void LX_Fetch (void)
{
int c;
pr_tokenclass = TK_NONE;
pr_token[0] = 0;
if (!pr_file_p)
{
pr_token_type = tt_eof;
return;
}
LexWhitespace();
c = *pr_file_p;
switch (ASCIIToChrCode[c])
{
case CHR_LETTER:
LexName();
return;
case CHR_NUMBER:
pr_token_type = tt_immediate;
pr_immediate_type = &type_float;
pr_immediate._float = LexNumber();
return;
case CHR_DQUOTE:
LexString();
return;
case CHR_SQUOTE:
LexVector();
return;
case CHR_DOLLARSIGN:
LexGrab();
return;
case CHR_EOF:
pr_token_type = tt_eof;
return;
case CHR_SPECIAL:
default:
LexPunctuation();
return;
}
}
static void LexVector (void)
{
int i;
pr_file_p++;
pr_token_type = tt_immediate;
pr_immediate_type = &type_vector;
for (i = 0; i < 3; i++)
{
LexWhitespace();
pr_immediate.vector[i] = LexNumber();
}
LexWhitespace();
if (*pr_file_p != '\'')
{
PR_ParseError("bad vector");
}
pr_file_p++;
}
////////////////////////////////////////////////////////////////////////////////
// NextToken: Scan the input and construct the next token.
// Inputs:
//
// Return Value:
// ExprToken The next token. Type will be Token::Eof on end-of-file,
// or Token::Error if a syntax error was encountered.
////////////////////////////////////////////////////////////////////////////////
RegularExprToken RegularExprLexer::NextToken()
{
// Skip whitespace
while (isspace(current)) {
Skip();
}
// Remember the starting position
int startLine = line;
int startColumn = column;
// Clear out the token
tokenText.erase();
RegularExprToken::Type type = RegularExprToken::Error;
try {
switch (current) {
case 0:
type = RegularExprToken::Eof;
break;
case '-':
type = RegularExprToken::Dash;
Consume();
break;
case '.':
type = RegularExprToken::Wildcard;
Consume();
break;
case '[':
type = RegularExprToken::LeftBracket;
Consume();
break;
case ']':
type = RegularExprToken::RightBracket;
Consume();
break;
case '^':
type = RegularExprToken::Caret;
Consume();
break;
case '(':
type = RegularExprToken::LeftParen;
Consume();
break;
case ')':
type = RegularExprToken::RightParen;
Consume();
break;
case '?':
type = RegularExprToken::Optional;
Consume();
break;
case '|':
type = RegularExprToken::Or;
Consume();
break;
case '*':
type = RegularExprToken::Star;
Consume();
break;
case '+':
type = RegularExprToken::Plus;
Consume();
break;
case '{':
type = RegularExprToken::LeftBrace;
Consume();
break;
case '}':
type = RegularExprToken::RightBrace;
Consume();
break;
case ',':
type = RegularExprToken::Comma;
Consume();
break;
case '=':
type = RegularExprToken::Equal;
Consume();
break;
default:
if (ISDIGIT(current)) {
type = LexNumber();
} else {
type = LexOther();
}
break;
}
} catch (int) {
// Don't get stuck on an error token
NextChar();
}
return RegularExprToken(type, tokenText, startLine, startColumn);
}
static void LexPunctuation (void)
{
pr_token_type = tt_punct;
*(int *)pr_token = 0;
*pr_token = *pr_file_p;
switch (*pr_file_p++)
{
case ';':
pr_tokenclass = TK_SEMICOLON;
return;
case '(':
if (*pr_file_p == '+' && pr_file_p[1] == ')')
{
pr_token[1] = *pr_file_p++;
pr_token[2] = *pr_file_p++;
pr_tokenclass = TK_BITSET;
return;
}
if (*pr_file_p == '-' && pr_file_p[1] == ')')
{
pr_token[1] = *pr_file_p++;
pr_token[2] = *pr_file_p++;
pr_tokenclass = TK_BITCLR;
return;
}
pr_tokenclass = TK_LPAREN;
return;
case ')':
pr_tokenclass = TK_RPAREN;
return;
case ',':
pr_tokenclass = TK_COMMA;
return;
case '+':
if (*pr_file_p == '=')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_ADDASSIGN;
return;
}
if (*pr_file_p == '+')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_INC;
return;
}
pr_tokenclass = TK_PLUS;
return;
case '*':
if (*pr_file_p == '=')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_MULASSIGN;
return;
}
pr_tokenclass = TK_ASTERISK;
return;
case '/':
if (*pr_file_p == '=')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_DIVASSIGN;
return;
}
pr_tokenclass = TK_SLASH;
return;
case '{':
pr_bracelevel++;
pr_tokenclass = TK_LBRACE;
return;
case '}':
pr_bracelevel--;
pr_tokenclass = TK_RBRACE;
return;
case '[':
pr_tokenclass = TK_LBRACKET;
return;
case ']':
pr_tokenclass = TK_RBRACKET;
return;
case ':':
pr_tokenclass = TK_COLON;
return;
case '#':
pr_tokenclass = TK_NUMBERSIGN;
return;
case '=':
if (*pr_file_p == '=')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_EQ;
return;
}
pr_tokenclass = TK_ASSIGN;
return;
case '&':
if (*pr_file_p == '&')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_AND;
return;
}
pr_tokenclass = TK_BITAND;
return;
case '|':
if (*pr_file_p == '|')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_OR;
return;
}
pr_tokenclass = TK_BITOR;
return;
case '!':
if (*pr_file_p == '=')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_NE;
return;
}
pr_tokenclass = TK_NOT;
return;
case '>':
if (*pr_file_p == '=')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_GE;
return;
}
pr_tokenclass = TK_GT;
return;
case '<':
if (*pr_file_p == '=')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_LE;
return;
}
pr_tokenclass = TK_LT;
return;
case '.':
if (ASCIIToChrCode[*pr_file_p] == CHR_NUMBER)
{
pr_file_p--;
pr_token_type = tt_immediate;
pr_immediate_type = &type_float;
pr_immediate._float = LexNumber();
return;
}
if (*pr_file_p == '.')
{
if (pr_file_p[1] == '.')
{
pr_token[1] = *pr_file_p++;
pr_token[2] = *pr_file_p++;
pr_tokenclass = TK_ELLIPSIS;
return;
}
else
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_RANGE;
return;
}
}
pr_tokenclass = TK_PERIOD;
return;
case '-':
if (*pr_file_p == '=')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_SUBASSIGN;
return;
}
if (*pr_file_p == '-')
{
pr_token[1] = *pr_file_p++;
pr_tokenclass = TK_DEC;
return;
}
if (ASCIIToChrCode[*pr_file_p] == CHR_NUMBER)
{
pr_file_p--;
pr_token_type = tt_immediate;
pr_immediate_type = &type_float;
pr_immediate._float = LexNumber();
return;
}
pr_tokenclass = TK_MINUS;
return;
default:
PR_ParseError("unknown punctuation");
}
}
