int XMLDocument::ParseDocument()
{
char Ch;
SkipWhiteSpaces();
while (currentChar!='\0' && !hasError)
{
XMLNodeParser * Node=new XMLNodeParser(this);
Ch=GetCh();
if (Ch=='<')
{
NextChar();
if (Node->ParseNode())
{
if (Node->type!=Type_Unknown) Childs.push_back(Node);
SkipWhiteSpaces();
}
else
{
delete Node;
break;
}
}
else
{
hasError=true;
Error="Document should consist only from nodes rounded with <name /> or <name></name> tags";
break;
}
}
return (int)hasError;
}
/*
================
Lexer::ReadToken
================
*/
const Token *Lexer::ReadToken( void ) {
if ( tokPos == -1 )
tokPos = 0;
else if ( tokIsUnread ) {
tokIsUnread = false;
return &token;
}
while ( bufPos < bufSize ) {
// Single line comments
if ( !lineComment.IsEmpty() &&
String::CmpPrefix((const char *)(buffer+bufPos), lineComment.c_str()) == 0 ) {
if ( !(flags & LEXER_FULL_LINES) && tokPos > 0 ) {
FinishToken();
return &token;
}
bufPos += 2;
if ( tokPos > 0 ) {
SkipRestOfLine();
return &token;
}
SkipRestOfLine();
continue;
}
// Multiline comments
else if ( !blockComment[0].IsEmpty() &&
String::CmpPrefix((const char *)(buffer+bufPos), blockComment[0].c_str()) == 0 ) {
if ( !(flags & LEXER_FULL_LINES) && tokPos > 0 ) {
FinishToken();
return &token;
}
bufPos += 2;
int l = line;
SkipToCommentEnd();
// If changed line during comment, finish the line.
if ( (flags & LEXER_FULL_LINES) && l != line && tokPos > 0 ) {
FinishToken();
return &token;
}
continue;
}
switch( buffer[bufPos] ) {
// Ignore whitespaces
case ' ':
case '\t':
SkipWhiteSpaces();
if ( flags & LEXER_FULL_LINES )
AddToToken( ' ' );
else if ( tokPos > 0 ) {
FinishToken();
return &token;
}
continue;
// Newlines
case '\n':
case '\r':
if ( !tokPos ) {
FinishLine();
continue;
}
FinishLine();
return &token;
// Quoted tokens
case '\"':
case '\'':
if ( flags & LEXER_FULL_LINES ) {
ParseQuoted();
continue;
}
if ( tokPos > 0 ) {
FinishToken();
return &token;
}
ParseQuoted();
return &token;
default:
// Check for single char tokens
if ( (flags & LEXER_FULL_LINES) == 0 && !singleTokenChars.IsEmpty() ) {
char c = buffer[bufPos];
const char *chars = singleTokenChars.c_str();
int num = singleTokenChars.Length();
for( int i=0; i<num; i++ ) {
if ( c != chars[i] )
continue;
if ( tokPos == 0 ) {
AddToToken( buffer[bufPos] );
bufPos++;
}
FinishToken();
return &token;
}
}
// Add this char to the token
AddToToken( buffer[bufPos] );
bufPos++;
continue;
}
}
if ( tokPos > 0 ) {
FinishToken();
return &token;
}
line = 0;
return NULL;
}
wyBool
LexHelper::GetObjectType(wyInt32 *obj)
{
wyChar obj_type[512] = {0};
if(!GetWordLex(obj_type, 512))
return wyFalse;
if(strnicmp(obj_type, "OR", 2) == 0)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
else
{
if(strnicmp(obj_type, "REPLACE", 7) == 0)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
else
return wyFalse;
}
}
if(strnicmp(obj_type, "ALGORITHM=", 10) == 0)
{
if(strlen(obj_type) == 10)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
else if(strnicmp(obj_type, "ALGORITHM", 9) == 0)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
else
{
if(strlen(obj_type) == 1)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
}
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
if(strnicmp(obj_type, "DEFINER=", 8) == 0)
{
if(strlen(obj_type) == 8)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
else if(strnicmp(obj_type, "DEFINER", 7) == 0)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
else
{
if(strlen(obj_type) == 1)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
}
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512))
return wyFalse;
}
if(strnicmp(obj_type, "SQL",3) == 0)
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512)) // get SECURITY key word for views
return wyFalse;
else
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type,512)) // get SQL SECURITY parameter for views
return wyFalse;
else
{
m_curpos = SkipWhiteSpaces();
if(!GetWordLex(obj_type, 512)) // get probable Object type if above keywords are used
return wyFalse;
}
}
}
if((stricmp(obj_type , "Table") == 0))
*obj = OBJ_TABLE;
else if(stricmp(obj_type , "View") == 0)
*obj = OBJ_VIEW;
else if(stricmp(obj_type , "Database") == 0)
*obj = OBJ_DATABASE;
else if(stricmp(obj_type , "Procedure") == 0)
*obj = OBJ_PROCEDURE;
else if(stricmp(obj_type , "Function") == 0)
*obj = OBJ_FUNCTION;
else if(stricmp(obj_type , "Trigger") == 0)
*obj = OBJ_TRIGGER;
else if(stricmp(obj_type , "Event") == 0)
*obj = OBJ_EVENT;
else
return wyFalse;
return wyTrue;
}
wyInt32
LexHelper::StartLex(wyChar * buffer, wyUInt32 dwcurpos, OPDETAILS *opdetails, wyBool isuse)
{
wyChar curr;
wyChar table[SIZE_512] = {0};
wyChar totable[SIZE_512] = {0};
wyBool bop = wyFalse, btype = wyFalse, btable = wyFalse, bif = wyFalse, btotable = wyFalse, bnot = wyFalse;
wyInt32 oper = 0, obj = 0;
const wyChar* tempsubstr;
m_buffer = buffer;
m_curpos = 0;
if(!m_buffer)
return -1;
while(1)
{
curr = *(m_buffer + m_curpos);
if(curr == C_NULL)
break;
m_curpos = SkipWhiteSpaces();
// Get words.
// The function will make a node depending upon of keywords or function name.
if(bop == wyFalse)
{
if(GetOperationType(&oper, isuse) == wyFalse)
break;
bop = wyTrue;
if(isuse == wyTrue)
btype = wyTrue;
}
else if(btype == wyFalse)
{
if(GetObjectType(&obj) == wyFalse)
break;
btype = wyTrue;
} else if(!btable)
{
if(!GetWordLex(table, SIZE_512))
break;
if(stricmp(table, "if")== 0)
{
bif = wyTrue;
}
else if(stricmp(table, "not") == 0 && bif == wyTrue)
{
bnot = wyTrue;
}
else if((stricmp(table, "exists")!= 0)|| bif == wyFalse)
{
btable = wyTrue;
m_all = wyTrue;
}
}
if(m_all == wyTrue && oper == OPER_RENAME && obj == OBJ_TABLE)
{
if(btotable == wyFalse)
{
btotable = wyTrue;
continue;
}
else
{
if(!GetWordLex(totable, SIZE_512))
break;
if(stricmp(totable, "TO")== 0)
continue;
wyString strto(totable);
wyString dbto, tblto;
wyInt32 pos;
if(strto.GetLength() == 0)
break;
pos = strto.Find(".", 0);
if(pos != -1)
{
dbto.SetAs(strto.Substr(0, pos));
if(dbto.GetCharAt(0) == '`')
{
dbto.Erase(0, 1);
dbto.Strip(1);
}
}
else
dbto.SetAs("");
tblto.SetAs(strto.Substr(pos + 1, strto.GetLength()));
if(tblto.GetCharAt(0) == '`')
{
tblto.Erase(0,1);
tblto.Strip(1);
}
strncpy(opdetails->totable, tblto.GetString(), SIZE_512-1);
strncpy(opdetails->todb, dbto.GetString(), SIZE_512-1);
}
}
if(btable)
break;
}
if(!m_all)
{
return 1;
}
else
{
opdetails->op_type =(OPERTYPE)oper;
opdetails->obj_type =(OBJTYPE)obj;
wyString str, db, tbl;
wyInt32 position;
str.SetAs(table);
if(str.GetLength() == 0)
return 1;
position = str.Find(".", 0);
if(position != -1)
{
db.SetAs(str.Substr(0, position));
if(db.GetCharAt(0) == '`')
{
db.Erase(0, 1);
db.Strip(1);
}
}
else
{
position = -1;
db.SetAs("");
}
if((tempsubstr = str.Substr(position + 1, str.GetLength())))
{
tbl.SetAs(tempsubstr);
}
if(tbl.GetCharAt(0) == '`')
{
tbl.Erase(0,1);
tbl.Strip(1);
}
strncpy(opdetails->table, tbl.GetString(), SIZE_512-1);
strncpy(opdetails->db, db.GetString(), SIZE_512-1);
}
return 0;
}
CXML_Element* CXML_Parser::ParseElement(CXML_Element* pParent, bool bStartTag) {
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
if (IsEOF()) {
return nullptr;
}
CFX_ByteString tag_name, tag_space;
bool bEndTag;
GetTagName(tag_space, tag_name, bEndTag, bStartTag);
if (tag_name.IsEmpty() || bEndTag) {
return nullptr;
}
CXML_Element* pElement = new CXML_Element;
pElement->m_pParent = pParent;
pElement->SetTag(tag_space.AsStringC(), tag_name.AsStringC());
do {
CFX_ByteString attr_space, attr_name;
while (m_dwIndex < m_dwBufferSize) {
SkipWhiteSpaces();
if (IsEOF()) {
break;
}
if (!g_FXCRT_XML_IsNameIntro(m_pBuffer[m_dwIndex])) {
break;
}
GetName(attr_space, attr_name);
SkipWhiteSpaces();
if (IsEOF()) {
break;
}
if (m_pBuffer[m_dwIndex] != '=') {
break;
}
m_dwIndex++;
SkipWhiteSpaces();
if (IsEOF()) {
break;
}
CFX_WideString attr_value;
GetAttrValue(attr_value);
pElement->m_AttrMap.SetAt(attr_space, attr_name, attr_value);
}
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
if (m_dwIndex < m_dwBufferSize || IsEOF()) {
break;
}
} while (ReadNextBlock());
SkipWhiteSpaces();
if (IsEOF()) {
return pElement;
}
uint8_t ch = m_pBuffer[m_dwIndex++];
if (ch == '/') {
m_dwIndex++;
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
return pElement;
}
if (ch != '>') {
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
delete pElement;
return nullptr;
}
SkipWhiteSpaces();
if (IsEOF()) {
return pElement;
}
CFX_UTF8Decoder decoder;
CFX_WideTextBuf content;
bool bCDATA = false;
int32_t iState = 0;
do {
while (m_dwIndex < m_dwBufferSize) {
ch = m_pBuffer[m_dwIndex++];
switch (iState) {
case 0:
if (ch == '<') {
iState = 1;
} else if (ch == '&') {
decoder.ClearStatus();
decoder.AppendChar(GetCharRef());
} else {
decoder.Input(ch);
}
break;
case 1:
if (ch == '!') {
iState = 2;
} else if (ch == '?') {
SkipLiterals("?>");
SkipWhiteSpaces();
iState = 0;
} else if (ch == '/') {
CFX_ByteString space, name;
GetName(space, name);
SkipWhiteSpaces();
m_dwIndex++;
iState = 10;
} else {
content << decoder.GetResult();
CFX_WideString dataStr = content.MakeString();
if (!bCDATA && !m_bSaveSpaceChars) {
dataStr.TrimRight(L" \t\r\n");
}
InsertContentSegment(bCDATA, dataStr.AsStringC(), pElement);
content.Clear();
decoder.Clear();
bCDATA = false;
iState = 0;
m_dwIndex--;
CXML_Element* pSubElement = ParseElement(pElement, true);
if (!pSubElement) {
break;
}
pSubElement->m_pParent = pElement;
pElement->m_Children.push_back(
{CXML_Element::Element, pSubElement});
SkipWhiteSpaces();
}
break;
case 2:
if (ch == '[') {
SkipLiterals("]]>");
} else if (ch == '-') {
m_dwIndex++;
SkipLiterals("-->");
} else {
SkipLiterals(">");
}
decoder.Clear();
SkipWhiteSpaces();
iState = 0;
break;
}
if (iState == 10) {
break;
}
}
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
if (iState == 10 || m_dwIndex < m_dwBufferSize || IsEOF()) {
break;
}
} while (ReadNextBlock());
content << decoder.GetResult();
CFX_WideString dataStr = content.MakeString();
if (!m_bSaveSpaceChars) {
dataStr.TrimRight(L" \t\r\n");
}
InsertContentSegment(bCDATA, dataStr.AsStringC(), pElement);
content.Clear();
decoder.Clear();
bCDATA = false;
return pElement;
}
CXML_Element* CXML_Parser::ParseElement(CXML_Element* pParent, FX_BOOL bStartTag)
{
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
if (IsEOF()) {
return NULL;
}
CFX_ByteString tag_name, tag_space;
FX_BOOL bEndTag;
GetTagName(tag_space, tag_name, bEndTag, bStartTag);
if (tag_name.IsEmpty() || bEndTag) {
return NULL;
}
CXML_Element* pElement;
pElement = FX_NEW CXML_Element;
if (pElement) {
pElement->m_pParent = pParent;
pElement->SetTag(tag_space, tag_name);
}
if (!pElement) {
return NULL;
}
do {
CFX_ByteString attr_space, attr_name;
while (m_dwIndex < m_dwBufferSize) {
SkipWhiteSpaces();
if (IsEOF()) {
break;
}
if (!g_FXCRT_XML_IsNameIntro(m_pBuffer[m_dwIndex])) {
break;
}
GetName(attr_space, attr_name);
SkipWhiteSpaces();
if (IsEOF()) {
break;
}
if (m_pBuffer[m_dwIndex] != '=') {
break;
}
m_dwIndex ++;
SkipWhiteSpaces();
if (IsEOF()) {
break;
}
CFX_WideString attr_value;
GetAttrValue(attr_value);
pElement->m_AttrMap.SetAt(attr_space, attr_name, attr_value);
}
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
if (m_dwIndex < m_dwBufferSize || IsEOF()) {
break;
}
} while (ReadNextBlock());
SkipWhiteSpaces();
if (IsEOF()) {
return pElement;
}
FX_BYTE ch = m_pBuffer[m_dwIndex ++];
if (ch == '/') {
m_dwIndex ++;
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
return pElement;
}
if (ch != '>') {
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
delete pElement;
return NULL;
}
SkipWhiteSpaces();
if (IsEOF()) {
return pElement;
}
CFX_UTF8Decoder decoder;
CFX_WideTextBuf content;
FX_BOOL bCDATA = FALSE;
FX_INT32 iState = 0;
do {
while (m_dwIndex < m_dwBufferSize) {
ch = m_pBuffer[m_dwIndex ++];
switch (iState) {
case 0:
if (ch == '<') {
iState = 1;
} else if (ch == '&') {
decoder.ClearStatus();
decoder.AppendChar(GetCharRef());
} else {
decoder.Input(ch);
}
break;
case 1:
if (ch == '!') {
iState = 2;
} else if (ch == '?') {
SkipLiterals(FX_BSTRC("?>"));
SkipWhiteSpaces();
iState = 0;
} else if (ch == '/') {
CFX_ByteString space, name;
GetName(space, name);
SkipWhiteSpaces();
m_dwIndex ++;
iState = 10;
} else {
content << decoder.GetResult();
CFX_WideString dataStr = content.GetWideString();
if (!bCDATA && !m_bSaveSpaceChars) {
dataStr.TrimRight((FX_LPCWSTR)L" \t\r\n");
}
InsertContentSegment(bCDATA, dataStr, pElement);
content.Clear();
decoder.Clear();
bCDATA = FALSE;
iState = 0;
m_dwIndex --;
CXML_Element* pSubElement = ParseElement(pElement, TRUE);
if (pSubElement == NULL) {
break;
}
pSubElement->m_pParent = pElement;
pElement->m_Children.Add((FX_LPVOID)CXML_Element::Element);
pElement->m_Children.Add(pSubElement);
SkipWhiteSpaces();
}
break;
case 2:
if (ch == '[') {
SkipLiterals(FX_BSTRC("]]>"));
} else if (ch == '-') {
m_dwIndex ++;
SkipLiterals(FX_BSTRC("-->"));
} else {
SkipLiterals(FX_BSTRC(">"));
}
decoder.Clear();
SkipWhiteSpaces();
iState = 0;
break;
}
if (iState == 10) {
break;
}
}
m_nOffset = m_nBufferOffset + (FX_FILESIZE)m_dwIndex;
if (iState == 10 || m_dwIndex < m_dwBufferSize || IsEOF()) {
break;
}
} while (ReadNextBlock());
content << decoder.GetResult();
CFX_WideString dataStr = content.GetWideString();
if (!m_bSaveSpaceChars) {
dataStr.TrimRight((FX_LPCWSTR)L" \t\r\n");
}
InsertContentSegment(bCDATA, dataStr, pElement);
content.Clear();
decoder.Clear();
bCDATA = FALSE;
return pElement;
}