
+ NextTok.Tok = TOK_GT;
}
break;
if (CurC == '=') {
SetTok (TOK_XOR_ASSIGN);
} else {
- nxttok = TOK_XOR;
+ NextTok.Tok = TOK_XOR;
}
break;
SetTok (TOK_OR_ASSIGN);
break;
default:
- nxttok = TOK_OR;
+ NextTok.Tok = TOK_OR;
}
break;
SetTok (TOK_COMP);
break;
- case '#':
- /* Skip it and following whitespace */
- do {
- NextChar ();
- } while (CurC == ' ');
- if (!IsSym (token) || strcmp (token, "pragma") != 0) {
- /* OOPS - should not happen */
- Error ("Preprocessor directive expected");
- }
- nxttok = TOK_PRAGMA;
- break;
-
default:
- unknown (CurC);
+ UnknownChar (CurC);
}
-void Consume (token_t Token, const char* ErrorMsg)
+void SkipTokens (const token_t* TokenList, unsigned TokenCount)
+/* Skip tokens until we reach TOK_CEOF or a token in the given token list.
+ * This routine is used for error recovery.
+ */
+{
+ while (CurTok.Tok != TOK_CEOF) {
+
+ /* Check if the current token is in the token list */
+ unsigned I;
+ for (I = 0; I < TokenCount; ++I) {
+ if (CurTok.Tok == TokenList[I]) {
+ /* Found a token in the list */
+ return;
+ }
+ }
+
+ /* Not in the list: Skip it */
+ NextToken ();
+
+ }
+}
+
+
+
+int Consume (token_t Token, const char* ErrorMsg)
/* Eat token if it is the next in the input stream, otherwise print an error
- * message.
+ * message. Returns true if the token was found and false otherwise.
*/
{
- if (curtok == Token) {
+ if (CurTok.Tok == Token) {
NextToken ();
+ return 1;
} else {
Error (ErrorMsg);
+ return 0;
}
}
-void ConsumeColon (void)
+int ConsumeColon (void)
/* Check for a colon and skip it. */
{
- Consume (TOK_COLON, "`:' expected");
+ return Consume (TOK_COLON, "`:' expected");
}
-void ConsumeSemi (void)
+int ConsumeSemi (void)
/* Check for a semicolon and skip it. */
{
/* Try do be smart about typos... */
- if (curtok == TOK_SEMI) {
- NextToken ();
+ if (CurTok.Tok == TOK_SEMI) {
+ NextToken ();
+ return 1;
} else {
Error ("`;' expected");
- if (curtok == TOK_COLON || curtok == TOK_COMMA) {
+ if (CurTok.Tok == TOK_COLON || CurTok.Tok == TOK_COMMA) {
NextToken ();
}
+ return 0;
}
}
-void ConsumeComma (void)
+int ConsumeComma (void)
/* Check for a comma and skip it. */
{
/* Try do be smart about typos... */
if (CurTok.Tok == TOK_COMMA) {
- NextToken ();
+ NextToken ();
+ return 1;
} else {
Error ("`,' expected");
if (CurTok.Tok == TOK_SEMI) {
NextToken ();
}
+ return 0;
}
}
-void ConsumeLParen (void)
+int ConsumeLParen (void)
/* Check for a left parenthesis and skip it */
{
- Consume (TOK_LPAREN, "`(' expected");
+ return Consume (TOK_LPAREN, "`(' expected");
}
-void ConsumeRParen (void)
+int ConsumeRParen (void)
/* Check for a right parenthesis and skip it */
{
- Consume (TOK_RPAREN, "`)' expected");
+ return Consume (TOK_RPAREN, "`)' expected");
}
-void ConsumeLBrack (void)
+int ConsumeLBrack (void)
/* Check for a left bracket and skip it */
{
- Consume (TOK_LBRACK, "`[' expected");
+ return Consume (TOK_LBRACK, "`[' expected");
}
-void ConsumeRBrack (void)
+int ConsumeRBrack (void)
/* Check for a right bracket and skip it */
{
- Consume (TOK_RBRACK, "`]' expected");
+ return Consume (TOK_RBRACK, "`]' expected");
}
-void ConsumeLCurly (void)
+int ConsumeLCurly (void)
/* Check for a left curly brace and skip it */
{
- Consume (TOK_LCURLY, "`{' expected");
+ return Consume (TOK_LCURLY, "`{' expected");
}
-void ConsumeRCurly (void)
+int ConsumeRCurly (void)
/* Check for a right curly brace and skip it */
{
- Consume (TOK_RCURLY, "`}' expected");
+ return Consume (TOK_RCURLY, "`}' expected");
}