From e2900ce0a1b11ec212aeaf42bbefb26a54697c25 Mon Sep 17 00:00:00 2001 From: Martin Sustrik Date: Mon, 14 Sep 2009 14:30:15 +0200 Subject: xmlParser added --- foreign/xmlParser/xmlParser.cpp | 2888 +++++++++++++++++++++++++++++++++++++++ foreign/xmlParser/xmlParser.hpp | 762 +++++++++++ 2 files changed, 3650 insertions(+) create mode 100644 foreign/xmlParser/xmlParser.cpp create mode 100644 foreign/xmlParser/xmlParser.hpp diff --git a/foreign/xmlParser/xmlParser.cpp b/foreign/xmlParser/xmlParser.cpp new file mode 100644 index 0000000..2414c01 --- /dev/null +++ b/foreign/xmlParser/xmlParser.cpp @@ -0,0 +1,2888 @@ +/** + **************************************************************************** + *

XML.c - implementation file for basic XML parser written in ANSI C++ + * for portability. It works by using recursion and a node tree for breaking + * down the elements of an XML document.

+ * + * @version V2.39 + * @author Frank Vanden Berghen + * + * NOTE: + * + * If you add "#define STRICT_PARSING", on the first line of this file + * the parser will see the following XML-stream: + * some textother text + * as an error. Otherwise, this tring will be equivalent to: + * some textother text + * + * NOTE: + * + * If you add "#define APPROXIMATE_PARSING" on the first line of this file + * the parser will see the following XML-stream: + * + * + * + * as equivalent to the following XML-stream: + * + * + * + * This can be useful for badly-formed XML-streams but prevent the use + * of the following XML-stream (problem is: tags at contiguous levels + * have the same names): + * + * + * + * + * + * + * NOTE: + * + * If you add "#define _XMLPARSER_NO_MESSAGEBOX_" on the first line of this file + * the "openFileHelper" function will always display error messages inside the + * console instead of inside a message-box-window. Message-box-windows are + * available on windows 9x/NT/2000/XP/Vista only. + * + * Copyright (c) 2002, Frank Vanden Berghen + * All rights reserved. + * + * The following license terms apply to projects that are in some way related to + * the "ZeroMQ project", including applications + * using "ZeroMQ project" and tools developed + * for enhancing "ZeroMQ project". All other projects + * (not related to "ZeroMQ project") have to use this + * code under the Aladdin Free Public License (AFPL) + * See the file "AFPL-license.txt" for more informations about the AFPL license. + * (see http://www.artifex.com/downloads/doc/Public.htm for detailed AFPL terms) + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are met: + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * * Neither the name of Frank Vanden Berghen nor the + * names of its contributors may be used to endorse or promote products + * derived from this software without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY Frank Vanden Berghen ``AS IS'' AND ANY + * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + * DISCLAIMED. IN NO EVENT SHALL BE LIABLE FOR ANY + * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + **************************************************************************** + */ +#ifndef _CRT_SECURE_NO_DEPRECATE +#define _CRT_SECURE_NO_DEPRECATE +#endif +#include +#ifdef _XMLWINDOWS +//#ifdef _DEBUG +//#define _CRTDBG_MAP_ALLOC +//#include +//#endif +#define WIN32_LEAN_AND_MEAN +#include // to have IsTextUnicode, MultiByteToWideChar, WideCharToMultiByte to handle unicode files + // to have "MessageBoxA" to display error messages for openFilHelper +#endif + +#include +#include +#include +#include +#include + +XMLCSTR XMLNode::getVersion() { return _CXML("v2.39"); } +void freeXMLString(XMLSTR t){if(t)free(t);} + +static XMLNode::XMLCharEncoding characterEncoding=XMLNode::char_encoding_UTF8; +static char guessWideCharChars=1, dropWhiteSpace=1, removeCommentsInMiddleOfText=1; + +inline int mmin( const int t1, const int t2 ) { return t1 < t2 ? t1 : t2; } + +// You can modify the initialization of the variable "XMLClearTags" below +// to change the clearTags that are currently recognized by the library. +// The number on the second columns is the length of the string inside the +// first column. The "") }, + { _CXML("") }, + { _CXML("") }, + { _CXML("
")    ,5,  _CXML("
") }, +// { _CXML("")}, + { NULL ,0, NULL } +}; + +// You can modify the initialization of the variable "XMLEntities" below +// to change the character entities that are currently recognized by the library. +// The number on the second columns is the length of the string inside the +// first column. Additionally, the syntaxes " " and " " are recognized. +typedef struct { XMLCSTR s; int l; XMLCHAR c;} XMLCharacterEntity; +static XMLCharacterEntity XMLEntities[] = +{ + { _CXML("&" ), 5, _CXML('&' )}, + { _CXML("<" ), 4, _CXML('<' )}, + { _CXML(">" ), 4, _CXML('>' )}, + { _CXML("""), 6, _CXML('\"')}, + { _CXML("'"), 6, _CXML('\'')}, + { NULL , 0, '\0' } +}; + +// When rendering the XMLNode to a string (using the "createXMLString" function), +// you can ask for a beautiful formatting. This formatting is using the +// following indentation character: +#define INDENTCHAR _CXML('\t') + +// The following function parses the XML errors into a user friendly string. +// You can edit this to change the output language of the library to something else. +XMLCSTR XMLNode::getError(XMLError xerror) +{ + switch (xerror) + { + case eXMLErrorNone: return _CXML("No error"); + case eXMLErrorMissingEndTag: return _CXML("Warning: Unmatched end tag"); + case eXMLErrorNoXMLTagFound: return _CXML("Warning: No XML tag found"); + case eXMLErrorEmpty: return _CXML("Error: No XML data"); + case eXMLErrorMissingTagName: return _CXML("Error: Missing start tag name"); + case eXMLErrorMissingEndTagName: return _CXML("Error: Missing end tag name"); + case eXMLErrorUnmatchedEndTag: return _CXML("Error: Unmatched end tag"); + case eXMLErrorUnmatchedEndClearTag: return _CXML("Error: Unmatched clear tag end"); + case eXMLErrorUnexpectedToken: return _CXML("Error: Unexpected token found"); + case eXMLErrorNoElements: return _CXML("Error: No elements found"); + case eXMLErrorFileNotFound: return _CXML("Error: File not found"); + case eXMLErrorFirstTagNotFound: return _CXML("Error: First Tag not found"); + case eXMLErrorUnknownCharacterEntity:return _CXML("Error: Unknown character entity"); + case eXMLErrorCharacterCodeAbove255: return _CXML("Error: Character code above 255 is forbidden in MultiByte char mode."); + case eXMLErrorCharConversionError: return _CXML("Error: unable to convert between WideChar and MultiByte chars"); + case eXMLErrorCannotOpenWriteFile: return _CXML("Error: unable to open file for writing"); + case eXMLErrorCannotWriteFile: return _CXML("Error: cannot write into file"); + + case eXMLErrorBase64DataSizeIsNotMultipleOf4: return _CXML("Warning: Base64-string length is not a multiple of 4"); + case eXMLErrorBase64DecodeTruncatedData: return _CXML("Warning: Base64-string is truncated"); + case eXMLErrorBase64DecodeIllegalCharacter: return _CXML("Error: Base64-string contains an illegal character"); + case eXMLErrorBase64DecodeBufferTooSmall: return _CXML("Error: Base64 decode output buffer is too small"); + }; + return _CXML("Unknown"); +} + +///////////////////////////////////////////////////////////////////////// +// Here start the abstraction layer to be OS-independent // +///////////////////////////////////////////////////////////////////////// + +// Here is an abstraction layer to access some common string manipulation functions. +// The abstraction layer is currently working for gcc, Microsoft Visual Studio 6.0, +// Microsoft Visual Studio .NET, CC (sun compiler) and Borland C++. +// If you plan to "port" the library to a new system/compiler, all you have to do is +// to edit the following lines. +#ifdef XML_NO_WIDE_CHAR +char myIsTextWideChar(const void *b, int len) { return FALSE; } +#else + #if defined (UNDER_CE) || !defined(_XMLWINDOWS) + char myIsTextWideChar(const void *b, int len) // inspired by the Wine API: RtlIsTextUnicode + { +#ifdef sun + // for SPARC processors: wchar_t* buffers must always be alligned, otherwise it's a char* buffer. + if ((((unsigned long)b)%sizeof(wchar_t))!=0) return FALSE; +#endif + const wchar_t *s=(const wchar_t*)b; + + // buffer too small: + if (len<(int)sizeof(wchar_t)) return FALSE; + + // odd length test + if (len&1) return FALSE; + + /* only checks the first 256 characters */ + len=mmin(256,len/sizeof(wchar_t)); + + // Check for the special byte order: + if (*((unsigned short*)s) == 0xFFFE) return TRUE; // IS_TEXT_UNICODE_REVERSE_SIGNATURE; + if (*((unsigned short*)s) == 0xFEFF) return TRUE; // IS_TEXT_UNICODE_SIGNATURE + + // checks for ASCII characters in the UNICODE stream + int i,stats=0; + for (i=0; ilen/2) return TRUE; + + // Check for UNICODE NULL chars + for (i=0; i + static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return wsncasecmp(c1,c2,l);} + static inline int xstrncmp(XMLCSTR c1, XMLCSTR c2, int l) { return wsncmp(c1,c2,l);} + static inline int xstricmp(XMLCSTR c1, XMLCSTR c2) { return wscasecmp(c1,c2); } + #else + // for gcc + static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return wcsncasecmp(c1,c2,l);} + static inline int xstrncmp(XMLCSTR c1, XMLCSTR c2, int l) { return wcsncmp(c1,c2,l);} + static inline int xstricmp(XMLCSTR c1, XMLCSTR c2) { return wcscasecmp(c1,c2); } + #endif + static inline XMLSTR xstrstr(XMLCSTR c1, XMLCSTR c2) { return (XMLSTR)wcsstr(c1,c2); } + static inline XMLSTR xstrcpy(XMLSTR c1, XMLCSTR c2) { return (XMLSTR)wcscpy(c1,c2); } + static inline FILE *xfopen(XMLCSTR filename,XMLCSTR mode) + { + char *filenameAscii=myWideCharToMultiByte(filename); + FILE *f; + if (mode[0]==_CXML('r')) f=fopen(filenameAscii,"rb"); + else f=fopen(filenameAscii,"wb"); + free(filenameAscii); + return f; + } + #else + static inline FILE *xfopen(XMLCSTR filename,XMLCSTR mode) { return fopen(filename,mode); } + static inline int xstrlen(XMLCSTR c) { return strlen(c); } + static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return strncasecmp(c1,c2,l);} + static inline int xstrncmp(XMLCSTR c1, XMLCSTR c2, int l) { return strncmp(c1,c2,l);} + static inline int xstricmp(XMLCSTR c1, XMLCSTR c2) { return strcasecmp(c1,c2); } + static inline XMLSTR xstrstr(XMLCSTR c1, XMLCSTR c2) { return (XMLSTR)strstr(c1,c2); } + static inline XMLSTR xstrcpy(XMLSTR c1, XMLCSTR c2) { return (XMLSTR)strcpy(c1,c2); } + #endif + static inline int _strnicmp(const char *c1,const char *c2, int l) { return strncasecmp(c1,c2,l);} +#endif + + +/////////////////////////////////////////////////////////////////////////////// +// the "xmltoc,xmltob,xmltoi,xmltol,xmltof,xmltoa" functions // +/////////////////////////////////////////////////////////////////////////////// +// These 6 functions are not used inside the XMLparser. +// There are only here as "convenience" functions for the user. +// If you don't need them, you can delete them without any trouble. +#ifdef _XMLWIDECHAR + #ifdef _XMLWINDOWS + // for Microsoft Visual Studio 6.0 and Microsoft Visual Studio .NET and Borland C++ Builder 6.0 + char xmltob(XMLCSTR t,int v){ if (t&&(*t)) return (char)_wtoi(t); return v; } + int xmltoi(XMLCSTR t,int v){ if (t&&(*t)) return _wtoi(t); return v; } + long xmltol(XMLCSTR t,long v){ if (t&&(*t)) return _wtol(t); return v; } + double xmltof(XMLCSTR t,double v){ if (t&&(*t)) wscanf(t, "%f", &v); /*v=_wtof(t);*/ return v; } + #else + #ifdef sun + // for CC + #include + char xmltob(XMLCSTR t,int v){ if (t) return (char)wstol(t,NULL,10); return v; } + int xmltoi(XMLCSTR t,int v){ if (t) return (int)wstol(t,NULL,10); return v; } + long xmltol(XMLCSTR t,long v){ if (t) return wstol(t,NULL,10); return v; } + #else + // for gcc + char xmltob(XMLCSTR t,int v){ if (t) return (char)wcstol(t,NULL,10); return v; } + int xmltoi(XMLCSTR t,int v){ if (t) return (int)wcstol(t,NULL,10); return v; } + long xmltol(XMLCSTR t,long v){ if (t) return wcstol(t,NULL,10); return v; } + #endif + double xmltof(XMLCSTR t,double v){ if (t&&(*t)) wscanf(t, "%f", &v); /*v=_wtof(t);*/ return v; } + #endif +#else + char xmltob(XMLCSTR t,char v){ if (t&&(*t)) return (char)atoi(t); return v; } + int xmltoi(XMLCSTR t,int v){ if (t&&(*t)) return atoi(t); return v; } + long xmltol(XMLCSTR t,long v){ if (t&&(*t)) return atol(t); return v; } + double xmltof(XMLCSTR t,double v){ if (t&&(*t)) return atof(t); return v; } +#endif +XMLCSTR xmltoa(XMLCSTR t,XMLCSTR v){ if (t) return t; return v; } +XMLCHAR xmltoc(XMLCSTR t,XMLCHAR v){ if (t&&(*t)) return *t; return v; } + +///////////////////////////////////////////////////////////////////////// +// the "openFileHelper" function // +///////////////////////////////////////////////////////////////////////// + +// Since each application has its own way to report and deal with errors, you should modify & rewrite +// the following "openFileHelper" function to get an "error reporting mechanism" tailored to your needs. +XMLNode XMLNode::openFileHelper(XMLCSTR filename, XMLCSTR tag) +{ + // guess the value of the global parameter "characterEncoding" + // (the guess is based on the first 200 bytes of the file). + FILE *f=xfopen(filename,_CXML("rb")); + if (f) + { + char bb[205]; + int l=(int)fread(bb,1,200,f); + setGlobalOptions(guessCharEncoding(bb,l),guessWideCharChars,dropWhiteSpace,removeCommentsInMiddleOfText); + fclose(f); + } + + // parse the file + XMLResults pResults; + XMLNode xnode=XMLNode::parseFile(filename,tag,&pResults); + + // display error message (if any) + if (pResults.error != eXMLErrorNone) + { + // create message + char message[2000],*s1=(char*)"",*s3=(char*)""; XMLCSTR s2=_CXML(""); + if (pResults.error==eXMLErrorFirstTagNotFound) { s1=(char*)"First Tag should be '"; s2=tag; s3=(char*)"'.\n"; } + sprintf(message, +#ifdef _XMLWIDECHAR + "XML Parsing error inside file '%S'.\n%S\nAt line %i, column %i.\n%s%S%s" +#else + "XML Parsing error inside file '%s'.\n%s\nAt line %i, column %i.\n%s%s%s" +#endif + ,filename,XMLNode::getError(pResults.error),pResults.nLine,pResults.nColumn,s1,s2,s3); + + // display message +#if defined(_XMLWINDOWS) && !defined(UNDER_CE) && !defined(_XMLPARSER_NO_MESSAGEBOX_) + MessageBoxA(NULL,message,"XML Parsing error",MB_OK|MB_ICONERROR|MB_TOPMOST); +#else + printf("%s",message); +#endif + exit(255); + } + return xnode; +} + +///////////////////////////////////////////////////////////////////////// +// Here start the core implementation of the XMLParser library // +///////////////////////////////////////////////////////////////////////// + +// You should normally not change anything below this point. + +#ifndef _XMLWIDECHAR +// If "characterEncoding=ascii" then we assume that all characters have the same length of 1 byte. +// If "characterEncoding=UTF8" then the characters have different lengths (from 1 byte to 4 bytes). +// If "characterEncoding=ShiftJIS" then the characters have different lengths (from 1 byte to 2 bytes). +// This table is used as lookup-table to know the length of a character (in byte) based on the +// content of the first byte of the character. +// (note: if you modify this, you must always have XML_utf8ByteTable[0]=0 ). +static const char XML_utf8ByteTable[256] = +{ + // 0 1 2 3 4 5 6 7 8 9 a b c d e f + 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x00 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x10 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x20 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x30 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x40 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x50 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x60 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x70 End of ASCII range + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x80 0x80 to 0xc1 invalid + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x90 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xa0 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xb0 + 1,1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xc0 0xc2 to 0xdf 2 byte + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xd0 + 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,// 0xe0 0xe0 to 0xef 3 byte + 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 +}; +static const char XML_legacyByteTable[256] = +{ + 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, + 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, + 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, + 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, + 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, + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1 +}; +static const char XML_sjisByteTable[256] = +{ + // 0 1 2 3 4 5 6 7 8 9 a b c d e f + 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x00 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x10 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x20 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x30 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x40 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x50 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x60 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x70 + 1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0x80 0x81 to 0x9F 2 bytes + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0x90 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xa0 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xb0 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xc0 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xd0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xe0 0xe0 to 0xef 2 bytes + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1 // 0xf0 +}; +static const char XML_gb2312ByteTable[256] = +{ +// 0 1 2 3 4 5 6 7 8 9 a b c d e f + 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x00 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x10 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x20 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x30 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x40 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x50 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x60 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x70 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x80 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x90 + 1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xa0 0xa1 to 0xf7 2 bytes + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xb0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xc0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xd0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xe0 + 2,2,2,2,2,2,2,2,1,1,1,1,1,1,1,1 // 0xf0 +}; +static const char XML_gbk_big5_ByteTable[256] = +{ + // 0 1 2 3 4 5 6 7 8 9 a b c d e f + 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x00 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x10 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x20 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x30 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x40 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x50 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x60 + 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x70 + 1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0x80 0x81 to 0xfe 2 bytes + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0x90 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xa0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xb0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xc0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xd0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xe0 + 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,1 // 0xf0 +}; +static const char *XML_ByteTable=(const char *)XML_utf8ByteTable; // the default is "characterEncoding=XMLNode::encoding_UTF8" +#endif + + +XMLNode XMLNode::emptyXMLNode; +XMLClear XMLNode::emptyXMLClear={ NULL, NULL, NULL}; +XMLAttribute XMLNode::emptyXMLAttribute={ NULL, NULL}; + +// Enumeration used to decipher what type a token is +typedef enum XMLTokenTypeTag +{ + eTokenText = 0, + eTokenQuotedText, + eTokenTagStart, /* "<" */ + eTokenTagEnd, /* "" */ + eTokenEquals, /* "=" */ + eTokenDeclaration, /* "" */ + eTokenClear, + eTokenError +} XMLTokenType; + +// Main structure used for parsing XML +typedef struct XML +{ + XMLCSTR lpXML; + XMLCSTR lpszText; + int nIndex,nIndexMissigEndTag; + enum XMLError error; + XMLCSTR lpEndTag; + int cbEndTag; + XMLCSTR lpNewElement; + int cbNewElement; + int nFirst; +} XML; + +typedef struct +{ + ALLXMLClearTag *pClr; + XMLCSTR pStr; +} NextToken; + +// Enumeration used when parsing attributes +typedef enum Attrib +{ + eAttribName = 0, + eAttribEquals, + eAttribValue +} Attrib; + +// Enumeration used when parsing elements to dictate whether we are currently +// inside a tag +typedef enum Status +{ + eInsideTag = 0, + eOutsideTag +} Status; + +XMLError XMLNode::writeToFile(XMLCSTR filename, const char *encoding, char nFormat) const +{ + if (!d) return eXMLErrorNone; + FILE *f=xfopen(filename,_CXML("wb")); + if (!f) return eXMLErrorCannotOpenWriteFile; +#ifdef _XMLWIDECHAR + unsigned char h[2]={ 0xFF, 0xFE }; + if (!fwrite(h,2,1,f)) return eXMLErrorCannotWriteFile; + if ((!isDeclaration())&&((d->lpszName)||(!getChildNode().isDeclaration()))) + { + if (!fwrite(L"\n",sizeof(wchar_t)*40,1,f)) + return eXMLErrorCannotWriteFile; + } +#else + if ((!isDeclaration())&&((d->lpszName)||(!getChildNode().isDeclaration()))) + { + if (characterEncoding==char_encoding_UTF8) + { + // header so that windows recognize the file as UTF-8: + unsigned char h[3]={0xEF,0xBB,0xBF}; if (!fwrite(h,3,1,f)) return eXMLErrorCannotWriteFile; + encoding="utf-8"; + } else if (characterEncoding==char_encoding_ShiftJIS) encoding="SHIFT-JIS"; + + if (!encoding) encoding="ISO-8859-1"; + if (fprintf(f,"\n",encoding)<0) return eXMLErrorCannotWriteFile; + } else + { + if (characterEncoding==char_encoding_UTF8) + { + unsigned char h[3]={0xEF,0xBB,0xBF}; if (!fwrite(h,3,1,f)) return eXMLErrorCannotWriteFile; + } + } +#endif + int i; + XMLSTR t=createXMLString(nFormat,&i); + if (!fwrite(t,sizeof(XMLCHAR)*i,1,f)) return eXMLErrorCannotWriteFile; + if (fclose(f)!=0) return eXMLErrorCannotWriteFile; + free(t); + return eXMLErrorNone; +} + +// Duplicate a given string. +XMLSTR stringDup(XMLCSTR lpszData, int cbData) +{ + if (lpszData==NULL) return NULL; + + XMLSTR lpszNew; + if (cbData==-1) cbData=(int)xstrlen(lpszData); + lpszNew = (XMLSTR)malloc((cbData+1) * sizeof(XMLCHAR)); + if (lpszNew) + { + memcpy(lpszNew, lpszData, (cbData) * sizeof(XMLCHAR)); + lpszNew[cbData] = (XMLCHAR)NULL; + } + return lpszNew; +} + +XMLSTR ToXMLStringTool::toXMLUnSafe(XMLSTR dest,XMLCSTR source) +{ + XMLSTR dd=dest; + XMLCHAR ch; + XMLCharacterEntity *entity; + while ((ch=*source)) + { + entity=XMLEntities; + do + { + if (ch==entity->c) {xstrcpy(dest,entity->s); dest+=entity->l; source++; goto out_of_loop1; } + entity++; + } while(entity->s); +#ifdef _XMLWIDECHAR + *(dest++)=*(source++); +#else + switch(XML_ByteTable[(unsigned char)ch]) + { + case 4: *(dest++)=*(source++); + case 3: *(dest++)=*(source++); + case 2: *(dest++)=*(source++); + case 1: *(dest++)=*(source++); + } +#endif +out_of_loop1: + ; + } + *dest=0; + return dd; +} + +// private (used while rendering): +int ToXMLStringTool::lengthXMLString(XMLCSTR source) +{ + int r=0; + XMLCharacterEntity *entity; + XMLCHAR ch; + while ((ch=*source)) + { + entity=XMLEntities; + do + { + if (ch==entity->c) { r+=entity->l; source++; goto out_of_loop1; } + entity++; + } while(entity->s); +#ifdef _XMLWIDECHAR + r++; source++; +#else + ch=XML_ByteTable[(unsigned char)ch]; r+=ch; source+=ch; +#endif +out_of_loop1: + ; + } + return r; +} + +ToXMLStringTool::~ToXMLStringTool(){ freeBuffer(); } +void ToXMLStringTool::freeBuffer(){ if (buf) free(buf); buf=NULL; buflen=0; } +XMLSTR ToXMLStringTool::toXML(XMLCSTR source) +{ + int l=lengthXMLString(source)+1; + if (l>buflen) { buflen=l; buf=(XMLSTR)realloc(buf,l*sizeof(XMLCHAR)); } + return toXMLUnSafe(buf,source); +} + +// private: +XMLSTR fromXMLString(XMLCSTR s, int lo, XML *pXML) +{ + // This function is the opposite of the function "toXMLString". It decodes the escape + // sequences &, ", ', <, > and replace them by the characters + // &,",',<,>. This function is used internally by the XML Parser. All the calls to + // the XML library will always gives you back "decoded" strings. + // + // in: string (s) and length (lo) of string + // out: new allocated string converted from xml + if (!s) return NULL; + + int ll=0,j; + XMLSTR d; + XMLCSTR ss=s; + XMLCharacterEntity *entity; + while ((lo>0)&&(*s)) + { + if (*s==_CXML('&')) + { + if ((lo>2)&&(s[1]==_CXML('#'))) + { + s+=2; lo-=2; + if ((*s==_CXML('X'))||(*s==_CXML('x'))) { s++; lo--; } + while ((*s)&&(*s!=_CXML(';'))&&((lo--)>0)) s++; + if (*s!=_CXML(';')) + { + pXML->error=eXMLErrorUnknownCharacterEntity; + return NULL; + } + s++; lo--; + } else + { + entity=XMLEntities; + do + { + if ((lo>=entity->l)&&(xstrnicmp(s,entity->s,entity->l)==0)) { s+=entity->l; lo-=entity->l; break; } + entity++; + } while(entity->s); + if (!entity->s) + { + pXML->error=eXMLErrorUnknownCharacterEntity; + return NULL; + } + } + } else + { +#ifdef _XMLWIDECHAR + s++; lo--; +#else + j=XML_ByteTable[(unsigned char)*s]; s+=j; lo-=j; ll+=j-1; +#endif + } + ll++; + } + + d=(XMLSTR)malloc((ll+1)*sizeof(XMLCHAR)); + s=d; + while (ll-->0) + { + if (*ss==_CXML('&')) + { + if (ss[1]==_CXML('#')) + { + ss+=2; j=0; + if ((*ss==_CXML('X'))||(*ss==_CXML('x'))) + { + ss++; + while (*ss!=_CXML(';')) + { + if ((*ss>=_CXML('0'))&&(*ss<=_CXML('9'))) j=(j<<4)+*ss-_CXML('0'); + else if ((*ss>=_CXML('A'))&&(*ss<=_CXML('F'))) j=(j<<4)+*ss-_CXML('A')+10; + else if ((*ss>=_CXML('a'))&&(*ss<=_CXML('f'))) j=(j<<4)+*ss-_CXML('a')+10; + else { free((void*)s); pXML->error=eXMLErrorUnknownCharacterEntity;return NULL;} + ss++; + } + } else + { + while (*ss!=_CXML(';')) + { + if ((*ss>=_CXML('0'))&&(*ss<=_CXML('9'))) j=(j*10)+*ss-_CXML('0'); + else { free((void*)s); pXML->error=eXMLErrorUnknownCharacterEntity;return NULL;} + ss++; + } + } +#ifndef _XMLWIDECHAR + if (j>255) { free((void*)s); pXML->error=eXMLErrorCharacterCodeAbove255;return NULL;} +#endif + (*d++)=(XMLCHAR)j; ss++; + } else + { + entity=XMLEntities; + do + { + if (xstrnicmp(ss,entity->s,entity->l)==0) { *(d++)=entity->c; ss+=entity->l; break; } + entity++; + } while(entity->s); + } + } else + { +#ifdef _XMLWIDECHAR + *(d++)=*(ss++); +#else + switch(XML_ByteTable[(unsigned char)*ss]) + { + case 4: *(d++)=*(ss++); ll--; + case 3: *(d++)=*(ss++); ll--; + case 2: *(d++)=*(ss++); ll--; + case 1: *(d++)=*(ss++); + } +#endif + } + } + *d=0; + return (XMLSTR)s; +} + +#define XML_isSPACECHAR(ch) ((ch==_CXML('\n'))||(ch==_CXML(' '))||(ch== _CXML('\t'))||(ch==_CXML('\r'))) + +// private: +char myTagCompare(XMLCSTR cclose, XMLCSTR copen) +// !!!! WARNING strange convention&: +// return 0 if equals +// return 1 if different +{ + if (!cclose) return 1; + int l=(int)xstrlen(cclose); + if (xstrnicmp(cclose, copen, l)!=0) return 1; + const XMLCHAR c=copen[l]; + if (XML_isSPACECHAR(c)|| + (c==_CXML('/' ))|| + (c==_CXML('<' ))|| + (c==_CXML('>' ))|| + (c==_CXML('=' ))) return 0; + return 1; +} + +// Obtain the next character from the string. +static inline XMLCHAR getNextChar(XML *pXML) +{ + XMLCHAR ch = pXML->lpXML[pXML->nIndex]; +#ifdef _XMLWIDECHAR + if (ch!=0) pXML->nIndex++; +#else + pXML->nIndex+=XML_ByteTable[(unsigned char)ch]; +#endif + return ch; +} + +// Find the next token in a string. +// pcbToken contains the number of characters that have been read. +static NextToken GetNextToken(XML *pXML, int *pcbToken, enum XMLTokenTypeTag *pType) +{ + NextToken result; + XMLCHAR ch; + XMLCHAR chTemp; + int indexStart,nFoundMatch,nIsText=FALSE; + result.pClr=NULL; // prevent warning + + // Find next non-white space character + do { indexStart=pXML->nIndex; ch=getNextChar(pXML); } while XML_isSPACECHAR(ch); + + if (ch) + { + // Cache the current string pointer + result.pStr = &pXML->lpXML[indexStart]; + + // First check whether the token is in the clear tag list (meaning it + // does not need formatting). + ALLXMLClearTag *ctag=XMLClearTags; + do + { + if (xstrncmp(ctag->lpszOpen, result.pStr, ctag->openTagLen)==0) + { + result.pClr=ctag; + pXML->nIndex+=ctag->openTagLen-1; + *pType=eTokenClear; + return result; + } + ctag++; + } while(ctag->lpszOpen); + + // If we didn't find a clear tag then check for standard tokens + switch(ch) + { + // Check for quotes + case _CXML('\''): + case _CXML('\"'): + // Type of token + *pType = eTokenQuotedText; + chTemp = ch; + + // Set the size + nFoundMatch = FALSE; + + // Search through the string to find a matching quote + while((ch = getNextChar(pXML))) + { + if (ch==chTemp) { nFoundMatch = TRUE; break; } + if (ch==_CXML('<')) break; + } + + // If we failed to find a matching quote + if (nFoundMatch == FALSE) + { + pXML->nIndex=indexStart+1; + nIsText=TRUE; + break; + } + +// 4.02.2002 +// if (FindNonWhiteSpace(pXML)) pXML->nIndex--; + + break; + + // Equals (used with attribute values) + case _CXML('='): + *pType = eTokenEquals; + break; + + // Close tag + case _CXML('>'): + *pType = eTokenCloseTag; + break; + + // Check for tag start and tag end + case _CXML('<'): + + // Peek at the next character to see if we have an end tag 'lpXML[pXML->nIndex]; + + // If we have a tag end... + if (chTemp == _CXML('/')) + { + // Set the type and ensure we point at the next character + getNextChar(pXML); + *pType = eTokenTagEnd; + } + + // If we have an XML declaration tag + else if (chTemp == _CXML('?')) + { + + // Set the type and ensure we point at the next character + getNextChar(pXML); + *pType = eTokenDeclaration; + } + + // Otherwise we must have a start tag + else + { + *pType = eTokenTagStart; + } + break; + + // Check to see if we have a short hand type end tag ('/>'). + case _CXML('/'): + + // 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 == _CXML('>')) + { + // Set the type and ensure we point at the next character + getNextChar(pXML); + *pType = eTokenShortHandClose; + 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; + while((ch = getNextChar(pXML))) + { + if XML_isSPACECHAR(ch) + { + indexStart++; break; + + } else if (ch==_CXML('/')) + { + // If we find a slash then this maybe text or a short hand end tag + // Peek at the next character to see it we have short hand end tag + ch=pXML->lpXML[pXML->nIndex]; + // If we found a short hand end tag then we need to exit the loop + if (ch==_CXML('>')) { pXML->nIndex--; break; } + + } else if ((ch==_CXML('<'))||(ch==_CXML('>'))||(ch==_CXML('='))) + { + pXML->nIndex--; break; + } + } + } + *pcbToken = pXML->nIndex-indexStart; + } else + { + // If we failed to obtain a valid character + *pcbToken = 0; + *pType = eTokenError; + result.pStr=NULL; + } + + return result; +} + +XMLCSTR XMLNode::updateName_WOSD(XMLSTR lpszName) +{ + if (!d) { free(lpszName); return NULL; } + if (d->lpszName&&(lpszName!=d->lpszName)) free((void*)d->lpszName); + d->lpszName=lpszName; + return lpszName; +} + +// private: +XMLNode::XMLNode(struct XMLNodeDataTag *p){ d=p; (p->ref_count)++; } +XMLNode::XMLNode(XMLNodeData *pParent, XMLSTR lpszName, char isDeclaration) +{ + d=(XMLNodeData*)malloc(sizeof(XMLNodeData)); + d->ref_count=1; + + d->lpszName=NULL; + d->nChild= 0; + d->nText = 0; + d->nClear = 0; + d->nAttribute = 0; + + d->isDeclaration = isDeclaration; + + d->pParent = pParent; + d->pChild= NULL; + d->pText= NULL; + d->pClear= NULL; + d->pAttribute= NULL; + d->pOrder= NULL; + + updateName_WOSD(lpszName); +} + +XMLNode XMLNode::createXMLTopNode_WOSD(XMLSTR lpszName, char isDeclaration) { return XMLNode(NULL,lpszName,isDeclaration); } +XMLNode XMLNode::createXMLTopNode(XMLCSTR lpszName, char isDeclaration) { return XMLNode(NULL,stringDup(lpszName),isDeclaration); } + +#define MEMORYINCREASE 50 + +static inline void myFree(void *p) { if (p) free(p); } +static inline void *myRealloc(void *p, int newsize, int memInc, int sizeofElem) +{ + if (p==NULL) { if (memInc) return malloc(memInc*sizeofElem); return malloc(sizeofElem); } + if ((memInc==0)||((newsize%memInc)==0)) p=realloc(p,(newsize+memInc)*sizeofElem); +// if (!p) +// { +// printf("XMLParser Error: Not enough memory! Aborting...\n"); exit(220); +// } + return p; +} + +// private: +XMLElementPosition XMLNode::findPosition(XMLNodeData *d, int index, XMLElementType xxtype) +{ + if (index<0) return -1; + int i=0,j=(int)((index<<2)+xxtype),*o=d->pOrder; while (o[i]!=j) i++; return i; +} + +// private: +// update "order" information when deleting a content of a XMLNode +int XMLNode::removeOrderElement(XMLNodeData *d, XMLElementType t, int index) +{ + int n=d->nChild+d->nText+d->nClear, *o=d->pOrder,i=findPosition(d,index,t); + memmove(o+i, o+i+1, (n-i)*sizeof(int)); + for (;ipOrder=(int)realloc(d->pOrder,n*sizeof(int)); + // but we skip reallocation because it's too time consuming. + // Anyway, at the end, it will be free'd completely at once. + return i; +} + +void *XMLNode::addToOrder(int memoryIncrease,int *_pos, int nc, void *p, int size, XMLElementType xtype) +{ + // in: *_pos is the position inside d->pOrder ("-1" means "EndOf") + // out: *_pos is the index inside p + p=myRealloc(p,(nc+1),memoryIncrease,size); + int n=d->nChild+d->nText+d->nClear; + d->pOrder=(int*)myRealloc(d->pOrder,n+1,memoryIncrease*3,sizeof(int)); + int pos=*_pos,*o=d->pOrder; + + if ((pos<0)||(pos>=n)) { *_pos=nc; o[n]=(int)((nc<<2)+xtype); return p; } + + int i=pos; + memmove(o+i+1, o+i, (n-i)*sizeof(int)); + + while ((pos>2; + memmove(((char*)p)+(pos+1)*size,((char*)p)+pos*size,(nc-pos)*size); + + return p; +} + +// Add a child node to the given element. +XMLNode XMLNode::addChild_priv(int memoryIncrease, XMLSTR lpszName, char isDeclaration, int pos) +{ + if (!lpszName) return emptyXMLNode; + d->pChild=(XMLNode*)addToOrder(memoryIncrease,&pos,d->nChild,d->pChild,sizeof(XMLNode),eNodeChild); + d->pChild[pos].d=NULL; + d->pChild[pos]=XMLNode(d,lpszName,isDeclaration); + d->nChild++; + return d->pChild[pos]; +} + +// Add an attribute to an element. +XMLAttribute *XMLNode::addAttribute_priv(int memoryIncrease,XMLSTR lpszName, XMLSTR lpszValuev) +{ + if (!lpszName) return &emptyXMLAttribute; + if (!d) { myFree(lpszName); myFree(lpszValuev); return &emptyXMLAttribute; } + int nc=d->nAttribute; + d->pAttribute=(XMLAttribute*)myRealloc(d->pAttribute,(nc+1),memoryIncrease,sizeof(XMLAttribute)); + XMLAttribute *pAttr=d->pAttribute+nc; + pAttr->lpszName = lpszName; + pAttr->lpszValue = lpszValuev; + d->nAttribute++; + return pAttr; +} + +// Add text to the element. +XMLCSTR XMLNode::addText_priv(int memoryIncrease, XMLSTR lpszValue, int pos) +{ + if (!lpszValue) return NULL; + if (!d) { myFree(lpszValue); return NULL; } + d->pText=(XMLCSTR*)addToOrder(memoryIncrease,&pos,d->nText,d->pText,sizeof(XMLSTR),eNodeText); + d->pText[pos]=lpszValue; + d->nText++; + return lpszValue; +} + +// Add clear (unformatted) text to the element. +XMLClear *XMLNode::addClear_priv(int memoryIncrease, XMLSTR lpszValue, XMLCSTR lpszOpen, XMLCSTR lpszClose, int pos) +{ + if (!lpszValue) return &emptyXMLClear; + if (!d) { myFree(lpszValue); return &emptyXMLClear; } + d->pClear=(XMLClear *)addToOrder(memoryIncrease,&pos,d->nClear,d->pClear,sizeof(XMLClear),eNodeClear); + XMLClear *pNewClear=d->pClear+pos; + pNewClear->lpszValue = lpszValue; + if (!lpszOpen) lpszOpen=XMLClearTags->lpszOpen; + if (!lpszClose) lpszClose=XMLClearTags->lpszClose; + pNewClear->lpszOpenTag = lpszOpen; + pNewClear->lpszCloseTag = lpszClose; + d->nClear++; + return pNewClear; +} + +// private: +// Parse a clear (unformatted) type node. +char XMLNode::parseClearTag(void *px, void *_pClear) +{ + XML *pXML=(XML *)px; + ALLXMLClearTag pClear=*((ALLXMLClearTag*)_pClear); + int cbTemp=0; + XMLCSTR lpszTemp=NULL; + XMLCSTR lpXML=&pXML->lpXML[pXML->nIndex]; + static XMLCSTR docTypeEnd=_CXML("]>"); + + // Find the closing tag + // Seems the ')) { lpszTemp=pCh; break; } +#ifdef _XMLWIDECHAR + pCh++; +#else + pCh+=XML_ByteTable[(unsigned char)(*pCh)]; +#endif + } + } else lpszTemp=xstrstr(lpXML, pClear.lpszClose); + + if (lpszTemp) + { + // Cache the size and increment the index + cbTemp = (int)(lpszTemp - lpXML); + + pXML->nIndex += cbTemp+(int)xstrlen(pClear.lpszClose); + + // Add the clear node to the current element + addClear_priv(MEMORYINCREASE,stringDup(lpXML,cbTemp), pClear.lpszOpen, pClear.lpszClose,-1); + return 0; + } + + // If we failed to find the end tag + pXML->error = eXMLErrorUnmatchedEndClearTag; + return 1; +} + +void XMLNode::exactMemory(XMLNodeData *d) +{ + if (d->pOrder) d->pOrder=(int*)realloc(d->pOrder,(d->nChild+d->nText+d->nClear)*sizeof(int)); + if (d->pChild) d->pChild=(XMLNode*)realloc(d->pChild,d->nChild*sizeof(XMLNode)); + if (d->pAttribute) d->pAttribute=(XMLAttribute*)realloc(d->pAttribute,d->nAttribute*sizeof(XMLAttribute)); + if (d->pText) d->pText=(XMLCSTR*)realloc(d->pText,d->nText*sizeof(XMLSTR)); + if (d->pClear) d->pClear=(XMLClear *)realloc(d->pClear,d->nClear*sizeof(XMLClear)); +} + +char XMLNode::maybeAddTxT(void *pa, XMLCSTR tokenPStr) +{ + XML *pXML=(XML *)pa; + XMLCSTR lpszText=pXML->lpszText; + if (!lpszText) return 0; + if (dropWhiteSpace) while (XML_isSPACECHAR(*lpszText)&&(lpszText!=tokenPStr)) lpszText++; + int cbText = (int)(tokenPStr - lpszText); + if (!cbText) { pXML->lpszText=NULL; return 0; } + if (dropWhiteSpace) { cbText--; while ((cbText)&&XML_isSPACECHAR(lpszText[cbText])) cbText--; cbText++; } + if (!cbText) { pXML->lpszText=NULL; return 0; } + XMLSTR lpt=fromXMLString(lpszText,cbText,pXML); + if (!lpt) return 1; + pXML->lpszText=NULL; + if (removeCommentsInMiddleOfText && d->nText && d->nClear) + { + // if the previous insertion was a comment () AND + // if the previous previous insertion was a text then, delete the comment and append the text + int n=d->nChild+d->nText+d->nClear-1,*o=d->pOrder; + if (((o[n]&3)==eNodeClear)&&((o[n-1]&3)==eNodeText)) + { + int i=o[n]>>2; + if (d->pClear[i].lpszOpenTag==XMLClearTags[2].lpszOpen) + { + deleteClear(i); + i=o[n-1]>>2; + n=xstrlen(d->pText[i]); + int n2=xstrlen(lpt)+1; + d->pText[i]=(XMLSTR)realloc((void*)d->pText[i],(n+n2)*sizeof(XMLCHAR)); + if (!d->pText[i]) return 1; + memcpy((void*)(d->pText[i]+n),lpt,n2*sizeof(XMLCHAR)); + free(lpt); + return 0; + } + } + } + addText_priv(MEMORYINCREASE,lpt,-1); + return 0; +} +// private: +// Recursively parse an XML element. +int XMLNode::ParseXMLElement(void *pa) +{ + XML *pXML=(XML *)pa; + int cbToken; + enum XMLTokenTypeTag xtype; + NextToken token; + XMLCSTR lpszTemp=NULL; + int cbTemp=0; + char nDeclaration; + XMLNode pNew; + enum Status status; // inside or outside a tag + enum Attrib attrib = eAttribName; + + assert(pXML); + + // If this is the first call to the function + if (pXML->nFirst) + { + // Assume we are outside of a tag definition + pXML->nFirst = FALSE; + status = eOutsideTag; + } else + { + // If this is not the first call then we should only be called when inside a tag. + status = eInsideTag; + } + + // Iterate through the tokens in the document + for(;;) + { + // Obtain the next token + token = GetNextToken(pXML, &cbToken, &xtype); + + if (xtype != eTokenError) + { + // Check the current status + switch(status) + { + + // If we are outside of a tag definition + case eOutsideTag: + + // Check what type of token we obtained + switch(xtype) + { + // If we have found text or quoted text + case eTokenText: + case eTokenCloseTag: /* '>' */ + case eTokenShortHandClose: /* '/>' */ + case eTokenQuotedText: + case eTokenEquals: + break; + + // If we found a start tag '<' and declarations 'error = eXMLErrorMissingTagName; + return FALSE; + } + + // If we found a new element which is the same as this + // element then we need to pass this back to the caller.. + +#ifdef APPROXIMATE_PARSING + if (d->lpszName && + myTagCompare(d->lpszName, token.pStr) == 0) + { + // Indicate to the caller that it needs to create a + // new element. + pXML->lpNewElement = token.pStr; + pXML->cbNewElement = cbToken; + return TRUE; + } else +#endif + { + // If the name of the new element differs from the name of + // the current element we need to add the new element to + // the current one and recurse + pNew = addChild_priv(MEMORYINCREASE,stringDup(token.pStr,cbToken), nDeclaration,-1); + + while (!pNew.isEmpty()) + { + // Callself to process the new node. If we return + // FALSE this means we dont have any more + // processing to do... + + if (!pNew.ParseXMLElement(pXML)) return FALSE; + else + { + // If the call to recurse this function + // evented in a end tag specified in XML then + // we need to unwind the calls to this + // function until we find the appropriate node + // (the element name and end tag name must + // match) + if (pXML->cbEndTag) + { + // If we are back at the root node then we + // have an unmatched end tag + if (!d->lpszName) + { + pXML->error=eXMLErrorUnmatchedEndTag; + return FALSE; + } + + // If the end tag matches the name of this + // element then we only need to unwind + // once more... + + if (myTagCompare(d->lpszName, pXML->lpEndTag)==0) + { + pXML->cbEndTag = 0; + } + + return TRUE; + } else + if (pXML->cbNewElement) + { + // If the call indicated a new element is to + // be created on THIS element. + + // If the name of this element matches the + // name of the element we need to create + // then we need to return to the caller + // and let it process the element. + + if (myTagCompare(d->lpszName, pXML->lpNewElement)==0) + { + return TRUE; + } + + // Add the new element and recurse + pNew = addChild_priv(MEMORYINCREASE,stringDup(pXML->lpNewElement,pXML->cbNewElement),0,-1); + pXML->cbNewElement = 0; + } + else + { + // If we didn't have a new element to create + pNew = emptyXMLNode; + + } + } + } + } + break; + + // If we found an end tag + case eTokenTagEnd: + + // If we have node text then add this to the element + if (maybeAddTxT(pXML,token.pStr)) return FALSE; + + // Find the name of the end tag + token = GetNextToken(pXML, &cbTemp, &xtype); + + // The end tag should be text + if (xtype != eTokenText) + { + pXML->error = eXMLErrorMissingEndTagName; + return FALSE; + } + lpszTemp = token.pStr; + + // After the end tag we should find a closing tag + token = GetNextToken(pXML, &cbToken, &xtype); + if (xtype != eTokenCloseTag) + { + pXML->error = eXMLErrorMissingEndTagName; + return FALSE; + } + pXML->lpszText=pXML->lpXML+pXML->nIndex; + + // We need to return to the previous caller. If the name + // of the tag cannot be found we need to keep returning to + // caller until we find a match + if (myTagCompare(d->lpszName, lpszTemp) != 0) +#ifdef STRICT_PARSING + { + pXML->error=eXMLErrorUnmatchedEndTag; + pXML->nIndexMissigEndTag=pXML->nIndex; + return FALSE; + } +#else + { + pXML->error=eXMLErrorMissingEndTag; + pXML->nIndexMissigEndTag=pXML->nIndex; + pXML->lpEndTag = lpszTemp; + pXML->cbEndTag = cbTemp; + } +#endif + + // Return to the caller + exactMemory(d); + return TRUE; + + // If we found a clear (unformatted) token + case eTokenClear: + // If we have node text then add this to the element + if (maybeAddTxT(pXML,token.pStr)) return FALSE; + if (parseClearTag(pXML, token.pClr)) return FALSE; + pXML->lpszText=pXML->lpXML+pXML->nIndex; + break; + + default: + break; + } + break; + + // If we are inside a tag definition we need to search for attributes + case eInsideTag: + + // Check what part of the attribute (name, equals, value) we + // are looking for. + switch(attrib) + { + // If we are looking for a new attribute + case eAttribName: + + // Check what the current token type is + switch(xtype) + { + // If the current type is text... + // Eg. 'attribute' + case eTokenText: + // Cache the token then indicate that we are next to + // look for the equals + lpszTemp = token.pStr; + cbTemp = cbToken; + attrib = eAttribEquals; + break; + + // If we found a closing tag... + // Eg. '>' + case eTokenCloseTag: + // We are now outside the tag + status = eOutsideTag; + pXML->lpszText=pXML->lpXML+pXML->nIndex; + break; + + // If we found a short hand '/>' closing tag then we can + // return to the caller + case eTokenShortHandClose: + exactMemory(d); + pXML->lpszText=pXML->lpXML+pXML->nIndex; + return TRUE; + + // Errors... + case eTokenQuotedText: /* '"SomeText"' */ + case eTokenTagStart: /* '<' */ + case eTokenTagEnd: /* 'error = eXMLErrorUnexpectedToken; + return FALSE; + default: break; + } + break; + + // If we are looking for an equals + case eAttribEquals: + // Check what the current token type is + switch(xtype) + { + // If the current type is text... + // Eg. 'Attribute AnotherAttribute' + case eTokenText: + // Add the unvalued attribute to the list + addAttribute_priv(MEMORYINCREASE,stringDup(lpszTemp,cbTemp), NULL); + // Cache the token then indicate. We are next to + // look for the equals attribute + lpszTemp = token.pStr; + cbTemp = cbToken; + break; + + // If we found a closing tag 'Attribute >' or a short hand + // closing tag 'Attribute />' + case eTokenShortHandClose: + case eTokenCloseTag: + // If we are a declaration element 'lpszText=pXML->lpXML+pXML->nIndex; + + if (d->isDeclaration && + (lpszTemp[cbTemp-1]) == _CXML('?')) + { + cbTemp--; + if (d->pParent && d->pParent->pParent) xtype = eTokenShortHandClose; + } + + if (cbTemp) + { + // Add the unvalued attribute to the list + addAttribute_priv(MEMORYINCREASE,stringDup(lpszTemp,cbTemp), NULL); + } + + // If this is the end of the tag then return to the caller + if (xtype == eTokenShortHandClose) + { + exactMemory(d); + return TRUE; + } + + // We are now outside the tag + status = eOutsideTag; + break; + + // If we found the equals token... + // Eg. 'Attr