/**
* Find the next token in a string.
*/
static XML::NextToken
XML::GetNextToken(Parser *pXML)
{
XML::NextToken result;
const TCHAR *lpXML;
TCHAR ch;
TCHAR temp_ch;
size_t size;
unsigned n;
bool found_match;
bool is_text = false;
// Find next non-white space character
ch = FindNonWhiteSpace(pXML);
if (gcc_unlikely(ch == 0)) {
// If we failed to obtain a valid character
return { nullptr, 0, eTokenError };
}
// Cache the current string pointer
lpXML = pXML->lpXML;
result.pStr = &lpXML[pXML->nIndex - 1];
switch (ch) {
// Check for quotes
case _T('\''):
case _T('\"'):
// Type of token
result.type = eTokenQuotedText;
temp_ch = ch;
n = pXML->nIndex;
// Set the size
size = 1;
found_match = false;
// Search through the string to find a matching quote
while (((ch = GetNextChar(pXML))) != 0) {
size++;
if (ch == temp_ch) {
found_match = true;
break;
}
if (ch == _T('<'))
break;
}
// If we failed to find a matching quote
if (!found_match) {
pXML->nIndex = n;
is_text = true;
break;
}
// 4.02.2002
if (FindNonWhiteSpace(pXML)) {
pXML->nIndex--;
}
break;
// Equals (used with attribute values)
case _T('='):
size = 1;
result.type = eTokenEquals;
break;
// Close tag
case _T('>'):
size = 1;
result.type = eTokenCloseTag;
break;
// Check for tag start and tag end
case _T('<'):
// Peek at the next character to see if we have an end tag '</',
// or an xml declaration '<?'
temp_ch = pXML->lpXML[pXML->nIndex];
// If we have a tag end...
if (temp_ch == _T('/')) {
// Set the type and ensure we point at the next character
GetNextChar(pXML);
result.type = eTokenTagEnd;
size = 2;
}
// If we have an XML declaration tag
else if (temp_ch == _T('?')) {
// Set the type and ensure we point at the next character
GetNextChar(pXML);
result.type = eTokenDeclaration;
size = 2;
}
// Otherwise we must have a start tag
else {
result.type = eTokenTagStart;
size = 1;
}
break;
// Check to see if we have a short hand type end tag ('/>').
case _T('/'):
// Peek at the next character to see if we have a short end tag '/>'
temp_ch = pXML->lpXML[pXML->nIndex];
// If we have a short hand end tag...
if (temp_ch == _T('>')) {
// Set the type and ensure we point at the next character
GetNextChar(pXML);
result.type = eTokenShortHandClose;
size = 2;
break;
}
// If we haven't found a short hand closing tag then drop into the
// text process
#if GCC_CHECK_VERSION(7,0)
[[fallthrough]];
#endif
// Other characters
default:
is_text = true;
}
// If this is a TEXT node
if (is_text) {
// Indicate we are dealing with text
result.type = eTokenText;
size = 1;
bool nExit = false;
while (!nExit && ((ch = GetNextChar(pXML)) != 0)) {
if (IsWhitespaceOrNull(ch))
// Break when we find white space
break;
switch (ch) {
// If we find a slash then this maybe text or a short hand end tag.
case _T('/'):
// Peek at the next character to see it we have short hand end tag
temp_ch = pXML->lpXML[pXML->nIndex];
// If we found a short hand end tag then we need to exit the loop
if (temp_ch == _T('>')) {
pXML->nIndex--; // 03.02.2002
nExit = true;
} else {
size++;
}
break;
// Break when we find a terminator and decrement the index and
// column count so that we are pointing at the right character
// the next time we are called.
case _T('<'):
case _T('>'):
case _T('='):
pXML->nIndex--;
nExit = true;
break;
case 0:
nExit = true;
break;
default:
size++;
}
}
}
result.length = size;
return result;
}
/**
* Find the next token in a string.
*
* @param pcbToken contains the number of characters that have been read
*/
static NextToken
GetNextToken(XML *pXML, size_t *pcbToken, enum TokenTypeTag *pType)
{
NextToken result;
const TCHAR *lpXML;
TCHAR ch;
TCHAR chTemp;
size_t nSize;
bool nFoundMatch;
unsigned n;
bool nIsText = false;
// Find next non-white space character
ch = FindNonWhiteSpace(pXML);
if (gcc_unlikely(ch == 0)) {
// If we failed to obtain a valid character
*pcbToken = 0;
*pType = eTokenError;
result.pStr = NULL;
return result;
}
// Cache the current string pointer
lpXML = pXML->lpXML;
result.pStr = &lpXML[pXML->nIndex - 1];
chTemp = 0;
switch (ch) {
// Check for quotes
case _T('\''):
case _T('\"'):
// Type of token
*pType = eTokenQuotedText;
chTemp = ch;
n = pXML->nIndex;
// Set the size
nSize = 1;
nFoundMatch = false;
// Search through the string to find a matching quote
while (((ch = getNextChar(pXML))) != 0) {
nSize++;
if (ch == chTemp) {
nFoundMatch = true;
break;
}
if (ch == _T('<'))
break;
}
// If we failed to find a matching quote
if (!nFoundMatch) {
pXML->nIndex = n - 1;
ch = getNextChar(pXML);
nIsText = true;
break;
}
// 4.02.2002
if (FindNonWhiteSpace(pXML)) {
pXML->nIndex--;
}
break;
// Equals (used with attribute values)
case _T('='):
nSize = 1;
*pType = eTokenEquals;
break;
// Close tag
case _T('>'):
nSize = 1;
*pType = eTokenCloseTag;
break;
// Check for tag start and tag end
case _T('<'):
// Peek at the next character to see if we have an end tag '</',
// or an xml declaration '<?'
chTemp = pXML->lpXML[pXML->nIndex];
// If we have a tag end...
if (chTemp == _T('/')) {
// Set the type and ensure we point at the next character
getNextChar(pXML);
*pType = eTokenTagEnd;
nSize = 2;
}
// If we have an XML declaration tag
else if (chTemp == _T('?')) {
// Set the type and ensure we point at the next character
getNextChar(pXML);
*pType = eTokenDeclaration;
nSize = 2;
}
// Otherwise we must have a start tag
else {
*pType = eTokenTagStart;
nSize = 1;
}
break;
// Check to see if we have a short hand type end tag ('/>').
case _T('/'):
// Peek at the next character to see if we have a short end tag '/>'
chTemp = pXML->lpXML[pXML->nIndex];
// If we have a short hand end tag...
if (chTemp == _T('>')) {
// Set the type and ensure we point at the next character
getNextChar(pXML);
*pType = eTokenShortHandClose;
nSize = 2;
break;
}
// If we haven't found a short hand closing tag then drop into the
// text process
// Other characters
default:
nIsText = true;
}
// If this is a TEXT node
if (nIsText) {
// Indicate we are dealing with text
*pType = eTokenText;
nSize = 1;
bool nExit = false;
while (!nExit && ((ch = getNextChar(pXML)) != 0)) {
switch (ch) {
// Break when we find white space
case _T('\n'):
case _T(' '):
case _T('\t'):
case _T('\r'):
nExit = true;
break;
// If we find a slash then this maybe text or a short hand end tag.
case _T('/'):
// Peek at the next character to see it we have short hand end tag
chTemp = pXML->lpXML[pXML->nIndex];
// If we found a short hand end tag then we need to exit the loop
if (chTemp == _T('>')) {
pXML->nIndex--; // 03.02.2002
nExit = true;
} else {
nSize++;
}
break;
// Break when we find a terminator and decrement the index and
// column count so that we are pointing at the right character
// the next time we are called.
case _T('<'):
case _T('>'):
case _T('='):
pXML->nIndex--;
nExit = true;
break;
case 0:
nExit = true;
break;
default:
nSize++;
}
}
}
*pcbToken = nSize;
return result;
}
