////////////////////////////////////////////////////////////////////////////////
// Lexer::Type::tag
// ^ | <isWhiteSpace> [ +|- ] <isIdentifierStart> [ <isIdentifierNext> ]*
bool Lexer::isTag (std::string& token, Lexer::Type& type)
{
std::size_t marker = _cursor;
// Lookbehind: ^ | <isWhiteSpace>
if (marker > 0 &&
! isWhitespace (_text[marker - 1]))
return false;
if (_text[marker] == '+' ||
_text[marker] == '-')
{
++marker;
if (isIdentifierStart (_text[marker]))
{
utf8_next_char (_text, marker);
while (isIdentifierNext (_text[marker]))
utf8_next_char (_text, marker);
token = _text.substr (_cursor, marker - _cursor);
type = Lexer::Type::tag;
_cursor = marker;
return true;
}
}
return false;
}
NABoolean
ComSqlTextHandle::isIdentifierPart(const char &aChar) const
{
if (isIdentifierStart(aChar) OR isDigit(aChar))
{
return TRUE;
}
return FALSE;
}
Token *parseIdentifier( Tokenizer *tokenizer ) {
StatefulString *ss = tokenizer->ss_;
assert( isIdentifierStart( ss, 0 ) );
// Delegate to the less-strict `parseName`
Token *t = parseName( tokenizer );
t->type = ( t->value[ t->length - 1 ] == L'(' ) ? FUNCTION : IDENTIFIER;
return t;
}
int main() {
std::cout << "Ascii letters are identifier start " <<
(isIdentifierStart('a') ? "[OK]" : "[FAIL]") <<
std::endl;
std::cout << "Ascii letters are identifier part " <<
(isIdentifierPart('a') ? "[OK]" : "[FAIL]") <<
std::endl;
std::cout << "Numbers are not identifier start " <<
(!isIdentifierStart('0') ? "[OK]" : "[FAIL]") <<
std::endl;
std::cout << "Numbers are identifier part " <<
(isIdentifierPart('0') ? "[OK]" : "[FAIL]") <<
std::endl;
}
///////////////////////////////////////////////////////////////////////////
//
// Parsing Interface
//
Token *tokenizer_next( Tokenizer *tokenizer ) {
wchar_t c, next;
Token *token = NULL;
next = ss_peek( tokenizer->ss_ );
while ( next != WEOF && !token ) {
// Whitespace
if ( isWhitespaceStart( tokenizer->ss_, 0 ) ) {
token = parseWhitespace( tokenizer );
}
// Strings
else if ( isStringStart( tokenizer->ss_, 0 ) ) {
token = parseString( tokenizer );
}
// Comments
else if ( isCommentStart( tokenizer->ss_, 0 ) ) {
token = parseComment( tokenizer );
}
// URL
else if ( isUrlStart( tokenizer->ss_, 0 ) ) {
token = parseUrl( tokenizer );
}
// SGML Comments
else if ( isSGMLCommentStart( tokenizer->ss_, 0 ) ) {
token = parseSGMLComment( tokenizer );
}
// Identifier
else if ( isIdentifierStart( tokenizer->ss_, 0 ) ) {
token = parseIdentifier( tokenizer );
}
// @keyword
else if ( isAtkeywordStart( tokenizer->ss_, 0 ) ) {
token = parseAtkeyword( tokenizer );
}
// #keyword
else if ( isHashkeywordStart( tokenizer->ss_, 0 ) ) {
token = parseHashkeyword( tokenizer );
}
// Number
else if ( isNumberStart( tokenizer->ss_, 0 ) ) {
token = parseNumber( tokenizer );
}
// Operators & Delims (everything else)
else {
token = parseEverythingElse( tokenizer );
}
}
if ( token ) {
return token;
} else {
return NULL;
}
}
bool tokeniseIdentifier(std::string::const_iterator& s, std::string::const_iterator& e, Token& tok)
{
// Be sure that first char is alphanumeric or _ or $
if ( s==e || !isIdentifierStart(*s) ) return false;
std::string::const_iterator t = s;
while ( s!=e && isIdentifierPart(*++s) );
tok = Token(T_IDENTIFIER, t, s);
return true;
}
Token Lexer::nextToken()
{
if (atEof_) {
return Token(TokenType::END_OF_INPUT, "");
}
if (std::isdigit(next_)) {
return parseNumber();
} else if (isIdentifierStart(next_)) {
return parseIdentifier();
} else if (isEol(next_)) {
return parseNewLine();
} else {
return parseOperator();
}
}
////////////////////////////////////////////////////////////////////////////////
// Lexer::Type::identifier
// <isIdentifierStart> [ <isIdentifierNext> ]*
bool Lexer::isIdentifier (std::string& token, Lexer::Type& type)
{
std::size_t marker = _cursor;
if (isIdentifierStart (_text[marker]))
{
utf8_next_char (_text, marker);
while (isIdentifierNext (_text[marker]))
utf8_next_char (_text, marker);
token = _text.substr (_cursor, marker - _cursor);
type = Lexer::Type::identifier;
_cursor = marker;
return true;
}
return false;
}
Token *parseNumber( Tokenizer *tokenizer ) {
StatefulString *ss = tokenizer->ss_;
assert( isNumberStart( ss, 0 ) );
int start, length;
StatefulStringPosition pos1, pos2;
start = ss->next_index;
length = 0;
pos1 = ss->next_position;
TokenType type = NUMBER;
int isFloat = 0;
wchar_t temp;
while (
isNumeric( ss_peek( ss ) ) ||
( ss_peek( ss ) == L'-' && length == 0 ) ||
( ss_peek( ss ) == L'.' && !isFloat )
) {
temp = ss_getchar( ss );
if ( temp == L'.' ) {
isFloat = 1;
}
length++;
}
// Is the number followed by a percentage?
if ( ss_peek( ss ) == L'%' ) {
ss_getchar( ss );
length++;
type = PERCENTAGE;
}
// Is the number followed by a dimension?
else if ( isIdentifierStart( ss, 0 ) ) {
while ( isNameChar( ss_peek( ss ) ) ) {
ss_getchar( ss );
length++;
}
type = DIMENSION;
}
pos2 = ss->next_position;
return token_new( ss_substr( ss, start, length ), length, type, pos1, pos2 );
}
/* Advances the parser one token, optionally skipping whitespace
* (otherwise it is concatenated and returned as a single whitespace token).
* Whitespace is needed to properly render function signatures. Unrecognized
* token starts are stored literally, e.g. token may equal to a character '#'. */
static int advanceToken (lexerState *lexer, boolean skip_whitspace)
{
boolean have_whitespace = FALSE;
lexer->line = getSourceLineNumber();
lexer->pos = getInputFilePosition();
while (lexer->cur_c != EOF)
{
if (isWhitespace(lexer->cur_c))
{
scanWhitespace(lexer);
have_whitespace = TRUE;
}
else if (lexer->cur_c == '/' && (lexer->next_c == '/' || lexer->next_c == '*'))
{
scanComments(lexer);
have_whitespace = TRUE;
}
else
{
if (have_whitespace && !skip_whitspace)
return lexer->cur_token = TOKEN_WHITESPACE;
break;
}
}
lexer->line = getSourceLineNumber();
lexer->pos = getInputFilePosition();
while (lexer->cur_c != EOF)
{
if (lexer->cur_c == '"')
{
scanString(lexer);
return lexer->cur_token = TOKEN_STRING;
}
else if (lexer->cur_c == 'r' && (lexer->next_c == '#' || lexer->next_c == '"'))
{
scanRawString(lexer);
return lexer->cur_token = TOKEN_STRING;
}
else if (lexer->cur_c == '\'')
{
scanCharacterOrLifetime(lexer);
return lexer->cur_token = TOKEN_STRING;
}
else if (isIdentifierStart(lexer->cur_c))
{
scanIdentifier(lexer);
return lexer->cur_token = TOKEN_IDENT;
}
/* These shift tokens aren't too important for tag-generation per se,
* but they confuse the skipUntil code which tracks the <> pairs. */
else if (lexer->cur_c == '>' && lexer->next_c == '>')
{
advanceNChar(lexer, 2);
return lexer->cur_token = TOKEN_RSHIFT;
}
else if (lexer->cur_c == '<' && lexer->next_c == '<')
{
advanceNChar(lexer, 2);
return lexer->cur_token = TOKEN_LSHIFT;
}
else if (lexer->cur_c == '-' && lexer->next_c == '>')
{
advanceNChar(lexer, 2);
return lexer->cur_token = TOKEN_RARROW;
}
else
{
int c = lexer->cur_c;
advanceChar(lexer);
return lexer->cur_token = c;
}
}
return lexer->cur_token = TOKEN_EOF;
}
bool Lexer::isIdentifierPart(char candidate) const
{
return isIdentifierStart(candidate) || std::isdigit(candidate);
}
int
getToken(LexStream *L, AST **ast_ptr)
{
// int base = 10;
int c;
AST *ast = NULL;
int at_startofline = (L->eoln == 1);
int peekc;
c = skipSpace(L, &ast);
if (c >= 127) {
*ast_ptr = last_ast = ast;
return c;
} else if (safe_isdigit(c)) {
lexungetc(L,c);
ast = NewAST(AST_INTEGER, NULL, NULL);
c = parseNumber(L, 10, &ast->d.ival);
if (c == T_FLOATNUM)
ast->kind = AST_FLOAT;
} else if (c == '$') {
ast = NewAST(AST_INTEGER, NULL, NULL);
c = parseNumber(L, 16, &ast->d.ival);
} else if (c == '%') {
ast = NewAST(AST_INTEGER, NULL, NULL);
c = lexgetc(L);
if (c == '%') {
c = parseNumber(L, 4, &ast->d.ival);
} else {
lexungetc(L, c);
c = parseNumber(L, 2, &ast->d.ival);
}
} else if (isIdentifierStart(c)) {
lexungetc(L, c);
c = parseIdentifier(L, &ast, NULL);
/* if in pasm, and at start of line, restart temporary
labels */
if (c == T_IDENTIFIER && InDatBlock(L) && at_startofline) {
L->lastGlobal = ast->d.string;
}
} else if (c == ':') {
peekc = lexgetc(L);
if (peekc == '=') {
c = T_ASSIGN;
} else if (!gl_p2 && isIdentifierStart(peekc) && InDatBlock(L)) {
lexungetc(L, peekc);
c = parseIdentifier(L, &ast, L->lastGlobal ? L->lastGlobal : "");
} else {
lexungetc(L, peekc);
}
} else if (gl_p2 && c == '.' && isIdentifierStart(lexpeekc(L)) && InDatBlock(L)) {
c = parseIdentifier(L, &ast, L->lastGlobal ? L->lastGlobal : "");
} else if (strchr(operator_chars, c) != NULL) {
char op[6];
int i;
int token;
Symbol *sym = NULL;
op[0] = token = c;
for (i = 1; i < sizeof(op)-1; i++) {
c = lexgetc(L);
if (c >= 128 || strchr(operator_chars, c) == NULL) {
lexungetc(L, c);
break;
}
op[i] = c;
op[i+1] = 0;
sym = FindSymbol(&reservedWords, op);
if (sym) {
token = INTVAL(sym);
} else {
lexungetc(L, c);
break;
}
}
c = token;
} else if (c == '"') {
c = parseString(L, &ast);
}
*ast_ptr = last_ast = ast;
return c;
}
int
isIdentifierChar(int c)
{
return isIdentifierStart(c) || safe_isdigit(c);
}
bool tokenise(std::string::const_iterator& s, std::string::const_iterator& e, Token& tok)
{
std::string::const_iterator t = s;
// Hand constructed state machine recogniser
enum {
START,
REJECT,
IDENTIFIER,
ZERO,
DIGIT,
HEXDIGIT_START,
HEXDIGIT,
OCTDIGIT,
BINDIGIT_START,
BINDIGIT,
DECIMAL_START,
DECIMAL,
EXPONENT_SIGN,
EXPONENT_START,
EXPONENT,
ACCEPT_IDENTIFIER,
ACCEPT_INC,
ACCEPT_NOINC
} state = START;
TokenType tokType = T_EOS;
while (true)
switch (state) {
case START:
if (t==e) {tok = Token(T_EOS, s, "<END>"); return true;}
else if (std::isspace(*t)) {++t; ++s; continue;}
else switch (*t) {
case '(': tokType = T_LPAREN; state = ACCEPT_INC; continue;
case ')': tokType = T_RPAREN; state = ACCEPT_INC; continue;
case ',': tokType = T_COMMA; state = ACCEPT_INC; continue;
case '+': tokType = T_PLUS; state = ACCEPT_INC; continue;
case '-': tokType = T_MINUS; state = ACCEPT_INC; continue;
case '*': tokType = T_MULT; state = ACCEPT_INC; continue;
case '/': tokType = T_DIV; state = ACCEPT_INC; continue;
case '=': tokType = T_EQUAL; state = ACCEPT_INC; continue;
case '<':
++t;
if (t==e || (*t!='>' && *t!='='))
{tokType = T_LESS; state = ACCEPT_NOINC; continue; }
else
{tokType = (*t=='>') ? T_NEQ : T_LSEQ; state = ACCEPT_INC; continue; }
case '>':
++t;
if (t==e || *t!='=')
{tokType = T_GRT; state = ACCEPT_NOINC; continue;}
else
{tokType = T_GREQ; state = ACCEPT_INC; continue;}
default:
break;
}
if (isIdentifierStart(*t)) {++t; state = IDENTIFIER;}
else if (*t=='\'') {return processString(s, e, '\'', T_STRING, tok);}
else if (*t=='\"') {return processString(s, e, '\"', T_IDENTIFIER, tok);}
else if (*t=='0') {++t; state = ZERO;}
else if (std::isdigit(*t)) {++t; state = DIGIT;}
else if (*t=='.') {++t; state = DECIMAL_START;}
else state = REJECT;
continue;
case IDENTIFIER:
if (t==e) {state = ACCEPT_IDENTIFIER;}
else if (isIdentifierPart(*t)) {++t; state = IDENTIFIER;}
else state = ACCEPT_IDENTIFIER;
continue;
case DECIMAL_START:
if (t==e) {state = REJECT;}
else if (std::isdigit(*t)) {++t; state = DECIMAL;}
else state = REJECT;
continue;
case EXPONENT_SIGN:
if (t==e) {state = REJECT;}
else if (*t=='-' || *t=='+') {++t; state = EXPONENT_START;}
else if (std::isdigit(*t)) {++t; state = EXPONENT;}
else state = REJECT;
continue;
case EXPONENT_START:
if (t==e) {state = REJECT;}
else if (std::isdigit(*t)) {++t; state = EXPONENT;}
else state = REJECT;
continue;
case ZERO:
if (t==e) {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
else if (*t=='.') {++t; state = DECIMAL;}
else if (*t=='x' || *t=='X') {++t; state = HEXDIGIT_START;}
else if (*t=='b' || *t=='B') {++t; state = BINDIGIT_START;}
else state = OCTDIGIT;
continue;
case HEXDIGIT_START:
if (t==e) {state = REJECT;}
else if (std::isxdigit(*t)) {++t; state = HEXDIGIT;}
else state = REJECT;
continue;
case HEXDIGIT:
if (t==e) {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
else if (*t=='l' || *t=='L') {tokType = T_NUMERIC_EXACT; state = ACCEPT_INC;}
else if (std::isxdigit(*t) || *t=='_') {++t; state = HEXDIGIT;}
else if (*t=='p' || *t=='P') {++t; state = EXPONENT_SIGN;}
else {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
continue;
case BINDIGIT_START:
if (t==e) {state = REJECT;}
else if (*t=='0' || *t=='1') {++t; state = BINDIGIT;}
else state = REJECT;
continue;
case BINDIGIT:
if (t==e) {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
else if (*t=='l' || *t=='L') {tokType = T_NUMERIC_EXACT; state = ACCEPT_INC;}
else if (*t=='0' || *t=='1' || *t=='_') {++t; state = BINDIGIT;}
else {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
continue;
case OCTDIGIT:
if (t==e) {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
else if (*t=='l' || *t=='L') {tokType = T_NUMERIC_EXACT; state = ACCEPT_INC;}
else if ((std::isdigit(*t) && *t<'8') || *t=='_') {++t; state = OCTDIGIT;}
else {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
continue;
case DIGIT:
if (t==e) {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
else if (*t=='l' || *t=='L') {tokType = T_NUMERIC_EXACT; state = ACCEPT_INC;}
else if (*t=='f' || *t=='F' || *t=='d' || *t=='D') {tokType = T_NUMERIC_APPROX; state = ACCEPT_INC;}
else if (std::isdigit(*t) || *t=='_') {++t; state = DIGIT;}
else if (*t=='.') {++t; state = DECIMAL;}
else if (*t=='e' || *t=='E') {++t; state = EXPONENT_SIGN;}
else {tokType = T_NUMERIC_EXACT; state = ACCEPT_NOINC;}
continue;
case DECIMAL:
if (t==e) {tokType = T_NUMERIC_APPROX; state = ACCEPT_NOINC;}
else if (std::isdigit(*t) || *t=='_') {++t; state = DECIMAL;}
else if (*t=='e' || *t=='E') {++t; state = EXPONENT_SIGN;}
else if (*t=='f' || *t=='F' || *t=='d' || *t=='D') {tokType = T_NUMERIC_APPROX; state = ACCEPT_INC;}
else {tokType = T_NUMERIC_APPROX; state = ACCEPT_NOINC;}
continue;
case EXPONENT:
if (t==e) {tokType = T_NUMERIC_APPROX; state = ACCEPT_NOINC;}
else if (std::isdigit(*t)) {++t; state = EXPONENT;}
else if (*t=='f' || *t=='F' || *t=='d' || *t=='D') {tokType = T_NUMERIC_APPROX; state = ACCEPT_INC;}
else {tokType = T_NUMERIC_APPROX; state = ACCEPT_NOINC;}
continue;
case ACCEPT_INC:
++t;
case ACCEPT_NOINC:
tok = Token(tokType, s, t);
s = t;
return true;
case ACCEPT_IDENTIFIER:
tok = Token(T_IDENTIFIER, s, t);
s = t;
tokeniseReservedWord(tok);
return true;
case REJECT:
return false;
};
}
int nextLexeme(char *s, int *length, int *type){
char *p = s;
if(isEndOfInput(*p)){
*length = 0;
*type = LT_INV;
return -1;
} else if(isNewLine(*p)){
*length = 1;
*type = LT_NL;
return 0;
} else if(isWhiteSpace(*p)){
while(isWhiteSpace(*p)){
p++;
}
*length = p - s;
*type = LT_WS;
return 0;
} else if(isNumberBegining(*p, *(p+1))){
while(isPPNumber(*p)){
if(isPPNumberExp(*p) && isPPNumberExpSign(*(p+1))){
p += 2;
} else {
p++;
}
}
*length = p - s;
*type = LT_NUM;
return 0;
} else if(isIdentifierStart(*p)){
while(isIdentifier(*p)){
p++;
}
*length = p - s;
*type = LT_ID;
return 0;
} else if(*p == '\''){
p++;
while(isSourceCharSet(*p) && *p != '\''){
if(!(*p == '\\' && *(p+1) == '\'')){
p++;
} else {
p += 2;
}
}
if(*p == '\''){
p++;
} else {
fprintf(stderr, "invalid token: %02x\n", *p);
exit(1);
}
*length = p - s;
*type = LT_CC;
return 0;
} else if(*p == '\"'){
p++;
while(isSourceCharSet(*p) && *p != '\"'){
if(!(*p == '\\' && *(p+1) == '\"')){
p++;
} else {
p += 2;
}
}
if(*p == '\"'){
p++;
} else {
fprintf(stderr, "invalid string literal token\n");
exit(1);
}
*length = p - s;
*type = LT_STRLIT;
return 0;
} else if(*p == '('){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == ')'){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == ','){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == ';'){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == '?'){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == '['){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == ']'){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == '{'){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == '}'){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == '~'){
p++;
*length = 1;
*type = LT_PUNC;
return 0;
} else if(*p == '!'){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '%'){
p++;
if(*p == ':'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '>'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '&'){
p++;
if(*p == '&'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '*'){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '+'){
p++;
if(*p == '+'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '-'){
p++;
if(*p == '-'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '>'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '.' && isDigit){
p++;
if(*p == '.' && *(p+1) == '.'){
p += 2;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '/'){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == ':'){
p++;
if(*p == '>'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '<'){
p++;
if(*p == '%'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == ':'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '<'){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '='){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '>'){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '>'){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '^'){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else if(*p == '|'){
p++;
if(*p == '='){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else if(*p == '|'){
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
} else {
*length = p - s;
*type = LT_PUNC;
return 0;
}
} else {
p++;
*length = p - s;
*type = LT_PUNC;
return 0;
}
return 0;
}