char *copy_token(char *p, char *tok)
{
char *ptok = tok;
int i = 0;
while (iscsym(*p) && i < MAX_IDEN_SIZE)
*ptok++ = *p++;
*ptok = '\0';
return p;
}
/**
* Return a pointer to the specified string, stripped of leading and trailing
* non-C symbol characters.
*
* Note that pszStr may be changed after return, the first trailing non-C symbol
* will be replaced with a null.
*
* @param pszStr The string to strip.
*
* @return A pointer to the start of the stripped string.
*/
static __inline PSZ stripNonCSyms( PSZ pszStr )
{
CHAR *pFront = pszStr;
CHAR *pBack = pszStr + strlen(pszStr);
while ( pBack > pFront && ! iscsym(*pBack) )
pBack--;
if ( ! iscsym(*pBack) )
{
// No C symbols in string.
*pBack = 0x0;
} else
{
*(++pBack) = 0x0;
while ( pFront < pBack && ! iscsym(*pFront) )
pFront++;
}
return pFront;
}
void GetName()
{
unsigned int v = 0;
char *NamePtr = Name;
while (iscsym(*FilePtr))
{
*NamePtr++ = *FilePtr++;
if (v++ >= NAME_MAX)
Error(Error_Fatal, "Symbol too int");
}
*NamePtr++ = 0;
return;
}
bool CSrScriptFile::TokenAlpha (const char*& pParse)
{
const char* pStartToken = pParse;
srscripttoken_t* pToken;
while (*pParse)
{
if (!iscsym(*pParse)) break;
++pParse;
}
pToken = m_Tokens.AddNew();
pToken->Token.Copy(pStartToken, pParse - pStartToken);
pToken->Type = SR_TOKEN_IDENTIFIER;
return true;
}
char * GetSym()
{
unsigned int v = 0;
char *NamePtr = xName;
char c;
while (iscsym(*FilePtr))
{
c = *FilePtr++;
if (c == '.')
c = '_';
*NamePtr++ = c;
if (v++ >= MOSYNC_NAME_MAX)
Error(Error_Fatal, "Symbol too int");
}
*NamePtr++ = 0;
return xName;
}
// ---
bool get_token( const char*& s, CStr& t ) {
if ( !s )
return false;
while( isspace(*s) )
s++;
while( skip_comment(s) )
;
switch( *s ) {
case TERM:
return false;
case STAR :
case BRACKET_O :
case BRACKET_C :
case BRACE_O :
case COMMA :
case SEMICOLON :
case COLON :
((char*)memcpy( t=new char[2], s++, 1 ))[1] = 0;
return true;
}
const char* b = s;
while( iscsym(*s) )
s++;
if ( s > b ) {
((char*)memcpy( t=new char[s-b+1], b, s-b ))[s-b] = 0;
return true;
}
else {
t = 0;
return false;
}
}
UBYTE *tokenize_word(register UBYTE *line,/* (in) -> current line position */
register UBYTE *word,/* (in) -> output token buffer */
UBYTE *qstring, /* (in) -> quoted string o/p buffer*/
SHORT *plen, /* (out) # of bytes put in word buf */
SHORT *ttype, /* (out) token type */
Boolean quote /* (in) preserve quotes on string? */
)
{
UBYTE *wrkptr;
UBYTE *sword = word;
int toklen;
SHORT toktype;
register unsigned int c;
UBYTE c1;
UBYTE c2;
/*----------------------------------------------------------------------------
* Skip leading whitespace, get the first character, return NULL if no char.
*--------------------------------------------------------------------------*/
while (isspace(*line) )
++line;
if ('\0' == (c = *line))
{
line = NULL;
goto OUT;
}
/*----------------------------------------------------------------------------
* Handle keywords, types, symbols
*--------------------------------------------------------------------------*/
if (iscsymf(c))
{
line = po_chop_csym(line, word, MAX_SYM_LEN-1, &wrkptr);
word = wrkptr;
toktype = TOK_UNDEF;
}
#ifdef DEADWOOD
else if (iscsymf(c))
{
toktype = TOK_UNDEF;
*word++ = c;
++line;
toklen = MAX_SYM_LEN;
for (;;)
{
c = *line;
if (iscsym(c))
{
++line;
if (toklen)
{
*word++ = c;
--toklen;
}
}
else
break;
}
}
#endif /* DEADWOOD */
/*----------------------------------------------------------------------------
* Handle numeric constants
*--------------------------------------------------------------------------*/
else if (isdigit(c))
{
toklen = get_digits(line,word,&toktype);
line += toklen;
word += toklen;
}
/*----------------------------------------------------------------------------
* Handle C operators and string/char constants...
* This processes non-alphanumeric characters. Most will be passed through
* as single character tokens. Some, like ==, !=, >= and <= are easier to
* handle here than in parser (which only has a one-token look-ahead).
* Also, quoted strings and chars are now handled in this switch statement.
*--------------------------------------------------------------------------*/
else
{
c1 = line[1]; /* lookahead characters */
c2 = line[2];
switch (c)
{
case '"': /* note: a shortcut in this loop */
/* assumes that a quoted string */
if (qstring != NULL) /* could never be longer than */
sword = word = qstring; /* SZTOKE-3 chars. */
toktype = TOK_QUO;
if (quote)
*word++ = '"';
++line;
toklen = SZTOKE-3; /* room for nullterm and two quotes */
while(--toklen)
{
c = *line++;
if (c == 0)
break;
else if (c == '\\')
{
if (quote)
{
*word++ = c;
*word++ = *line++;
}
else
{
wrkptr = line;
*word++ = translate_escape(&wrkptr);
line = wrkptr;
}
}
else if (c == '\"')
{
break;
}
else
{
*word++ = c;
}
}
if (quote)
*word++ = '"';
break;
case '\'':
if (quote)
*word++ = '\'';
toktype = TOK_SQUO;
++line;
for (;;)
{
c = *line++;
if (c == 0)
break;
else if (c == '\\')
{
if (quote)
{
*word++ = c;
*word++ = *line++;
}
else
{
wrkptr = line;
*word++ = translate_escape(&wrkptr);
line = wrkptr;
}
}
else if (c == '\'')
{
break;
}
else
{
*word++ = c;
}
}
if (quote)
*word++ = '\'';
break;
case '.':
if (c1 == '.' && c2 == '.')
{
toktype = TOK_UNDEF;
goto THREECHAR;
}
else
goto SIMPLE;
case '%':
if (c1 == '=') /* mod-equals */
{
toktype = TOK_MOD_EQUALS;
goto TWOCHAR;
}
else
goto SIMPLE;
case '/':
if (c1 == '=') /* div-equals */
{
toktype = TOK_DIV_EQUALS;
goto TWOCHAR;
}
else
goto SIMPLE;
case '*':
if (c1 == '=')
{
toktype = TOK_MUL_EQUALS;
goto TWOCHAR;
}
else
goto SIMPLE;
case '+':
if (c1 == '+')
{
toktype = TOK_PLUS_PLUS;
goto TWOCHAR;
}
else if (c1 == '=')
{
toktype = TOK_PLUS_EQUALS;
goto TWOCHAR;
}
else
goto SIMPLE;
case '-':
if (c1 == '-')
{
toktype = TOK_MINUS_MINUS;
goto TWOCHAR;
}
else if (c1 == '=')
{
toktype = TOK_MINUS_EQUALS;
goto TWOCHAR;
}
else if (c1 == '>')
{
toktype = TOK_ARROW;
goto TWOCHAR;
}
else
goto SIMPLE;
case '=':
if (c1 == '=') /* double equals */
{
toktype = TOK_EQ;
goto TWOCHAR;
}
else
goto SIMPLE;
case '!':
if (c1 == '=') /* != */
{
toktype = TOK_NE;
goto TWOCHAR;
}
else
goto SIMPLE;
case '<':
if (c1 == '=') /* <= */
{
toktype = TOK_LE;
goto TWOCHAR;
}
else if (c1 == '<') /* << */
{
if (c2 == '=')
{
toktype = TOK_LSHIFT_EQUALS;
goto THREECHAR;
}
else
{
toktype = TOK_LSHIFT;
goto TWOCHAR;
}
}
else
goto SIMPLE;
case '>':
if (c1 == '=') /* >= */
{
toktype = TOK_GE;
goto TWOCHAR;
}
else if (c1 == '>') /* >> */
{
if (c2 == '=')
{
toktype = TOK_RSHIFT_EQUALS;
goto THREECHAR;
}
else
{
toktype = TOK_RSHIFT;
goto TWOCHAR;
}
}
else
goto SIMPLE;
case '&':
if (c1 == '&') /* logical and - && */
{
toktype = TOK_LAND;
goto TWOCHAR;
}
else if (c1 == '=')
{
toktype = TOK_AND_EQUALS;
goto TWOCHAR;
}
else
goto SIMPLE;
case '|':
if (c1 == '|') /* logical or - || */
{
toktype = TOK_LOR;
goto TWOCHAR;
}
else if (c1 == '=')
{
toktype = TOK_OR_EQUALS;
goto TWOCHAR;
}
else
goto SIMPLE;
case '^':
if (c1 == '=')
{
toktype = TOK_XOR_EQUALS;
goto TWOCHAR;
}
else
goto SIMPLE;
default:
SIMPLE:
toktype = *word++ = c;
++line;
break;
}
}
OUT:
*ttype = toktype;
*word = 0;
if (plen != NULL)
*plen = word - sword;
return line;
TWOCHAR:
*word++ = c;
*word++ = c1;
line += 2;
goto OUT;
THREECHAR:
*word++ = c;
*word++ = c1;
*word++ = c2;
line += 3;
goto OUT;
}
