TToken ASNLexer::LookupToken(void)
{
char c = Char();
switch ( c ) {
case ':':
if ( Char(1) == ':' && Char(2) == '=' ) {
StartToken();
AddChars(3);
return T_DEFINE;
}
return T_SYMBOL;
case '-':
case '+':
if ( IsDigit(Char(1)) ) {
StartToken();
AddChar();
return LookupNumber();
}
return T_SYMBOL;
case '\"':
StartToken();
AddChar();
StartString();
LookupString();
return T_STRING;
case '\'':
StartToken();
AddChar();
return LookupBinHexString();
case '[':
StartToken();
AddChar();
LookupTag();
return T_TAG;
default:
if ( IsDigit(c) ) {
StartToken();
AddChar();
return LookupNumber();
}
else if ( c >= 'a' && c <= 'z' ) {
StartToken();
AddChar();
LookupIdentifier();
return T_IDENTIFIER;
}
else if ( c >= 'A' && c <= 'Z' ) {
StartToken();
AddChar();
LookupIdentifier();
return LookupKeyword();
}
return T_SYMBOL;
}
}
RTFStatus RTFReader::ParseToken(RTFToken& r_token, int4& r_parameter)
{
RTFStatus t_status;
t_status = kRTFStatusSuccess;
while(m_input < m_input_end && (*m_input == '\n' || *m_input == '\r'))
m_input++;
switch(m_input_state)
{
case kRTFInputStateNormal:
{
// MW-2013-07-31: [[ Bug 10972 ]] Don't use '\0' to indicate end
// of RTF, just break directly (Adobe Reader 9 on Mac puts \0 in
// font names sometimes causing premature end of RTF parsing).
if (m_input == m_input_end)
{
r_token = kRTFTokenEnd;
break;
}
int t_char;
t_char = *m_input;
switch(t_char)
{
case '{':
r_token = kRTFTokenBeginGroup;
m_input++;
break;
case '}':
r_token = kRTFTokenEndGroup;
m_input++;
break;
case '\\':
{
m_input++;
if (m_input == m_input_end)
t_char = '\0';
else
t_char = *m_input;
char t_keyword[32];
int t_keyword_length;
t_keyword_length = 0;
char t_parameter[20];
int t_parameter_length;
t_parameter_length = 0;
if (t_char == '\0')
{
// End of stream in a keyword is an error
t_status = kRTFStatusIncompleteKeyword;
}
else if (t_char == '\'')
{
// Handle hex characters here (simplifies skipping)
m_input++;
if (m_input == m_input_end)
t_char = '\0';
else
t_char = *m_input;
int t_first_nibble;
t_status = LookupNibble(t_char, t_first_nibble);
int t_second_nibble;
if (t_status == kRTFStatusSuccess)
{
m_input++;
if (m_input == m_input_end)
t_char = '\0';
else
t_char = *m_input;
t_status = LookupNibble(t_char, t_second_nibble);
}
if (t_status == kRTFStatusSuccess)
{
m_input++;
r_token = kRTFTokenCharacter | kRTFTokenHasParameter;
r_parameter = (t_first_nibble << 4) | t_second_nibble;
}
}
else if (t_char == '{' || t_char == '}' || t_char == '\\')
{
m_input++;
r_token = kRTFTokenCharacter | kRTFTokenHasParameter;
r_parameter = (uint1)t_char;
}
else
{
if ((t_char < 'a' || t_char > 'z' ) && (t_char < 'A' || t_char > 'Z'))
{
// Any non-letter is a control symbol
t_keyword[0] = t_char;
t_keyword_length = 1;
m_input++;
}
else
{
// We have a letter so parse a control word
while((t_char >= 'a' && t_char <= 'z') || (t_char >= 'A' && t_char <= 'Z'))
{
if (t_char == '\0')
{
t_status = kRTFStatusIncompleteKeyword;
break;
}
else if (t_keyword_length == 31)
{
t_status = kRTFStatusKeywordTooLong;
break;
}
else
t_keyword[t_keyword_length++] = t_char;
m_input++;
if (m_input == m_input_end)
t_char = '\0';
else
t_char = *m_input;
}
// Now we parse a parameter if the current char is '-' or a digit
if (t_char == '-' || (t_char >= '0' && t_char <= '9'))
{
t_parameter[t_parameter_length++] = t_char;
m_input++;
if (m_input == m_input_end)
t_char = '\0';
else
t_char = *m_input;
while(t_char >= '0' && t_char <= '9')
{
if (t_char == '\0')
{
t_status = kRTFStatusIncompleteKeyword;
break;
}
else if (t_parameter_length == 19)
{
t_status = kRTFStatusParameterTooLong;
break;
}
else
t_parameter[t_parameter_length++] = t_char;
m_input++;
if (m_input == m_input_end)
t_char = '\0';
else
t_char = *m_input;
}
}
// Now skip the delimiter if its a space
if (t_char == ' ')
m_input++;
}
if (t_status == kRTFStatusSuccess)
t_status = LookupKeyword(t_keyword, t_keyword_length, r_token);
if (t_status == kRTFStatusSuccess && t_parameter_length != 0)
{
r_token |= kRTFTokenHasParameter;
r_parameter = StringToInteger(t_parameter, t_parameter_length);
}
}
}
break;
default:
m_input++;
r_token = kRTFTokenCharacter | kRTFTokenHasParameter;
r_parameter = (int)((unsigned char)t_char);
break;
}
}
break;
case kRTFInputStateBinary:
{
int4 t_char;
if (m_input != m_input_end)
{
t_char = *m_input;
m_input++;
}
m_input_binary_count -= 1;
if (m_input_binary_count == 0)
m_input_state = kRTFInputStateNormal;
r_token = kRTFTokenCharacter | kRTFTokenHasParameter;
r_parameter = (int)((unsigned char)t_char);
}
break;
case kRTFInputStateHex:
{
int4 t_char;
if (m_input != m_input_end)
t_char = *m_input;
else
t_char = '\0';
m_input++;
int t_first_nibble;
t_status = LookupNibble(t_char, t_first_nibble);
int t_second_nibble;
if (t_status == kRTFStatusSuccess)
{
if (m_input != m_input_end)
t_char = *m_input;
else
t_char = '\0';
m_input++;
t_status = LookupNibble(t_char, t_second_nibble);
}
if (t_status == kRTFStatusSuccess)
{
r_token = kRTFTokenCharacter | kRTFTokenHasParameter;
r_parameter = (t_first_nibble << 4) | t_second_nibble;
m_input_state = kRTFInputStateNormal;
}
}
break;
}
return t_status;
}
void
HangStruct::ScanState(const char *code, const char *state)
{
if (code != 0)
{
for (int n = 0; code[n] != NULLC; n++)
{
if (state[n] == Normal)
{
int c = code[n];
if (isName(c)
&& (n == 0 || !isName(code[n-1])))
{
char *name = parseKeyword(code, n);
int findWord = LookupKeyword(name);
TRACE(("lookup '%s' ->%d\n", name, findWord));
do_aggreg = False;
if (findWord >= 0)
{
indent = 0;
until_parn = 1;
if (pIndentWords[findWord].code == oneLine)
stmt_level++;
else
stmt_level = 0;
}
else
{
if (!strcmp(name, "enum"))
until_curl = 1;
if (parn_level == 0)
until_parn = 0;
indent = 1;
}
delete[] name;
}
else if (!isspace(code[n]))
{
if (do_aggreg && code[n] != L_CURL)
do_aggreg = False;
switch (code[n])
{
case '=':
if (parn_level == 0)
do_aggreg = True;
break;
case L_CURL:
curl_level++;
indent = 0;
stmt_level = 0;
until_parn = 0;
if (do_aggreg)
in_aggreg = curl_level;
break;
case R_CURL:
curl_level--;
indent = 0;
stmt_level = 0;
until_curl = 0;
break;
case ':':
// "::" means something different entirely
if (code[n+1] == ':')
{
n++;
}
else
{
indent = 0;
stmt_level = 0;
until_parn = 0;
}
break;
case SEMICOLON:
if (parn_level == 0)
{
indent = 0;
stmt_level = 0;
until_parn = 0;
until_curl = 0;
if (in_aggreg > curl_level)
in_aggreg = 0;
}
break;
case L_PAREN:
parn_level++;
indent = 1;
break;
case R_PAREN:
parn_level--;
if (until_parn && !parn_level)
indent = 0;
else
indent = 1;
break;
case ESCAPE:
break;
default:
indent = 1;
break;
}
}
}
}
}
}
