2 ****************************************************************************
3 * <P> XML.c - implementation file for basic XML parser written in ANSI C++
4 * for portability. It works by using recursion and a node tree for breaking
5 * down the elements of an XML document. </P>
8 * @author Frank Vanden Berghen
12 * If you add "#define STRICT_PARSING", on the first line of this file
13 * the parser will see the following XML-stream:
14 * <a><b>some text</b><b>other text </a>
15 * as an error. Otherwise, this tring will be equivalent to:
16 * <a><b>some text</b><b>other text</b></a>
20 * If you add "#define APPROXIMATE_PARSING" on the first line of this file
21 * the parser will see the following XML-stream:
25 * as equivalent to the following XML-stream:
29 * This can be useful for badly-formed XML-streams but prevent the use
30 * of the following XML-stream (problem is: tags at contiguous levels
31 * have the same names):
40 * If you add "#define _XMLPARSER_NO_MESSAGEBOX_" on the first line of this file
41 * the "openFileHelper" function will always display error messages inside the
42 * console instead of inside a message-box-window. Message-box-windows are
43 * available on windows 9x/NT/2000/XP/Vista only.
46 * Copyright (c) 2002, Frank Vanden Berghen
47 * All rights reserved.
48 * Redistribution and use in source and binary forms, with or without
49 * modification, are permitted provided that the following conditions are met:
51 * * Redistributions of source code must retain the above copyright
52 * notice, this list of conditions and the following disclaimer.
53 * * Redistributions in binary form must reproduce the above copyright
54 * notice, this list of conditions and the following disclaimer in the
55 * documentation and/or other materials provided with the distribution.
56 * * Neither the name of the Frank Vanden Berghen nor the
57 * names of its contributors may be used to endorse or promote products
58 * derived from this software without specific prior written permission.
60 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND ANY
61 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
62 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
63 * DISCLAIMED. IN NO EVENT SHALL THE REGENTS AND CONTRIBUTORS BE LIABLE FOR ANY
64 * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
65 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
66 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
67 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
68 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
69 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
71 ****************************************************************************
73 #ifndef _CRT_SECURE_NO_DEPRECATE
74 #define _CRT_SECURE_NO_DEPRECATE
76 #include "xmlParser.h"
79 //#define _CRTDBG_MAP_ALLOC
82 #define WIN32_LEAN_AND_MEAN
83 #include <Windows.h> // to have IsTextUnicode, MultiByteToWideChar, WideCharToMultiByte to handle unicode files
84 // to have "MessageBoxA" to display error messages for openFilHelper
93 XMLCSTR XMLNode::getVersion() { return _T("v2.23"); }
94 void free_XMLDLL(void *t){free(t);}
96 static char strictUTF8Parsing=1, guessUnicodeChars=1, dropWhiteSpace=1;
98 inline int mmin( const int t1, const int t2 ) { return t1 < t2 ? t1 : t2; }
100 // You can modify the initialization of the variable "XMLClearTags" below
101 // to change the clearTags that are currently recognized by the library.
102 // The number on the second columns is the length of the string inside the
103 // first column. The "<!DOCTYPE" declaration must be the second in the list.
104 static ALLXMLClearTag XMLClearTags[] =
106 { _T("<![CDATA["),9, _T("]]>") },
107 { _T("<!DOCTYPE"),9, _T(">") },
108 { _T("<PRE>") ,5, _T("</PRE>") },
109 { _T("<Script>") ,8, _T("</Script>")},
110 { _T("<!--") ,4, _T("-->") },
113 ALLXMLClearTag* XMLNode::getClearTagTable() { return XMLClearTags; }
115 // You can modify the initialization of the variable "XMLEntities" below
116 // to change the character entities that are currently recognized by the library.
117 // The number on the second columns is the length of the string inside the
118 // first column. Additionally, the syntaxes " " and " " are recognized.
119 typedef struct { XMLCSTR s; int l; XMLCHAR c;} XMLCharacterEntity;
120 static XMLCharacterEntity XMLEntities[] =
122 { _T("&" ), 5, _T('&' )},
123 { _T("<" ), 4, _T('<' )},
124 { _T(">" ), 4, _T('>' )},
125 { _T("""), 6, _T('\"')},
126 { _T("'"), 6, _T('\'')},
130 // When rendering the XMLNode to a string (using the "createXMLString" function),
131 // you can ask for a beautiful formatting. This formatting is using the
132 // following indentation character:
133 #define INDENTCHAR _T('\t')
135 // The following function parses the XML errors into a user friendly string.
136 // You can edit this to change the output language of the library to something else.
137 XMLCSTR XMLNode::getError(XMLError xerror)
141 case eXMLErrorNone: return _T("No error");
142 case eXMLErrorMissingEndTag: return _T("Warning: Unmatched end tag");
143 case eXMLErrorEmpty: return _T("Error: No XML data");
144 case eXMLErrorFirstNotStartTag: return _T("Error: First token not start tag");
145 case eXMLErrorMissingTagName: return _T("Error: Missing start tag name");
146 case eXMLErrorMissingEndTagName: return _T("Error: Missing end tag name");
147 case eXMLErrorNoMatchingQuote: return _T("Error: Unmatched quote");
148 case eXMLErrorUnmatchedEndTag: return _T("Error: Unmatched end tag");
149 case eXMLErrorUnmatchedEndClearTag: return _T("Error: Unmatched clear tag end");
150 case eXMLErrorUnexpectedToken: return _T("Error: Unexpected token found");
151 case eXMLErrorInvalidTag: return _T("Error: Invalid tag found");
152 case eXMLErrorNoElements: return _T("Error: No elements found");
153 case eXMLErrorFileNotFound: return _T("Error: File not found");
154 case eXMLErrorFirstTagNotFound: return _T("Error: First Tag not found");
155 case eXMLErrorUnknownCharacterEntity:return _T("Error: Unknown character entity");
156 case eXMLErrorCharConversionError: return _T("Error: unable to convert between UNICODE and MultiByte chars");
157 case eXMLErrorCannotOpenWriteFile: return _T("Error: unable to open file for writing");
158 case eXMLErrorCannotWriteFile: return _T("Error: cannot write into file");
160 case eXMLErrorBase64DataSizeIsNotMultipleOf4: return _T("Warning: Base64-string length is not a multiple of 4");
161 case eXMLErrorBase64DecodeTruncatedData: return _T("Warning: Base64-string is truncated");
162 case eXMLErrorBase64DecodeIllegalCharacter: return _T("Error: Base64-string contains an illegal character");
163 case eXMLErrorBase64DecodeBufferTooSmall: return _T("Error: Base64 decode output buffer is too small");
165 return _T("Unknown");
168 // Here is an abstraction layer to access some common string manipulation functions.
169 // The abstraction layer is currently working for gcc, Microsoft Visual Studio 6.0,
170 // Microsoft Visual Studio .NET, CC (sun compiler) and Borland C++.
171 // If you plan to "port" the library to a new system/compiler, all you have to do is
172 // to edit the following lines.
173 #ifdef XML_NO_WIDE_CHAR
174 char myIsTextUnicode(const void *b, int len) { return FALSE; }
176 #if defined (UNDER_CE) || !defined(WIN32)
177 char myIsTextUnicode(const void *b, int len) // inspired by the Wine API: RtlIsTextUnicode
180 // for SPARC processors: wchar_t* buffers must always be alligned, otherwise it's a char* buffer.
181 if ((((unsigned long)b)%sizeof(wchar_t))!=0) return FALSE;
183 const wchar_t *s=(const wchar_t*)b;
186 if (len<(int)sizeof(wchar_t)) return FALSE;
189 if (len&1) return FALSE;
191 /* only checks the first 256 characters */
192 len=mmin(256,len/sizeof(wchar_t));
194 // Check for the special byte order:
195 if (*s == 0xFFFE) return FALSE; // IS_TEXT_UNICODE_REVERSE_SIGNATURE;
196 if (*s == 0xFEFF) return TRUE; // IS_TEXT_UNICODE_SIGNATURE
198 // checks for ASCII characters in the UNICODE stream
200 for (i=0; i<len; i++) if (s[i]<=(unsigned short)255) stats++;
201 if (stats>len/2) return TRUE;
203 // Check for UNICODE NULL chars
204 for (i=0; i<len; i++) if (!s[i]) return TRUE;
209 char myIsTextUnicode(const void *b,int l) { return (char)IsTextUnicode((CONST LPVOID)b,l,NULL); };
214 // for Microsoft Visual Studio 6.0 and Microsoft Visual Studio .NET,
216 wchar_t *myMultiByteToWideChar(const char *s,int l)
219 if (strictUTF8Parsing) i=(int)MultiByteToWideChar(CP_UTF8,0 ,s,l,NULL,0);
220 else i=(int)MultiByteToWideChar(CP_ACP ,MB_PRECOMPOSED,s,l,NULL,0);
221 if (i<0) return NULL;
222 wchar_t *d=(wchar_t *)malloc((i+1)*sizeof(XMLCHAR));
223 if (strictUTF8Parsing) i=(int)MultiByteToWideChar(CP_UTF8,0 ,s,l,d,i);
224 else i=(int)MultiByteToWideChar(CP_ACP ,MB_PRECOMPOSED,s,l,d,i);
229 char *myWideCharToMultiByte(const wchar_t *s,int l)
231 UINT codePage=CP_ACP; if (strictUTF8Parsing) codePage=CP_UTF8;
232 int i=(int)WideCharToMultiByte(codePage, // code page
233 0, // performance and mapping flags
234 s, // wide-character string
235 l, // number of chars in string
236 NULL, // buffer for new string
238 NULL, // default for unmappable chars
239 NULL // set when default char used
241 if (i<0) return NULL;
242 char *d=(char*)malloc(i+1);
243 WideCharToMultiByte(codePage, // code page
244 0, // performance and mapping flags
245 s, // wide-character string
246 l, // number of chars in string
247 d, // buffer for new string
249 NULL, // default for unmappable chars
250 NULL // set when default char used
257 int _strnicmp(char *c1, char *c2, int l){ return strnicmp(c1,c2,l);}
261 #ifdef XML_NO_WIDE_CHAR
262 char *myWideCharToMultiByte(const wchar_t *s, int l) { return NULL; }
264 char *myWideCharToMultiByte(const wchar_t *s, int l)
267 int i=(int)wcsrtombs(NULL,&ss,0,NULL);
268 if (i<0) return NULL;
269 char *d=(char *)malloc(i+1);
270 wcsrtombs(d,&s,i,NULL);
276 wchar_t *myMultiByteToWideChar(const char *s, int l)
279 int i=(int)mbsrtowcs(NULL,&ss,0,NULL);
280 if (i<0) return NULL;
281 wchar_t *d=(wchar_t *)malloc((i+1)*sizeof(wchar_t));
282 mbsrtowcs(d,&s,l,NULL);
286 int _tcslen(XMLCSTR c) { return wcslen(c); }
290 int _tcsnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return wsncasecmp(c1,c2,l);}
291 int _tcsicmp(XMLCSTR c1, XMLCSTR c2) { return wscasecmp(c1,c2); }
294 int _tcsnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return wcsncasecmp(c1,c2,l);}
295 int _tcsicmp(XMLCSTR c1, XMLCSTR c2) { return wcscasecmp(c1,c2); }
297 XMLSTR _tcsstr(XMLCSTR c1, XMLCSTR c2) { return (XMLSTR)wcsstr(c1,c2); }
298 XMLSTR _tcscpy(XMLSTR c1, XMLCSTR c2) { return (XMLSTR)wcscpy(c1,c2); }
299 FILE *_tfopen(XMLCSTR filename,XMLCSTR mode)
301 char *filenameAscii=myWideCharToMultiByte(filename,0);
303 if (mode[0]==_T('r')) f=fopen(filenameAscii,"rb");
304 else f=fopen(filenameAscii,"wb");
309 FILE *_tfopen(XMLCSTR filename,XMLCSTR mode) { return fopen(filename,mode); }
310 int _tcslen(XMLCSTR c) { return strlen(c); }
311 int _tcsnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return strncasecmp(c1,c2,l);}
312 int _tcsicmp(XMLCSTR c1, XMLCSTR c2) { return strcasecmp(c1,c2); }
313 XMLSTR _tcsstr(XMLCSTR c1, XMLCSTR c2) { return (XMLSTR)strstr(c1,c2); }
314 XMLSTR _tcscpy(XMLSTR c1, XMLCSTR c2) { return (XMLSTR)strcpy(c1,c2); }
316 int _strnicmp(const char *c1,const char *c2, int l) { return strncasecmp(c1,c2,l);}
319 /////////////////////////////////////////////////////////////////////////
320 // Here start the core implementation of the XMLParser library //
321 /////////////////////////////////////////////////////////////////////////
323 // You should normally not change anything below this point.
324 // For your own information, I suggest that you read the openFileHelper below:
325 XMLNode XMLNode::openFileHelper(XMLCSTR filename, XMLCSTR tag)
327 // guess the value of the global parameter "strictUTF8Parsing"
328 // (the guess is based on the first 200 bytes of the file).
329 FILE *f=_tfopen(filename,_T("rb"));
333 int l=(int)fread(bb,1,200,f);
334 setGlobalOptions(guessUnicodeChars,guessUTF8ParsingParameterValue(bb,l),dropWhiteSpace);
340 XMLNode xnode=XMLNode::parseFile(filename,tag,&pResults);
342 // display error message (if any)
343 if (pResults.error != eXMLErrorNone)
346 char message[2000],*s1=(char*)"",*s3=(char*)""; XMLCSTR s2=_T("");
347 if (pResults.error==eXMLErrorFirstTagNotFound) { s1=(char*)"First Tag should be '"; s2=tag; s3=(char*)"'.\n"; }
350 "XML Parsing error inside file '%S'.\n%S\nAt line %i, column %i.\n%s%S%s"
352 "XML Parsing error inside file '%s'.\n%s\nAt line %i, column %i.\n%s%s%s"
354 ,filename,XMLNode::getError(pResults.error),pResults.nLine,pResults.nColumn,s1,s2,s3);
357 #if defined(WIN32) && !defined(UNDER_CE) && !defined(_XMLPARSER_NO_MESSAGEBOX_)
358 MessageBoxA(NULL,message,"XML Parsing error",MB_OK|MB_ICONERROR|MB_TOPMOST);
360 printf("%s",message);
368 // If "strictUTF8Parsing=0" then we assume that all characters have the same length of 1 byte.
369 // If "strictUTF8Parsing=1" then the characters have different lengths (from 1 byte to 4 bytes).
370 // This table is used as lookup-table to know the length of a character (in byte) based on the
371 // content of the first byte of the character.
372 // (note: if you modify this, you must always have XML_utf8ByteTable[0]=0 ).
373 static const char XML_utf8ByteTable[256] =
375 // 0 1 2 3 4 5 6 7 8 9 a b c d e f
376 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x00
377 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x10
378 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x20
379 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x30
380 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x40
381 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x50
382 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x60
383 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x70End of ASCII range
384 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x80 0x80 to 0xc1 invalid
385 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x90
386 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xa0
387 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xb0
388 1,1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xc0 0xc2 to 0xdf 2 byte
389 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xd0
390 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,// 0xe0 0xe0 to 0xef 3 byte
391 4,4,4,4,4,1,1,1,1,1,1,1,1,1,1,1 // 0xf0 0xf0 to 0xf4 4 byte, 0xf5 and higher invalid
393 static const char XML_asciiByteTable[256] =
395 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
396 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
397 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
398 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
399 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,
400 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1
402 static const char *XML_ByteTable=(const char *)XML_utf8ByteTable; // the default is "strictUTF8Parsing=1"
405 XMLError XMLNode::writeToFile(XMLCSTR filename, const char *encoding, char nFormat) const
408 XMLSTR t=createXMLString(nFormat,&i);
409 FILE *f=_tfopen(filename,_T("wb"));
410 if (!f) return eXMLErrorCannotOpenWriteFile;
412 unsigned char h[2]={ 0xFF, 0xFE };
413 if (!fwrite(h,2,1,f)) return eXMLErrorCannotWriteFile;
414 if (!isDeclaration())
416 if (!fwrite(_T("<?xml version=\"1.0\" encoding=\"utf-16\"?>\n"),sizeof(wchar_t)*40,1,f))
417 return eXMLErrorCannotWriteFile;
420 if (!isDeclaration())
422 if ((!encoding)||(XML_ByteTable==XML_utf8ByteTable))
424 // header so that windows recognize the file as UTF-8:
425 unsigned char h[3]={0xEF,0xBB,0xBF};
426 if (!fwrite(h,3,1,f)) return eXMLErrorCannotWriteFile;
427 if (!fwrite("<?xml version=\"1.0\" encoding=\"utf-8\"?>\n",39,1,f)) return eXMLErrorCannotWriteFile;
430 if (fprintf(f,"<?xml version=\"1.0\" encoding=\"%s\"?>\n",encoding)<0) return eXMLErrorCannotWriteFile;
433 if (XML_ByteTable==XML_utf8ByteTable) // test if strictUTF8Parsing==1"
435 unsigned char h[3]={0xEF,0xBB,0xBF}; if (!fwrite(h,3,1,f)) return eXMLErrorCannotWriteFile;
439 if (!fwrite(t,sizeof(XMLCHAR)*i,1,f)) return eXMLErrorCannotWriteFile;
440 if (fclose(f)!=0) return eXMLErrorCannotWriteFile;
442 return eXMLErrorNone;
445 // Duplicate a given string.
446 XMLSTR stringDup(XMLCSTR lpszData, int cbData)
448 if (lpszData==NULL) return NULL;
451 if (cbData==0) cbData=(int)_tcslen(lpszData);
452 lpszNew = (XMLSTR)malloc((cbData+1) * sizeof(XMLCHAR));
455 memcpy(lpszNew, lpszData, (cbData) * sizeof(XMLCHAR));
456 lpszNew[cbData] = (XMLCHAR)NULL;
461 XMLNode XMLNode::emptyXMLNode;
462 XMLClear XMLNode::emptyXMLClear={ NULL, NULL, NULL};
463 XMLAttribute XMLNode::emptyXMLAttribute={ NULL, NULL};
465 // Enumeration used to decipher what type a token is
466 typedef enum XMLTokenTypeTag
470 eTokenTagStart, /* "<" */
471 eTokenTagEnd, /* "</" */
472 eTokenCloseTag, /* ">" */
473 eTokenEquals, /* "=" */
474 eTokenDeclaration, /* "<?" */
475 eTokenShortHandClose, /* "/>" */
480 // Main structure used for parsing XML
485 int nIndex,nIndexMissigEndTag;
489 XMLCSTR lpNewElement;
496 ALLXMLClearTag *pClr;
500 // Enumeration used when parsing attributes
508 // Enumeration used when parsing elements to dictate whether we are currently
516 // private (used while rendering):
517 XMLSTR toXMLString(XMLSTR dest,XMLCSTR source)
521 XMLCharacterEntity *entity;
527 if (ch==entity->c) {_tcscpy(dest,entity->s); dest+=entity->l; source++; goto out_of_loop1; }
531 *(dest++)=*(source++);
533 switch(XML_ByteTable[(unsigned char)ch])
535 case 4: *(dest++)=*(source++);
536 case 3: *(dest++)=*(source++);
537 case 2: *(dest++)=*(source++);
538 case 1: *(dest++)=*(source++);
548 // private (used while rendering):
549 int lengthXMLString(XMLCSTR source)
552 XMLCharacterEntity *entity;
559 if (ch==entity->c) { r+=entity->l; source++; goto out_of_loop1; }
565 ch=XML_ByteTable[(unsigned char)ch]; r+=ch; source+=ch;
573 XMLSTR toXMLString(XMLCSTR source)
575 XMLSTR dest=(XMLSTR)malloc((lengthXMLString(source)+1)*sizeof(XMLCHAR));
576 return toXMLString(dest,source);
579 XMLSTR toXMLStringFast(XMLSTR *dest,int *destSz, XMLCSTR source)
581 int l=lengthXMLString(source)+1;
582 if (l>*destSz) { *destSz=l; *dest=(XMLSTR)realloc(*dest,l*sizeof(XMLCHAR)); }
583 return toXMLString(*dest,source);
587 XMLSTR fromXMLString(XMLCSTR s, int lo, XML *pXML)
589 // This function is the opposite of the function "toXMLString". It decodes the escape
590 // sequences &, ", ', <, > and replace them by the characters
591 // &,",',<,>. This function is used internally by the XML Parser. All the calls to
592 // the XML library will always gives you back "decoded" strings.
594 // in: string (s) and length (lo) of string
595 // out: new allocated string converted from xml
601 XMLCharacterEntity *entity;
606 if ((lo>2)&&(s[1]==_T('#')))
609 if ((*s==_T('X'))||(*s==_T('x'))) { s++; lo--; }
610 while ((*s)&&(*s!=_T(';'))&&((lo--)>0)) s++;
613 pXML->error=eXMLErrorUnknownCharacterEntity;
622 if ((lo>=entity->l)&&(_tcsnicmp(s,entity->s,entity->l)==0)) { s+=entity->l; lo-=entity->l; break; }
627 pXML->error=eXMLErrorUnknownCharacterEntity;
636 j=XML_ByteTable[(unsigned char)*s]; s+=j; lo-=j; ll+=j-1;
642 d=(XMLSTR)malloc((ll+1)*sizeof(XMLCHAR));
651 if ((*ss==_T('X'))||(*ss==_T('x')))
656 if ((*ss>=_T('0'))&&(*ss<=_T('9'))) j=(j<<4)+*ss-_T('0');
657 else if ((*ss>=_T('A'))&&(*ss<=_T('F'))) j=(j<<4)+*ss-_T('A')+10;
658 else if ((*ss>=_T('a'))&&(*ss<=_T('f'))) j=(j<<4)+*ss-_T('a')+10;
659 else { free((void*)s); pXML->error=eXMLErrorUnknownCharacterEntity;return NULL;}
666 if ((*ss>=_T('0'))&&(*ss<=_T('9'))) j=(j*10)+*ss-_T('0');
667 else { free((void*)s); pXML->error=eXMLErrorUnknownCharacterEntity;return NULL;}
671 (*d++)=(XMLCHAR)j; ss++;
677 if (_tcsnicmp(ss,entity->s,entity->l)==0) { *(d++)=entity->c; ss+=entity->l; break; }
686 switch(XML_ByteTable[(unsigned char)*ss])
688 case 4: *(d++)=*(ss++); ll--;
689 case 3: *(d++)=*(ss++); ll--;
690 case 2: *(d++)=*(ss++); ll--;
691 case 1: *(d++)=*(ss++);
700 #define XML_isSPACECHAR(ch) ((ch==_T('\n'))||(ch==_T(' '))||(ch== _T('\t'))||(ch==_T('\r')))
703 char myTagCompare(XMLCSTR cclose, XMLCSTR copen)
704 // !!!! WARNING strange convention&:
705 // return 0 if equals
706 // return 1 if different
708 if (!cclose) return 1;
709 int l=(int)_tcslen(cclose);
710 if (_tcsnicmp(cclose, copen, l)!=0) return 1;
711 const XMLCHAR c=copen[l];
712 if (XML_isSPACECHAR(c)||
716 (c==_T('=' ))) return 0;
720 // Obtain the next character from the string.
721 static inline XMLCHAR getNextChar(XML *pXML)
723 XMLCHAR ch = pXML->lpXML[pXML->nIndex];
725 if (ch!=0) pXML->nIndex++;
727 pXML->nIndex+=XML_ByteTable[(unsigned char)ch];
732 // Find the next token in a string.
733 // pcbToken contains the number of characters that have been read.
734 static NextToken GetNextToken(XML *pXML, int *pcbToken, enum XMLTokenTypeTag *pType)
739 int indexStart,nFoundMatch,nIsText=FALSE;
740 result.pClr=NULL; // prevent warning
742 // Find next non-white space character
743 do { indexStart=pXML->nIndex; ch=getNextChar(pXML); } while XML_isSPACECHAR(ch);
747 // Cache the current string pointer
748 result.pStr = &pXML->lpXML[indexStart];
750 // First check whether the token is in the clear tag list (meaning it
751 // does not need formatting).
752 ALLXMLClearTag *ctag=XMLClearTags;
755 if (_tcsnicmp(ctag->lpszOpen, result.pStr, ctag->openTagLen)==0)
758 pXML->nIndex+=ctag->openTagLen-1;
763 } while(ctag->lpszOpen);
765 // If we didn't find a clear tag then check for standard tokens
772 *pType = eTokenQuotedText;
778 // Search through the string to find a matching quote
779 while((ch = getNextChar(pXML)))
781 if (ch==chTemp) { nFoundMatch = TRUE; break; }
782 if (ch==_T('<')) break;
785 // If we failed to find a matching quote
786 if (nFoundMatch == FALSE)
788 pXML->nIndex=indexStart+1;
794 // if (FindNonWhiteSpace(pXML)) pXML->nIndex--;
798 // Equals (used with attribute values)
800 *pType = eTokenEquals;
805 *pType = eTokenCloseTag;
808 // Check for tag start and tag end
811 // Peek at the next character to see if we have an end tag '</',
812 // or an xml declaration '<?'
813 chTemp = pXML->lpXML[pXML->nIndex];
815 // If we have a tag end...
816 if (chTemp == _T('/'))
818 // Set the type and ensure we point at the next character
820 *pType = eTokenTagEnd;
823 // If we have an XML declaration tag
824 else if (chTemp == _T('?'))
827 // Set the type and ensure we point at the next character
829 *pType = eTokenDeclaration;
832 // Otherwise we must have a start tag
835 *pType = eTokenTagStart;
839 // Check to see if we have a short hand type end tag ('/>').
842 // Peek at the next character to see if we have a short end tag '/>'
843 chTemp = pXML->lpXML[pXML->nIndex];
845 // If we have a short hand end tag...
846 if (chTemp == _T('>'))
848 // Set the type and ensure we point at the next character
850 *pType = eTokenShortHandClose;
854 // If we haven't found a short hand closing tag then drop into the
862 // If this is a TEXT node
865 // Indicate we are dealing with text
867 while((ch = getNextChar(pXML)))
869 if XML_isSPACECHAR(ch)
873 } else if (ch==_T('/'))
875 // If we find a slash then this maybe text or a short hand end tag
876 // Peek at the next character to see it we have short hand end tag
877 ch=pXML->lpXML[pXML->nIndex];
878 // If we found a short hand end tag then we need to exit the loop
879 if (ch==_T('>')) { pXML->nIndex--; break; }
881 } else if ((ch==_T('<'))||(ch==_T('>'))||(ch==_T('=')))
883 pXML->nIndex--; break;
887 *pcbToken = pXML->nIndex-indexStart;
890 // If we failed to obtain a valid character
892 *pType = eTokenError;
899 XMLCSTR XMLNode::updateName_WOSD(XMLCSTR lpszName)
901 if (d->lpszName&&(lpszName!=d->lpszName)) free((void*)d->lpszName);
902 d->lpszName=lpszName;
907 XMLNode::XMLNode(struct XMLNodeDataTag *p){ d=p; (p->ref_count)++; }
908 XMLNode::XMLNode(XMLNodeData *pParent, XMLCSTR lpszName, char isDeclaration)
910 d=(XMLNodeData*)malloc(sizeof(XMLNodeData));
919 d->isDeclaration = isDeclaration;
921 d->pParent = pParent;
928 updateName_WOSD(lpszName);
931 XMLNode XMLNode::createXMLTopNode_WOSD(XMLCSTR lpszName, char isDeclaration) { return XMLNode(NULL,lpszName,isDeclaration); }
932 XMLNode XMLNode::createXMLTopNode(XMLCSTR lpszName, char isDeclaration) { return XMLNode(NULL,stringDup(lpszName),isDeclaration); }
934 #define MEMORYINCREASE 50
936 static inline void *myRealloc(void *p, int newsize, int memInc, int sizeofElem)
938 if (p==NULL) { if (memInc) return malloc(memInc*sizeofElem); return malloc(sizeofElem); }
939 if ((memInc==0)||((newsize%memInc)==0)) p=realloc(p,(newsize+memInc)*sizeofElem);
942 // printf("XMLParser Error: Not enough memory! Aborting...\n"); exit(220);
948 int XMLNode::findPosition(XMLNodeData *d, int index, XMLElementType xtype)
950 if (index<0) return -1;
951 int i=0,j=(int)((index<<2)+xtype),*o=d->pOrder; while (o[i]!=j) i++; return i;
955 // update "order" information when deleting a content of a XMLNode
956 int XMLNode::removeOrderElement(XMLNodeData *d, XMLElementType t, int index)
958 int n=d->nChild+d->nText+d->nClear, *o=d->pOrder,i=findPosition(d,index,t);
959 memmove(o+i, o+i+1, (n-i)*sizeof(int));
961 if ((o[i]&3)==(int)t) o[i]-=4;
962 // We should normally do:
963 // d->pOrder=(int)realloc(d->pOrder,n*sizeof(int));
964 // but we skip reallocation because it's too time consuming.
965 // Anyway, at the end, it will be free'd completely at once.
969 void *XMLNode::addToOrder(int memoryIncrease,int *_pos, int nc, void *p, int size, XMLElementType xtype)
971 // in: *_pos is the position inside d->pOrder ("-1" means "EndOf")
972 // out: *_pos is the index inside p
973 p=myRealloc(p,(nc+1),memoryIncrease,size);
974 int n=d->nChild+d->nText+d->nClear;
975 d->pOrder=(int*)myRealloc(d->pOrder,n+1,memoryIncrease*3,sizeof(int));
976 int pos=*_pos,*o=d->pOrder;
978 if ((pos<0)||(pos>=n)) { *_pos=nc; o[n]=(int)((nc<<2)+xtype); return p; }
981 memmove(o+i+1, o+i, (n-i)*sizeof(int));
983 while ((pos<n)&&((o[pos]&3)!=(int)xtype)) pos++;
984 if (pos==n) { *_pos=nc; o[n]=(int)((nc<<2)+xtype); return p; }
987 for (i=pos+1;i<=n;i++) if ((o[i]&3)==(int)xtype) o[i]+=4;
990 memmove(((char*)p)+(pos+1)*size,((char*)p)+pos*size,(nc-pos)*size);
995 // Add a child node to the given element.
996 XMLNode XMLNode::addChild_priv(int memoryIncrease, XMLCSTR lpszName, char isDeclaration, int pos)
998 if (!lpszName) return emptyXMLNode;
999 d->pChild=(XMLNode*)addToOrder(memoryIncrease,&pos,d->nChild,d->pChild,sizeof(XMLNode),eNodeChild);
1000 d->pChild[pos].d=NULL;
1001 d->pChild[pos]=XMLNode(d,lpszName,isDeclaration);
1003 return d->pChild[pos];
1006 // Add an attribute to an element.
1007 XMLAttribute *XMLNode::addAttribute_priv(int memoryIncrease,XMLCSTR lpszName, XMLCSTR lpszValuev)
1009 if (!lpszName) return &emptyXMLAttribute;
1010 int nc=d->nAttribute;
1011 d->pAttribute=(XMLAttribute*)myRealloc(d->pAttribute,(nc+1),memoryIncrease,sizeof(XMLAttribute));
1012 XMLAttribute *pAttr=d->pAttribute+nc;
1013 pAttr->lpszName = lpszName;
1014 pAttr->lpszValue = lpszValuev;
1019 // Add text to the element.
1020 XMLCSTR XMLNode::addText_priv(int memoryIncrease, XMLCSTR lpszValue, int pos)
1022 if (!lpszValue) return NULL;
1023 d->pText=(XMLCSTR*)addToOrder(memoryIncrease,&pos,d->nText,d->pText,sizeof(XMLSTR),eNodeText);
1024 d->pText[pos]=lpszValue;
1029 // Add clear (unformatted) text to the element.
1030 XMLClear *XMLNode::addClear_priv(int memoryIncrease, XMLCSTR lpszValue, XMLCSTR lpszOpen, XMLCSTR lpszClose, int pos)
1032 if (!lpszValue) return &emptyXMLClear;
1033 d->pClear=(XMLClear *)addToOrder(memoryIncrease,&pos,d->nClear,d->pClear,sizeof(XMLClear),eNodeClear);
1034 XMLClear *pNewClear=d->pClear+pos;
1035 pNewClear->lpszValue = lpszValue;
1036 if (!lpszOpen) lpszOpen=getClearTagTable()->lpszOpen;
1037 if (!lpszClose) lpszOpen=getClearTagTable()->lpszClose;
1038 pNewClear->lpszOpenTag = lpszOpen;
1039 pNewClear->lpszCloseTag = lpszClose;
1045 // Parse a clear (unformatted) type node.
1046 char XMLNode::parseClearTag(void *px, ALLXMLClearTag *pClear)
1048 XML *pXML=(XML *)px;
1050 XMLCSTR lpszTemp=NULL;
1051 XMLCSTR lpXML=&pXML->lpXML[pXML->nIndex];
1052 static XMLCSTR docTypeEnd=_T("]>");
1054 // Find the closing tag
1055 // Seems the <!DOCTYPE need a better treatment so lets handle it
1056 if (pClear->lpszOpen==XMLClearTags[1].lpszOpen)
1061 if (*pCh==_T('<')) { pClear->lpszClose=docTypeEnd; lpszTemp=_tcsstr(lpXML,docTypeEnd); break; }
1062 else if (*pCh==_T('>')) { lpszTemp=pCh; break; }
1066 pCh+=XML_ByteTable[(unsigned char)(*pCh)];
1069 } else lpszTemp=_tcsstr(lpXML, pClear->lpszClose);
1073 // Cache the size and increment the index
1074 cbTemp = (int)(lpszTemp - lpXML);
1076 pXML->nIndex += cbTemp+(int)_tcslen(pClear->lpszClose);
1078 // Add the clear node to the current element
1079 addClear_priv(MEMORYINCREASE,stringDup(lpXML,cbTemp), pClear->lpszOpen, pClear->lpszClose,-1);
1083 // If we failed to find the end tag
1084 pXML->error = eXMLErrorUnmatchedEndClearTag;
1088 void XMLNode::exactMemory(XMLNodeData *d)
1090 if (d->pOrder) d->pOrder=(int*)realloc(d->pOrder,(d->nChild+d->nText+d->nClear)*sizeof(int));
1091 if (d->pChild) d->pChild=(XMLNode*)realloc(d->pChild,d->nChild*sizeof(XMLNode));
1092 if (d->pAttribute) d->pAttribute=(XMLAttribute*)realloc(d->pAttribute,d->nAttribute*sizeof(XMLAttribute));
1093 if (d->pText) d->pText=(XMLCSTR*)realloc(d->pText,d->nText*sizeof(XMLSTR));
1094 if (d->pClear) d->pClear=(XMLClear *)realloc(d->pClear,d->nClear*sizeof(XMLClear));
1097 char XMLNode::maybeAddTxT(void *pa, XMLCSTR tokenPStr)
1099 XML *pXML=(XML *)pa;
1100 XMLCSTR lpszText=pXML->lpszText;
1101 if (!lpszText) return 0;
1102 if (dropWhiteSpace) while (XML_isSPACECHAR(*lpszText)&&(lpszText!=tokenPStr)) lpszText++;
1103 int cbText = (int)(tokenPStr - lpszText);
1104 if (!cbText) { pXML->lpszText=NULL; return 0; }
1105 if (dropWhiteSpace) { cbText--; while ((cbText)&&XML_isSPACECHAR(lpszText[cbText])) cbText--; cbText++; }
1106 if (!cbText) { pXML->lpszText=NULL; return 0; }
1107 lpszText=fromXMLString(lpszText,cbText,pXML);
1108 if (!lpszText) return 1;
1109 addText_priv(MEMORYINCREASE,lpszText,-1);
1110 pXML->lpszText=NULL;
1114 // Recursively parse an XML element.
1115 int XMLNode::ParseXMLElement(void *pa)
1117 XML *pXML=(XML *)pa;
1119 enum XMLTokenTypeTag type;
1121 XMLCSTR lpszTemp=NULL;
1125 enum Status status; // inside or outside a tag
1126 enum Attrib attrib = eAttribName;
1130 // If this is the first call to the function
1133 // Assume we are outside of a tag definition
1134 pXML->nFirst = FALSE;
1135 status = eOutsideTag;
1138 // If this is not the first call then we should only be called when inside a tag.
1139 status = eInsideTag;
1142 // Iterate through the tokens in the document
1145 // Obtain the next token
1146 token = GetNextToken(pXML, &cbToken, &type);
1148 if (type != eTokenError)
1150 // Check the current status
1154 // If we are outside of a tag definition
1157 // Check what type of token we obtained
1160 // If we have found text or quoted text
1162 case eTokenCloseTag: /* '>' */
1163 case eTokenShortHandClose: /* '/>' */
1164 case eTokenQuotedText:
1168 // If we found a start tag '<' and declarations '<?'
1169 case eTokenTagStart:
1170 case eTokenDeclaration:
1172 // Cache whether this new element is a declaration or not
1173 nDeclaration = (type == eTokenDeclaration);
1175 // If we have node text then add this to the element
1176 if (maybeAddTxT(pXML,token.pStr)) return FALSE;
1178 // Find the name of the tag
1179 token = GetNextToken(pXML, &cbToken, &type);
1181 // Return an error if we couldn't obtain the next token or
1183 if (type != eTokenText)
1185 pXML->error = eXMLErrorMissingTagName;
1189 // If we found a new element which is the same as this
1190 // element then we need to pass this back to the caller..
1192 #ifdef APPROXIMATE_PARSING
1194 myTagCompare(d->lpszName, token.pStr) == 0)
1196 // Indicate to the caller that it needs to create a
1198 pXML->lpNewElement = token.pStr;
1199 pXML->cbNewElement = cbToken;
1204 // If the name of the new element differs from the name of
1205 // the current element we need to add the new element to
1206 // the current one and recurse
1207 pNew = addChild_priv(MEMORYINCREASE,stringDup(token.pStr,cbToken), nDeclaration,-1);
1209 while (!pNew.isEmpty())
1211 // Callself to process the new node. If we return
1212 // FALSE this means we dont have any more
1213 // processing to do...
1215 if (!pNew.ParseXMLElement(pXML)) return FALSE;
1218 // If the call to recurse this function
1219 // evented in a end tag specified in XML then
1220 // we need to unwind the calls to this
1221 // function until we find the appropriate node
1222 // (the element name and end tag name must
1226 // If we are back at the root node then we
1227 // have an unmatched end tag
1230 pXML->error=eXMLErrorUnmatchedEndTag;
1234 // If the end tag matches the name of this
1235 // element then we only need to unwind
1238 if (myTagCompare(d->lpszName, pXML->lpEndTag)==0)
1245 if (pXML->cbNewElement)
1247 // If the call indicated a new element is to
1248 // be created on THIS element.
1250 // If the name of this element matches the
1251 // name of the element we need to create
1252 // then we need to return to the caller
1253 // and let it process the element.
1255 if (myTagCompare(d->lpszName, pXML->lpNewElement)==0)
1260 // Add the new element and recurse
1261 pNew = addChild_priv(MEMORYINCREASE,stringDup(pXML->lpNewElement,pXML->cbNewElement),0,-1);
1262 pXML->cbNewElement = 0;
1266 // If we didn't have a new element to create
1267 pNew = emptyXMLNode;
1275 // If we found an end tag
1278 // If we have node text then add this to the element
1279 if (maybeAddTxT(pXML,token.pStr)) return FALSE;
1281 // Find the name of the end tag
1282 token = GetNextToken(pXML, &cbTemp, &type);
1284 // The end tag should be text
1285 if (type != eTokenText)
1287 pXML->error = eXMLErrorMissingEndTagName;
1290 lpszTemp = token.pStr;
1292 // After the end tag we should find a closing tag
1293 token = GetNextToken(pXML, &cbToken, &type);
1294 if (type != eTokenCloseTag)
1296 pXML->error = eXMLErrorMissingEndTagName;
1299 pXML->lpszText=pXML->lpXML+pXML->nIndex;
1301 // We need to return to the previous caller. If the name
1302 // of the tag cannot be found we need to keep returning to
1303 // caller until we find a match
1304 if (myTagCompare(d->lpszName, lpszTemp) != 0)
1305 #ifdef STRICT_PARSING
1307 pXML->error=eXMLErrorUnmatchedEndTag;
1308 pXML->nIndexMissigEndTag=pXML->nIndex;
1313 pXML->error=eXMLErrorMissingEndTag;
1314 pXML->nIndexMissigEndTag=pXML->nIndex;
1315 pXML->lpEndTag = lpszTemp;
1316 pXML->cbEndTag = cbTemp;
1320 // Return to the caller
1324 // If we found a clear (unformatted) token
1326 // If we have node text then add this to the element
1327 if (maybeAddTxT(pXML,token.pStr)) return FALSE;
1328 if (parseClearTag(pXML, token.pClr)) return FALSE;
1329 pXML->lpszText=pXML->lpXML+pXML->nIndex;
1337 // If we are inside a tag definition we need to search for attributes
1340 // Check what part of the attribute (name, equals, value) we
1344 // If we are looking for a new attribute
1347 // Check what the current token type is
1350 // If the current type is text...
1353 // Cache the token then indicate that we are next to
1354 // look for the equals
1355 lpszTemp = token.pStr;
1357 attrib = eAttribEquals;
1360 // If we found a closing tag...
1362 case eTokenCloseTag:
1363 // We are now outside the tag
1364 status = eOutsideTag;
1365 pXML->lpszText=pXML->lpXML+pXML->nIndex;
1368 // If we found a short hand '/>' closing tag then we can
1369 // return to the caller
1370 case eTokenShortHandClose:
1372 pXML->lpszText=pXML->lpXML+pXML->nIndex;
1376 case eTokenQuotedText: /* '"SomeText"' */
1377 case eTokenTagStart: /* '<' */
1378 case eTokenTagEnd: /* '</' */
1379 case eTokenEquals: /* '=' */
1380 case eTokenDeclaration: /* '<?' */
1382 pXML->error = eXMLErrorUnexpectedToken;
1388 // If we are looking for an equals
1390 // Check what the current token type is
1393 // If the current type is text...
1394 // Eg. 'Attribute AnotherAttribute'
1396 // Add the unvalued attribute to the list
1397 addAttribute_priv(MEMORYINCREASE,stringDup(lpszTemp,cbTemp), NULL);
1398 // Cache the token then indicate. We are next to
1399 // look for the equals attribute
1400 lpszTemp = token.pStr;
1404 // If we found a closing tag 'Attribute >' or a short hand
1405 // closing tag 'Attribute />'
1406 case eTokenShortHandClose:
1407 case eTokenCloseTag:
1408 // If we are a declaration element '<?' then we need
1409 // to remove extra closing '?' if it exists
1410 pXML->lpszText=pXML->lpXML+pXML->nIndex;
1412 if (d->isDeclaration &&
1413 (lpszTemp[cbTemp-1]) == _T('?'))
1420 // Add the unvalued attribute to the list
1421 addAttribute_priv(MEMORYINCREASE,stringDup(lpszTemp,cbTemp), NULL);
1424 // If this is the end of the tag then return to the caller
1425 if (type == eTokenShortHandClose)
1431 // We are now outside the tag
1432 status = eOutsideTag;
1435 // If we found the equals token...
1436 // Eg. 'Attribute ='
1438 // Indicate that we next need to search for the value
1439 // for the attribute
1440 attrib = eAttribValue;
1444 case eTokenQuotedText: /* 'Attribute "InvalidAttr"'*/
1445 case eTokenTagStart: /* 'Attribute <' */
1446 case eTokenTagEnd: /* 'Attribute </' */
1447 case eTokenDeclaration: /* 'Attribute <?' */
1449 pXML->error = eXMLErrorUnexpectedToken;
1455 // If we are looking for an attribute value
1457 // Check what the current token type is
1460 // If the current type is text or quoted text...
1461 // Eg. 'Attribute = "Value"' or 'Attribute = Value' or
1462 // 'Attribute = 'Value''.
1464 case eTokenQuotedText:
1465 // If we are a declaration element '<?' then we need
1466 // to remove extra closing '?' if it exists
1467 if (d->isDeclaration &&
1468 (token.pStr[cbToken-1]) == _T('?'))
1475 // Add the valued attribute to the list
1476 if (type==eTokenQuotedText) { token.pStr++; cbToken-=2; }
1477 XMLCSTR attrVal=token.pStr;
1480 attrVal=fromXMLString(attrVal,cbToken,pXML);
1481 if (!attrVal) return FALSE;
1483 addAttribute_priv(MEMORYINCREASE,stringDup(lpszTemp,cbTemp),attrVal);
1486 // Indicate we are searching for a new attribute
1487 attrib = eAttribName;
1491 case eTokenTagStart: /* 'Attr = <' */
1492 case eTokenTagEnd: /* 'Attr = </' */
1493 case eTokenCloseTag: /* 'Attr = >' */
1494 case eTokenShortHandClose: /* "Attr = />" */
1495 case eTokenEquals: /* 'Attr = =' */
1496 case eTokenDeclaration: /* 'Attr = <?' */
1498 pXML->error = eXMLErrorUnexpectedToken;
1506 // If we failed to obtain the next token
1509 if ((!d->isDeclaration)&&(d->pParent))
1511 #ifdef STRICT_PARSING
1512 pXML->error=eXMLErrorUnmatchedEndTag;
1514 pXML->error=eXMLErrorMissingEndTag;
1516 pXML->nIndexMissigEndTag=pXML->nIndex;
1523 // Count the number of lines and columns in an XML string.
1524 static void CountLinesAndColumns(XMLCSTR lpXML, int nUpto, XMLResults *pResults)
1530 struct XML xml={ lpXML,lpXML, 0, 0, eXMLErrorNone, NULL, 0, NULL, 0, TRUE };
1532 pResults->nLine = 1;
1533 pResults->nColumn = 1;
1534 while (xml.nIndex<nUpto)
1536 ch = getNextChar(&xml);
1537 if (ch != _T('\n')) pResults->nColumn++;
1541 pResults->nColumn=1;
1546 // Parse XML and return the root element.
1547 XMLNode XMLNode::parseString(XMLCSTR lpszXML, XMLCSTR tag, XMLResults *pResults)
1553 pResults->error=eXMLErrorNoElements;
1555 pResults->nColumn=0;
1557 return emptyXMLNode;
1560 XMLNode xnode(NULL,NULL,FALSE);
1561 struct XML xml={ lpszXML, lpszXML, 0, 0, eXMLErrorNone, NULL, 0, NULL, 0, TRUE };
1563 // Create header element
1564 xnode.ParseXMLElement(&xml);
1565 enum XMLError error = xml.error;
1566 if ((xnode.nChildNode()==1)&&(xnode.nElement()==1)) xnode=xnode.getChildNode(); // skip the empty node
1568 // If no error occurred
1569 if ((error==eXMLErrorNone)||(error==eXMLErrorMissingEndTag))
1571 XMLCSTR name=xnode.getName();
1572 if (tag&&_tcslen(tag)&&((!name)||(_tcsicmp(xnode.getName(),tag))))
1576 while (i<xnode.nChildNode())
1578 nodeTmp=xnode.getChildNode(i);
1579 if (_tcsicmp(nodeTmp.getName(),tag)==0) break;
1580 if (nodeTmp.isDeclaration()) { xnode=nodeTmp; i=0; } else i++;
1582 if (i>=xnode.nChildNode())
1586 pResults->error=eXMLErrorFirstTagNotFound;
1588 pResults->nColumn=0;
1590 return emptyXMLNode;
1596 // Cleanup: this will destroy all the nodes
1597 xnode = emptyXMLNode;
1601 // If we have been given somewhere to place results
1604 pResults->error = error;
1606 // If we have an error
1607 if (error!=eXMLErrorNone)
1609 if (error==eXMLErrorMissingEndTag) xml.nIndex=xml.nIndexMissigEndTag;
1610 // Find which line and column it starts on.
1611 CountLinesAndColumns(xml.lpXML, xml.nIndex, pResults);
1617 XMLNode XMLNode::parseFile(XMLCSTR filename, XMLCSTR tag, XMLResults *pResults)
1619 if (pResults) { pResults->nLine=0; pResults->nColumn=0; }
1620 FILE *f=_tfopen(filename,_T("rb"));
1621 if (f==NULL) { if (pResults) pResults->error=eXMLErrorFileNotFound; return emptyXMLNode; }
1622 fseek(f,0,SEEK_END);
1623 int l=ftell(f),headerSz=0;
1624 if (!l) { if (pResults) pResults->error=eXMLErrorEmpty; return emptyXMLNode; }
1625 fseek(f,0,SEEK_SET);
1626 unsigned char *buf=(unsigned char*)malloc(l+1);
1631 if (guessUnicodeChars)
1633 if (!myIsTextUnicode(buf,l))
1635 if ((buf[0]==0xef)&&(buf[1]==0xbb)&&(buf[2]==0xbf)) headerSz=3;
1636 XMLSTR b2=myMultiByteToWideChar((const char*)(buf+headerSz),l-headerSz);
1637 free(buf); buf=(unsigned char*)b2; headerSz=0;
1640 if ((buf[0]==0xef)&&(buf[1]==0xff)) headerSz=2;
1641 if ((buf[0]==0xff)&&(buf[1]==0xfe)) headerSz=2;
1645 if (guessUnicodeChars)
1647 if (myIsTextUnicode(buf,l))
1650 if ((buf[0]==0xef)&&(buf[1]==0xff)) headerSz=2;
1651 if ((buf[0]==0xff)&&(buf[1]==0xfe)) headerSz=2;
1652 char *b2=myWideCharToMultiByte((const wchar_t*)(buf+headerSz),l-headerSz);
1653 free(buf); buf=(unsigned char*)b2; headerSz=0;
1656 if ((buf[0]==0xef)&&(buf[1]==0xbb)&&(buf[2]==0xbf)) headerSz=3;
1661 if (!buf) { if (pResults) pResults->error=eXMLErrorCharConversionError; return emptyXMLNode; }
1662 XMLNode x=parseString((XMLSTR)(buf+headerSz),tag,pResults);
1667 static inline void charmemset(XMLSTR dest,XMLCHAR c,int l) { while (l--) *(dest++)=c; }
1669 // Creates an user friendly XML string from a given element with
1670 // appropriate white space and carriage returns.
1672 // This recurses through all subnodes then adds contents of the nodes to the
1674 int XMLNode::CreateXMLStringR(XMLNodeData *pEntry, XMLSTR lpszMarker, int nFormat)
1679 int nChildFormat=-1;
1680 int nElementI=pEntry->nChild+pEntry->nText+pEntry->nClear;
1685 #define LENSTR(lpsz) (lpsz ? _tcslen(lpsz) : 0)
1687 // If the element has no name then assume this is the head node.
1688 cbElement = (int)LENSTR(pEntry->lpszName);
1693 cb = nFormat == -1 ? 0 : nFormat;
1697 if (cb) charmemset(lpszMarker, INDENTCHAR, sizeof(XMLCHAR)*cb);
1699 lpszMarker[nResult++]=_T('<');
1700 if (pEntry->isDeclaration) lpszMarker[nResult++]=_T('?');
1701 _tcscpy(&lpszMarker[nResult], pEntry->lpszName);
1703 lpszMarker[nResult++]=_T(' ');
1707 nResult+=cbElement+2+cb;
1708 if (pEntry->isDeclaration) nResult++;
1711 // Enumerate attributes and add them to the string
1712 XMLAttribute *pAttr=pEntry->pAttribute;
1713 for (i=0; i<pEntry->nAttribute; i++)
1716 cb = (int)LENSTR(pAttr->lpszName);
1719 if (lpszMarker) _tcscpy(&lpszMarker[nResult], pAttr->lpszName);
1722 if (pAttr->lpszValue)
1724 cb=(int)lengthXMLString(pAttr->lpszValue);
1727 lpszMarker[nResult]=_T('=');
1728 lpszMarker[nResult+1]=_T('"');
1729 if (cb) toXMLString(&lpszMarker[nResult+2],pAttr->lpszValue);
1730 lpszMarker[nResult+cb+2]=_T('"');
1734 if (lpszMarker) lpszMarker[nResult] = _T(' ');
1740 if (pEntry->isDeclaration)
1744 lpszMarker[nResult-1]=_T('?');
1745 lpszMarker[nResult]=_T('>');
1750 if (lpszMarker) lpszMarker[nResult]=_T('\n');
1754 // If there are child nodes we need to terminate the start tag
1757 if (lpszMarker) lpszMarker[nResult-1]=_T('>');
1760 if (lpszMarker) lpszMarker[nResult]=_T('\n');
1766 // Calculate the child format for when we recurse. This is used to
1767 // determine the number of spaces used for prefixes.
1770 if (cbElement&&(!pEntry->isDeclaration)) nChildFormat=nFormat+1;
1771 else nChildFormat=nFormat;
1774 // Enumerate through remaining children
1775 for (i=0; i<nElementI; i++)
1777 j=pEntry->pOrder[i];
1778 switch((XMLElementType)(j&3))
1784 XMLCSTR pChild=pEntry->pText[j>>2];
1785 cb = (int)lengthXMLString(pChild);
1792 charmemset(&lpszMarker[nResult],INDENTCHAR,sizeof(XMLCHAR)*(nFormat + 1));
1793 toXMLString(&lpszMarker[nResult+nFormat+1],pChild);
1794 lpszMarker[nResult+nFormat+1+cb]=_T('\n');
1796 nResult+=cb+nFormat+2;
1799 if (lpszMarker) toXMLString(&lpszMarker[nResult], pChild);
1809 XMLClear *pChild=pEntry->pClear+(j>>2);
1811 cb = (int)LENSTR(pChild->lpszOpenTag);
1818 charmemset(&lpszMarker[nResult], INDENTCHAR, sizeof(XMLCHAR)*(nFormat + 1));
1819 _tcscpy(&lpszMarker[nResult+nFormat+1], pChild->lpszOpenTag);
1821 nResult+=cb+nFormat+1;
1825 if (lpszMarker)_tcscpy(&lpszMarker[nResult], pChild->lpszOpenTag);
1831 cb = (int)LENSTR(pChild->lpszValue);
1834 if (lpszMarker) _tcscpy(&lpszMarker[nResult], pChild->lpszValue);
1838 // "OpenTag Value CloseTag"
1839 cb = (int)LENSTR(pChild->lpszCloseTag);
1842 if (lpszMarker) _tcscpy(&lpszMarker[nResult], pChild->lpszCloseTag);
1848 if (lpszMarker) lpszMarker[nResult] = _T('\n');
1857 // Recursively add child nodes
1858 nResult += CreateXMLStringR(pEntry->pChild[j>>2].d, lpszMarker ? lpszMarker + nResult : 0, nChildFormat);
1865 if ((cbElement)&&(!pEntry->isDeclaration))
1867 // If we have child entries we need to use long XML notation for
1868 // closing the element - "<elementname>blah blah blah</elementname>"
1871 // "</elementname>\0"
1878 charmemset(&lpszMarker[nResult], INDENTCHAR,sizeof(XMLCHAR)*nFormat);
1883 _tcscpy(&lpszMarker[nResult], _T("</"));
1885 _tcscpy(&lpszMarker[nResult], pEntry->lpszName);
1886 nResult += cbElement;
1890 _tcscpy(&lpszMarker[nResult], _T(">"));
1894 _tcscpy(&lpszMarker[nResult], _T(">\n"));
1899 if (nFormat != -1) nResult+=cbElement+4+nFormat;
1900 else nResult+=cbElement+3;
1904 // If there are no children we can use shorthand XML notation -
1911 _tcscpy(&lpszMarker[nResult], _T("/>"));
1916 _tcscpy(&lpszMarker[nResult], _T("/>\n"));
1922 nResult += nFormat == -1 ? 2 : 3;
1932 // Create an XML string
1933 // @param int nFormat - 0 if no formatting is required
1934 // otherwise nonzero for formatted text
1935 // with carriage returns and indentation.
1936 // @param int *pnSize - [out] pointer to the size of the
1937 // returned string not including the
1939 // @return XMLSTR - Allocated XML string, you must free
1940 // this with free().
1941 XMLSTR XMLNode::createXMLString(int nFormat, int *pnSize) const
1943 if (!d) { if (pnSize) *pnSize=0; return NULL; }
1945 XMLSTR lpszResult = NULL;
1948 // Recursively Calculate the size of the XML string
1949 if (!dropWhiteSpace) nFormat=0;
1950 nFormat = nFormat ? 0 : -1;
1951 cbStr = CreateXMLStringR(d, 0, nFormat);
1953 // Alllocate memory for the XML string + the NULL terminator and
1954 // create the recursively XML string.
1955 lpszResult=(XMLSTR)malloc((cbStr+1)*sizeof(XMLCHAR));
1956 CreateXMLStringR(d, lpszResult, nFormat);
1957 if (pnSize) *pnSize = cbStr;
1961 XMLNode::~XMLNode() { deleteNodeContent(); }
1963 int XMLNode::detachFromParent(XMLNodeData *d)
1965 XMLNode *pa=d->pParent->pChild;
1967 while (((void*)(pa[i].d))!=((void*)d)) i++;
1968 d->pParent->nChild--;
1969 if (d->pParent->nChild) memmove(pa+i,pa+i+1,(d->pParent->nChild-i)*sizeof(XMLNode));
1970 else { free(pa); d->pParent->pChild=NULL; }
1971 return removeOrderElement(d->pParent,eNodeChild,i);
1974 void XMLNode::deleteNodeContent(char force)
1978 if ((d->ref_count==0)||force)
1981 if (d->pParent) detachFromParent(d);
1982 for(i=0; i<d->nChild; i++) { d->pChild[i].d->pParent=NULL; d->pChild[i].deleteNodeContent(force); }
1984 for(i=0; i<d->nText; i++) free((void*)d->pText[i]);
1986 for(i=0; i<d->nClear; i++) free((void*)d->pClear[i].lpszValue);
1988 for(i=0; i<d->nAttribute; i++)
1990 free((void*)d->pAttribute[i].lpszName);
1991 if (d->pAttribute[i].lpszValue) free((void*)d->pAttribute[i].lpszValue);
1993 free(d->pAttribute);
1995 free((void*)d->lpszName);
2001 XMLNode XMLNode::addChild(XMLNode childNode, int pos)
2003 XMLNodeData *dc=childNode.d;
2004 if ((!dc)||(!d)) return childNode;
2005 if (dc->pParent) { if ((detachFromParent(dc)<=pos)&&(dc->pParent==d)) pos--; } else dc->ref_count++;
2007 // int nc=d->nChild;
2008 // d->pChild=(XMLNode*)myRealloc(d->pChild,(nc+1),memoryIncrease,sizeof(XMLNode));
2009 d->pChild=(XMLNode*)addToOrder(0,&pos,d->nChild,d->pChild,sizeof(XMLNode),eNodeChild);
2010 d->pChild[pos].d=dc;
2015 void XMLNode::deleteAttribute(int i)
2017 if ((!d)||(i<0)||(i>=d->nAttribute)) return;
2019 XMLAttribute *p=d->pAttribute+i;
2020 free((void*)p->lpszName);
2021 if (p->lpszValue) free((void*)p->lpszValue);
2022 if (d->nAttribute) memmove(p,p+1,(d->nAttribute-i)*sizeof(XMLAttribute)); else { free(p); d->pAttribute=NULL; }
2025 void XMLNode::deleteAttribute(XMLAttribute *a){ if (a) deleteAttribute(a->lpszName); }
2026 void XMLNode::deleteAttribute(XMLCSTR lpszName)
2029 getAttribute(lpszName,&j);
2030 if (j) deleteAttribute(j-1);
2033 XMLAttribute *XMLNode::updateAttribute_WOSD(XMLCSTR lpszNewValue, XMLCSTR lpszNewName,int i)
2035 if (!d) return NULL;
2036 if (i>=d->nAttribute)
2038 if (lpszNewName) return addAttribute_WOSD(lpszNewName,lpszNewValue);
2041 XMLAttribute *p=d->pAttribute+i;
2042 if (p->lpszValue&&p->lpszValue!=lpszNewValue) free((void*)p->lpszValue);
2043 p->lpszValue=lpszNewValue;
2044 if (lpszNewName&&p->lpszName!=lpszNewName) { free((void*)p->lpszName); p->lpszName=lpszNewName; };
2048 XMLAttribute *XMLNode::updateAttribute_WOSD(XMLAttribute *newAttribute, XMLAttribute *oldAttribute)
2050 if (oldAttribute) return updateAttribute_WOSD(newAttribute->lpszValue,newAttribute->lpszName,oldAttribute->lpszName);
2051 return addAttribute_WOSD(newAttribute->lpszName,newAttribute->lpszValue);
2054 XMLAttribute *XMLNode::updateAttribute_WOSD(XMLCSTR lpszNewValue, XMLCSTR lpszNewName,XMLCSTR lpszOldName)
2057 getAttribute(lpszOldName,&j);
2058 if (j) return updateAttribute_WOSD(lpszNewValue,lpszNewName,j-1);
2061 if (lpszNewName) return addAttribute_WOSD(lpszNewName,lpszNewValue);
2062 else return addAttribute_WOSD(stringDup(lpszOldName),lpszNewValue);
2066 int XMLNode::indexText(XMLCSTR lpszValue) const
2070 if (!lpszValue) { if (l) return 0; return -1; }
2071 XMLCSTR *p=d->pText;
2072 for (i=0; i<l; i++) if (lpszValue==p[i]) return i;
2076 void XMLNode::deleteText(int i)
2078 if ((!d)||(i<0)||(i>=d->nText)) return;
2080 XMLCSTR *p=d->pText+i;
2082 if (d->nText) memmove(p,p+1,(d->nText-i)*sizeof(XMLCSTR)); else { free(p); d->pText=NULL; }
2083 removeOrderElement(d,eNodeText,i);
2086 void XMLNode::deleteText(XMLCSTR lpszValue) { deleteText(indexText(lpszValue)); }
2088 XMLCSTR XMLNode::updateText_WOSD(XMLCSTR lpszNewValue, int i)
2090 if (!d) return NULL;
2091 if (i>=d->nText) return addText_WOSD(lpszNewValue);
2092 XMLCSTR *p=d->pText+i;
2093 if (*p!=lpszNewValue) { free((void*)*p); *p=lpszNewValue; }
2094 return lpszNewValue;
2097 XMLCSTR XMLNode::updateText_WOSD(XMLCSTR lpszNewValue, XMLCSTR lpszOldValue)
2099 if (!d) return NULL;
2100 int i=indexText(lpszOldValue);
2101 if (i>=0) return updateText_WOSD(lpszNewValue,i);
2102 return addText_WOSD(lpszNewValue);
2105 void XMLNode::deleteClear(int i)
2107 if ((!d)||(i<0)||(i>=d->nClear)) return;
2109 XMLClear *p=d->pClear+i;
2110 free((void*)p->lpszValue);
2111 if (d->nClear) memmove(p,p+1,(d->nText-i)*sizeof(XMLClear)); else { free(p); d->pClear=NULL; }
2112 removeOrderElement(d,eNodeClear,i);
2115 int XMLNode::indexClear(XMLCSTR lpszValue) const
2119 if (!lpszValue) { if (l) return 0; return -1; }
2120 XMLClear *p=d->pClear;
2121 for (i=0; i<l; i++) if (lpszValue==p[i].lpszValue) return i;
2125 void XMLNode::deleteClear(XMLCSTR lpszValue) { deleteClear(indexClear(lpszValue)); }
2126 void XMLNode::deleteClear(XMLClear *a) { if (a) deleteClear(a->lpszValue); }
2128 XMLClear *XMLNode::updateClear_WOSD(XMLCSTR lpszNewContent, int i)
2130 if (!d) return NULL;
2133 return addClear_WOSD(XMLClearTags[0].lpszOpen,lpszNewContent,XMLClearTags[0].lpszClose);
2135 XMLClear *p=d->pClear+i;
2136 if (lpszNewContent!=p->lpszValue) { free((void*)p->lpszValue); p->lpszValue=lpszNewContent; }
2140 XMLClear *XMLNode::updateClear_WOSD(XMLCSTR lpszNewValue, XMLCSTR lpszOldValue)
2142 if (!d) return NULL;
2143 int i=indexClear(lpszOldValue);
2144 if (i>=0) return updateClear_WOSD(lpszNewValue,i);
2145 return addClear_WOSD(lpszNewValue,XMLClearTags[0].lpszOpen,XMLClearTags[0].lpszClose);
2148 XMLClear *XMLNode::updateClear_WOSD(XMLClear *newP,XMLClear *oldP)
2150 if (oldP) return updateClear_WOSD(newP->lpszValue,oldP->lpszValue);
2154 XMLNode& XMLNode::operator=( const XMLNode& A )
2159 deleteNodeContent();
2161 if (d) (d->ref_count) ++ ;
2166 XMLNode::XMLNode(const XMLNode &A)
2170 if (d) (d->ref_count)++ ;
2173 int XMLNode::nChildNode(XMLCSTR name) const
2176 int i,j=0,n=d->nChild;
2177 XMLNode *pc=d->pChild;
2180 if (_tcsicmp(pc->d->lpszName, name)==0) j++;
2186 XMLNode XMLNode::getChildNode(XMLCSTR name, int *j) const
2188 if (!d) return emptyXMLNode;
2189 int i=0,n=d->nChild;
2191 XMLNode *pc=d->pChild+i;
2194 if (_tcsicmp(pc->d->lpszName, name)==0)
2201 return emptyXMLNode;
2204 XMLNode XMLNode::getChildNode(XMLCSTR name, int j) const
2206 if (!d) return emptyXMLNode;
2208 while (j-->0) getChildNode(name,&i);
2209 return getChildNode(name,&i);
2212 int XMLNode::positionOfText (int i) const { if (i>=d->nText ) i=d->nText-1; return findPosition(d,i,eNodeText ); }
2213 int XMLNode::positionOfClear (int i) const { if (i>=d->nClear) i=d->nClear-1; return findPosition(d,i,eNodeClear); }
2214 int XMLNode::positionOfChildNode(int i) const { if (i>=d->nChild) i=d->nChild-1; return findPosition(d,i,eNodeChild); }
2215 int XMLNode::positionOfText (XMLCSTR lpszValue) const { return positionOfText (indexText (lpszValue)); }
2216 int XMLNode::positionOfClear(XMLCSTR lpszValue) const { return positionOfClear(indexClear(lpszValue)); }
2217 int XMLNode::positionOfClear(XMLClear *a) const { if (a) return positionOfClear(a->lpszValue); return positionOfClear(); }
2218 int XMLNode::positionOfChildNode(XMLNode x) const
2220 if ((!d)||(!x.d)) return -1;
2221 XMLNodeData *dd=x.d;
2222 XMLNode *pc=d->pChild;
2224 while (i--) if (pc[i].d==dd) return findPosition(d,i,eNodeChild);
2227 int XMLNode::positionOfChildNode(XMLCSTR name, int count) const
2229 if (!name) return positionOfChildNode(count);
2231 do { getChildNode(name,&j); if (j<0) return -1; } while (count--);
2232 return findPosition(d,j-1,eNodeChild);
2235 XMLNode XMLNode::getChildNodeWithAttribute(XMLCSTR name,XMLCSTR attributeName,XMLCSTR attributeValue, int *k) const
2243 x=getChildNode(name,&i);
2251 t=x.getAttribute(attributeName,&j);
2252 if (t&&(_tcsicmp(attributeValue,t)==0)) { if (k) *k=i+1; return x; }
2256 if (x.isAttributeSet(attributeName)) { if (k) *k=i+1; return x; }
2259 } while (!x.isEmpty());
2260 return emptyXMLNode;
2263 // Find an attribute on an node.
2264 XMLCSTR XMLNode::getAttribute(XMLCSTR lpszAttrib, int *j) const
2266 if (!d) return NULL;
2267 int i=0,n=d->nAttribute;
2269 XMLAttribute *pAttr=d->pAttribute+i;
2272 if (_tcsicmp(pAttr->lpszName, lpszAttrib)==0)
2275 return pAttr->lpszValue;
2282 char XMLNode::isAttributeSet(XMLCSTR lpszAttrib) const
2284 if (!d) return FALSE;
2285 int i,n=d->nAttribute;
2286 XMLAttribute *pAttr=d->pAttribute;
2289 if (_tcsicmp(pAttr->lpszName, lpszAttrib)==0)
2298 XMLCSTR XMLNode::getAttribute(XMLCSTR name, int j) const
2300 if (!d) return NULL;
2302 while (j-->0) getAttribute(name,&i);
2303 return getAttribute(name,&i);
2306 XMLNodeContents XMLNode::enumContents(int i) const
2309 if (!d) { c.type=eNodeNULL; return c; }
2310 if (i<d->nAttribute)
2312 c.type=eNodeAttribute;
2313 c.attrib=d->pAttribute[i];
2317 c.type=(XMLElementType)(d->pOrder[i]&3);
2318 i=(d->pOrder[i])>>2;
2321 case eNodeChild: c.child = d->pChild[i]; break;
2322 case eNodeText: c.text = d->pText[i]; break;
2323 case eNodeClear: c.clear = d->pClear[i]; break;
2329 XMLCSTR XMLNode::getName() const { if (!d) return NULL; return d->lpszName; }
2330 int XMLNode::nText() const { if (!d) return 0; return d->nText; }
2331 int XMLNode::nChildNode() const { if (!d) return 0; return d->nChild; }
2332 int XMLNode::nAttribute() const { if (!d) return 0; return d->nAttribute; }
2333 int XMLNode::nClear() const { if (!d) return 0; return d->nClear; }
2334 int XMLNode::nElement() const { if (!d) return 0; return d->nAttribute+d->nChild+d->nText+d->nClear; }
2335 XMLClear XMLNode::getClear (int i) const { if ((!d)||(i>=d->nClear )) return emptyXMLClear; return d->pClear[i]; }
2336 XMLAttribute XMLNode::getAttribute (int i) const { if ((!d)||(i>=d->nAttribute)) return emptyXMLAttribute; return d->pAttribute[i]; }
2337 XMLCSTR XMLNode::getAttributeName (int i) const { if ((!d)||(i>=d->nAttribute)) return NULL; return d->pAttribute[i].lpszName; }
2338 XMLCSTR XMLNode::getAttributeValue(int i) const { if ((!d)||(i>=d->nAttribute)) return NULL; return d->pAttribute[i].lpszValue; }
2339 XMLCSTR XMLNode::getText (int i) const { if ((!d)||(i>=d->nText )) return NULL; return d->pText[i]; }
2340 XMLNode XMLNode::getChildNode (int i) const { if ((!d)||(i>=d->nChild )) return emptyXMLNode; return d->pChild[i]; }
2341 XMLNode XMLNode::getParentNode ( ) const { if ((!d)||(!d->pParent )) return emptyXMLNode; return XMLNode(d->pParent); }
2342 char XMLNode::isDeclaration ( ) const { if (!d) return 0; return d->isDeclaration; }
2343 char XMLNode::isEmpty ( ) const { return (d==NULL); }
2345 XMLNode XMLNode::addChild(XMLCSTR lpszName, char isDeclaration, int pos)
2346 { return addChild_priv(0,stringDup(lpszName),isDeclaration,pos); }
2347 XMLNode XMLNode::addChild_WOSD(XMLCSTR lpszName, char isDeclaration, int pos)
2348 { return addChild_priv(0,lpszName,isDeclaration,pos); }
2349 XMLAttribute *XMLNode::addAttribute(XMLCSTR lpszName, XMLCSTR lpszValue)
2350 { return addAttribute_priv(0,stringDup(lpszName),stringDup(lpszValue)); }
2351 XMLAttribute *XMLNode::addAttribute_WOSD(XMLCSTR lpszName, XMLCSTR lpszValuev)
2352 { return addAttribute_priv(0,lpszName,lpszValuev); }
2353 XMLCSTR XMLNode::addText(XMLCSTR lpszValue, int pos)
2354 { return addText_priv(0,stringDup(lpszValue),pos); }
2355 XMLCSTR XMLNode::addText_WOSD(XMLCSTR lpszValue, int pos)
2356 { return addText_priv(0,lpszValue,pos); }
2357 XMLClear *XMLNode::addClear(XMLCSTR lpszValue, XMLCSTR lpszOpen, XMLCSTR lpszClose, int pos)
2358 { return addClear_priv(0,stringDup(lpszValue),lpszOpen,lpszClose,pos); }
2359 XMLClear *XMLNode::addClear_WOSD(XMLCSTR lpszValue, XMLCSTR lpszOpen, XMLCSTR lpszClose, int pos)
2360 { return addClear_priv(0,lpszValue,lpszOpen,lpszClose,pos); }
2361 XMLCSTR XMLNode::updateName(XMLCSTR lpszName)
2362 { return updateName_WOSD(stringDup(lpszName)); }
2363 XMLAttribute *XMLNode::updateAttribute(XMLAttribute *newAttribute, XMLAttribute *oldAttribute)
2364 { return updateAttribute_WOSD(stringDup(newAttribute->lpszValue),stringDup(newAttribute->lpszName),oldAttribute->lpszName); }
2365 XMLAttribute *XMLNode::updateAttribute(XMLCSTR lpszNewValue, XMLCSTR lpszNewName,int i)
2366 { return updateAttribute_WOSD(stringDup(lpszNewValue),stringDup(lpszNewName),i); }
2367 XMLAttribute *XMLNode::updateAttribute(XMLCSTR lpszNewValue, XMLCSTR lpszNewName,XMLCSTR lpszOldName)
2368 { return updateAttribute_WOSD(stringDup(lpszNewValue),stringDup(lpszNewName),lpszOldName); }
2369 XMLCSTR XMLNode::updateText(XMLCSTR lpszNewValue, int i)
2370 { return updateText_WOSD(stringDup(lpszNewValue),i); }
2371 XMLCSTR XMLNode::updateText(XMLCSTR lpszNewValue, XMLCSTR lpszOldValue)
2372 { return updateText_WOSD(stringDup(lpszNewValue),lpszOldValue); }
2373 XMLClear *XMLNode::updateClear(XMLCSTR lpszNewContent, int i)
2374 { return updateClear_WOSD(stringDup(lpszNewContent),i); }
2375 XMLClear *XMLNode::updateClear(XMLCSTR lpszNewValue, XMLCSTR lpszOldValue)
2376 { return updateClear_WOSD(stringDup(lpszNewValue),lpszOldValue); }
2377 XMLClear *XMLNode::updateClear(XMLClear *newP,XMLClear *oldP)
2378 { return updateClear_WOSD(stringDup(newP->lpszValue),oldP->lpszValue); }
2380 void XMLNode::setGlobalOptions(char _guessUnicodeChars, char _strictUTF8Parsing, char _dropWhiteSpace)
2382 guessUnicodeChars=_guessUnicodeChars; dropWhiteSpace=_dropWhiteSpace; strictUTF8Parsing=_strictUTF8Parsing;
2384 if (_strictUTF8Parsing) XML_ByteTable=XML_utf8ByteTable; else XML_ByteTable=XML_asciiByteTable;
2388 char XMLNode::guessUTF8ParsingParameterValue(void *buf,int l, char useXMLEncodingAttribute)
2394 if (myIsTextUnicode(buf,l)) return 0;
2395 unsigned char *b=(unsigned char*)buf;
2396 if ((b[0]==0xef)&&(b[1]==0xbb)&&(b[2]==0xbf)) return 1;
2398 // Match utf-8 model ?
2401 switch (XML_utf8ByteTable[b[i]])
2403 case 4: i++; if ((i<l)&&(b[i]& 0xC0)!=0x80) return 0; // 10bbbbbb ?
2404 case 3: i++; if ((i<l)&&(b[i]& 0xC0)!=0x80) return 0; // 10bbbbbb ?
2405 case 2: i++; if ((i<l)&&(b[i]& 0xC0)!=0x80) return 0; // 10bbbbbb ?
2409 if (!useXMLEncodingAttribute) return 1;
2410 // if encoding is specified and different from utf-8 than it's non-utf8
2411 // otherwise it's utf-8
2414 memcpy(bb,buf,l); // copy buf into bb to be able to do "bb[l]=0"
2416 b=(unsigned char*)strstr(bb,"encoding");
2418 b+=8; while XML_isSPACECHAR(*b) b++; if (*b!='=') return 1;
2419 b++; while XML_isSPACECHAR(*b) b++; if ((*b!='\'')&&(*b!='"')) return 1;
2420 b++; while XML_isSPACECHAR(*b) b++; if ((_strnicmp((char*)b,"utf-8",5)==0)||
2421 (_strnicmp((char*)b,"utf8",4)==0)) return 1;
2425 #undef XML_isSPACECHAR
2427 //////////////////////////////////////////////////////////
2428 // Here starts the base64 conversion functions. //
2429 //////////////////////////////////////////////////////////
2431 static const char base64Fillchar = _T('='); // used to mark partial words at the end
2433 // this lookup table defines the base64 encoding
2434 XMLCSTR base64EncodeTable=_T("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/");
2436 // Decode Table gives the index of any valid base64 character in the Base64 table]
2437 // 96: '=' - 97: space char - 98: illegal char - 99: end of string
2438 const unsigned char base64DecodeTable[] = {
2439 99,98,98,98,98,98,98,98,98,97, 97,98,98,97,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, //00 -29
2440 98,98,97,98,98,98,98,98,98,98, 98,98,98,62,98,98,98,63,52,53, 54,55,56,57,58,59,60,61,98,98, //30 -59
2441 98,96,98,98,98, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,10,11,12,13,14, 15,16,17,18,19,20,21,22,23,24, //60 -89
2442 25,98,98,98,98,98,98,26,27,28, 29,30,31,32,33,34,35,36,37,38, 39,40,41,42,43,44,45,46,47,48, //90 -119
2443 49,50,51,98,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, //120 -149
2444 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, //150 -179
2445 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, //180 -209
2446 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98,98,98,98,98, //210 -239
2447 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98 //240 -255
2450 XMLParserBase64Tool::~XMLParserBase64Tool(){ freeBuffer(); }
2452 void XMLParserBase64Tool::freeBuffer(){ if (buf) free(buf); buf=NULL; buflen=0; }
2454 int XMLParserBase64Tool::encodeLength(int inlen, char formatted)
2456 unsigned int i=((inlen-1)/3*4+4+1);
2457 if (formatted) i+=inlen/54;
2461 XMLSTR XMLParserBase64Tool::encode(unsigned char *inbuf, unsigned int inlen, char formatted)
2463 int i=encodeLength(inlen,formatted),k=17,eLen=inlen/3,j;
2464 alloc(i*sizeof(XMLCHAR));
2465 XMLSTR curr=(XMLSTR)buf;
2468 // Copy next three bytes into lower 24 bits of int, paying attention to sign.
2469 j=(inbuf[0]<<16)|(inbuf[1]<<8)|inbuf[2]; inbuf+=3;
2470 // Encode the int into four chars
2471 *(curr++)=base64EncodeTable[ j>>18 ];
2472 *(curr++)=base64EncodeTable[(j>>12)&0x3f];
2473 *(curr++)=base64EncodeTable[(j>> 6)&0x3f];
2474 *(curr++)=base64EncodeTable[(j )&0x3f];
2475 if (formatted) { if (!k) { *(curr++)=_T('\n'); k=18; } k--; }
2477 eLen=inlen-eLen*3; // 0 - 2.
2480 *(curr++)=base64EncodeTable[ inbuf[0]>>2 ];
2481 *(curr++)=base64EncodeTable[(inbuf[0]<<4)&0x3F];
2482 *(curr++)=base64Fillchar;
2483 *(curr++)=base64Fillchar;
2486 j=(inbuf[0]<<8)|inbuf[1];
2487 *(curr++)=base64EncodeTable[ j>>10 ];
2488 *(curr++)=base64EncodeTable[(j>> 4)&0x3f];
2489 *(curr++)=base64EncodeTable[(j<< 2)&0x3f];
2490 *(curr++)=base64Fillchar;
2496 unsigned int XMLParserBase64Tool::decodeSize(XMLCSTR data,XMLError *xe)
2498 if (xe) *xe=eXMLErrorNone;
2501 //skip any extra characters (e.g. newlines or spaces)
2505 if (*data>255) { if (xe) *xe=eXMLErrorBase64DecodeIllegalCharacter; return 0; }
2507 c=base64DecodeTable[(unsigned char)(*data)];
2509 else if (c==98) { if (xe) *xe=eXMLErrorBase64DecodeIllegalCharacter; return 0; }
2512 if (xe&&(size%4!=0)) *xe=eXMLErrorBase64DataSizeIsNotMultipleOf4;
2513 if (size==0) return 0;
2514 do { data--; size--; } while(*data==base64Fillchar); size++;
2515 return (unsigned int)((size*3)/4);
2518 unsigned char XMLParserBase64Tool::decode(XMLCSTR data, unsigned char *buf, int len, XMLError *xe)
2520 if (xe) *xe=eXMLErrorNone;
2527 #define BASE64DECODE_READ_NEXT_CHAR(c) \
2529 if (data[i]>255){ c=98; break; } \
2530 c=base64DecodeTable[(unsigned char)data[i++]]; \
2532 if(c==98){ if(xe)*xe=eXMLErrorBase64DecodeIllegalCharacter; return 0; }
2534 #define BASE64DECODE_READ_NEXT_CHAR(c) \
2535 do { c=base64DecodeTable[(unsigned char)data[i++]]; }while (c==97); \
2536 if(c==98){ if(xe)*xe=eXMLErrorBase64DecodeIllegalCharacter; return 0; }
2539 BASE64DECODE_READ_NEXT_CHAR(c)
2540 if (c==99) { return 2; }
2543 if (p==(int)len) return 2;
2544 if (xe) *xe=eXMLErrorBase64DecodeTruncatedData;
2548 BASE64DECODE_READ_NEXT_CHAR(d)
2549 if ((d==99)||(d==96)) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; }
2550 if (p==(int)len) { if (xe) *xe=eXMLErrorBase64DecodeBufferTooSmall; return 0; }
2551 buf[p++]=(c<<2)|((d>>4)&0x3);
2553 BASE64DECODE_READ_NEXT_CHAR(c)
2554 if (c==99) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; }
2557 if (c==96) return 2;
2558 if (xe) *xe=eXMLErrorBase64DecodeBufferTooSmall;
2561 if (c==96) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; }
2562 buf[p++]=((d<<4)&0xf0)|((c>>2)&0xf);
2564 BASE64DECODE_READ_NEXT_CHAR(d)
2565 if (d==99 ) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; }
2568 if (d==96) return 2;
2569 if (xe) *xe=eXMLErrorBase64DecodeBufferTooSmall;
2572 if (d==96) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; }
2573 buf[p++]=((c<<6)&0xc0)|d;
2576 #undef BASE64DECODE_READ_NEXT_CHAR
2578 void XMLParserBase64Tool::alloc(int newsize)
2580 if ((!buf)&&(newsize)) { buf=malloc(newsize); buflen=newsize; return; }
2581 if (newsize>buflen) { buf=realloc(buf,newsize); buflen=newsize; }
2584 unsigned char *XMLParserBase64Tool::decode(XMLCSTR data, int *outlen, XMLError *xe)
2586 if (xe) *xe=eXMLErrorNone;
2587 unsigned int len=decodeSize(data,xe);
2588 if (outlen) *outlen=len;
2589 if (!len) return NULL;
2591 if(!decode(data,(unsigned char*)buf,len,xe)){ return NULL; }
2592 return (unsigned char*)buf;