xmlparser.cpp

xml文件解析器 解析所有的xml文件 支持unicode

/**
 ****************************************************************************
 * <P> 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.  </P>
 *
 * @version     V2.31
 * @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:
 *      <a><b>some text</b><b>other text    </a>
 *   as an error. Otherwise, this tring will be equivalent to:
 *      <a><b>some text</b><b>other text</b></a>
 *
 * NOTE:
 *
 *   If you add "#define APPROXIMATE_PARSING" on the first line of this file
 *   the parser will see the following XML-stream:
 *     <data name="n1">
 *     <data name="n2">
 *     <data name="n3" />
 *   as equivalent to the following XML-stream:
 *     <data name="n1" />
 *     <data name="n2" />
 *     <data name="n3" />
 *   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):
 *     <data name="n1">
 *        <data name="n2">
 *            <data name="n3" />
 *        </data>
 *     </data>
 *
 * 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.
 *
 * BSD license:
 * Copyright (c) 2002, Frank Vanden Berghen
 * All rights reserved.
 * 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 the 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 THE REGENTS AND CONTRIBUTORS ``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 THE REGENTS AND CONTRIBUTORS 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 "xmlParser.h"
#ifdef _XMLWINDOWS
//#ifdef _DEBUG
//#define _CRTDBG_MAP_ALLOC
//#include <crtdbg.h>
//#endif
#define WIN32_LEAN_AND_MEAN
#include <Windows.h> // to have IsTextUnicode, MultiByteToWideChar, WideCharToMultiByte to handle unicode files
                     // to have "MessageBoxA" to display error messages for openFilHelper
#endif

#include <memory.h>
#include <assert.h>
#include <stdio.h>
#include <string.h>
#include <stdlib.h>

XMLCSTR XMLNode::getVersion() { return _X("v2.30"); }
void freeXMLString(XMLSTR t){free(t);}

static XMLNode::XMLCharEncoding characterEncoding=XMLNode::encoding_UTF8;
static char guessWideCharChars=1, dropWhiteSpace=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 "<!DOCTYPE" declaration must be the second in the list.
typedef struct { XMLCSTR lpszOpen; int openTagLen; XMLCSTR lpszClose;} ALLXMLClearTag;
static ALLXMLClearTag XMLClearTags[] =
{
    {    _X("<![CDATA["),9,  _X("]]>")      },
    {    _X("<!DOCTYPE"),9,  _X(">")        },
    {    _X("<PRE>")    ,5,  _X("</PRE>")   },
    {    _X("<Script>") ,8,  _X("</Script>")},
    {    _X("<!--")     ,4,  _X("-->")      },
    {    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 "&#xA0;" and "&#160;" are recognized.
typedef struct { XMLCSTR s; int l; XMLCHAR c;} XMLCharacterEntity;
static XMLCharacterEntity XMLEntities[] =
{
    { _X("&amp;" ), 5, _X('&' )},
    { _X("&lt;"  ), 4, _X('<' )},
    { _X("&gt;"  ), 4, _X('>' )},
    { _X("&quot;"), 6, _X('\"')},
    { _X("&apos;"), 6, _X('\'')},
    { 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 _X('\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 _X("No error");
    case eXMLErrorMissingEndTag:         return _X("Warning: Unmatched end tag");
    case eXMLErrorNoXMLTagFound:         return _X("Warning: No XML tag found");
    case eXMLErrorEmpty:                 return _X("Error: No XML data");
    case eXMLErrorMissingTagName:        return _X("Error: Missing start tag name");
    case eXMLErrorMissingEndTagName:     return _X("Error: Missing end tag name");
    case eXMLErrorUnmatchedEndTag:       return _X("Error: Unmatched end tag");
    case eXMLErrorUnmatchedEndClearTag:  return _X("Error: Unmatched clear tag end");
    case eXMLErrorUnexpectedToken:       return _X("Error: Unexpected token found");
    case eXMLErrorNoElements:            return _X("Error: No elements found");
    case eXMLErrorFileNotFound:          return _X("Error: File not found");
    case eXMLErrorFirstTagNotFound:      return _X("Error: First Tag not found");
    case eXMLErrorUnknownCharacterEntity:return _X("Error: Unknown character entity");
    case eXMLErrorCharConversionError:   return _X("Error: unable to convert between WideChar and MultiByte chars");
    case eXMLErrorCannotOpenWriteFile:   return _X("Error: unable to open file for writing");
    case eXMLErrorCannotWriteFile:       return _X("Error: cannot write into file");

    case eXMLErrorBase64DataSizeIsNotMultipleOf4: return _X("Warning: Base64-string length is not a multiple of 4");
    case eXMLErrorBase64DecodeTruncatedData:      return _X("Warning: Base64-string is truncated");
    case eXMLErrorBase64DecodeIllegalCharacter:   return _X("Error: Base64-string contains an illegal character");
    case eXMLErrorBase64DecodeBufferTooSmall:     return _X("Error: Base64 decode output buffer is too small");
    };
    return _X("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; i<len; i++) if (s[i]<=(unsigned short)255) stats++;
        if (stats>len/2) return TRUE;

        // Check for UNICODE NULL chars
        for (i=0; i<len; i++) if (!s[i]) return TRUE;

        return FALSE;
    }
    #else
    char myIsTextWideChar(const void *b,int l) { return (char)IsTextUnicode((CONST LPVOID)b,l,NULL); };
    #endif
#endif

#ifdef _XMLWINDOWS
// for Microsoft Visual Studio 6.0 and Microsoft Visual Studio .NET,
    #ifdef _XMLWIDECHAR
        wchar_t *myMultiByteToWideChar(const char *s)
        {
            int i;
            if (characterEncoding==XMLNode::encoding_UTF8) i=(int)MultiByteToWideChar(CP_UTF8,0             ,s,-1,NULL,0);
            else                                           i=(int)MultiByteToWideChar(CP_ACP ,MB_PRECOMPOSED,s,-1,NULL,0);
            if (i<0) return NULL;
            wchar_t *d=(wchar_t *)malloc((i+1)*sizeof(XMLCHAR));
            if (characterEncoding==XMLNode::encoding_UTF8) i=(int)MultiByteToWideChar(CP_UTF8,0             ,s,-1,d,i);
            else                                           i=(int)MultiByteToWideChar(CP_ACP ,MB_PRECOMPOSED,s,-1,d,i);
            d[i]=0;
            return d;
        }
        static inline FILE *xfopen(XMLCSTR filename,XMLCSTR mode) { return _wfopen(filename,mode); }
        static inline int xstrlen(XMLCSTR c)   { return (int)wcslen(c); }
        static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return _wcsnicmp(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 _wcsicmp(c1,c2); }
        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); }
    #else
        char *myWideCharToMultiByte(const wchar_t *s)
        {
            UINT codePage=CP_ACP; if (characterEncoding==XMLNode::encoding_UTF8) codePage=CP_UTF8;
            int i=(int)WideCharToMultiByte(codePage,  // code page
                0,                       // performance and mapping flags
                s,                       // wide-character string
                -1,                       // number of chars in string
                NULL,                       // buffer for new string
                0,                       // size of buffer
                NULL,                    // default for unmappable chars
                NULL                     // set when default char used
                );
            if (i<0) return NULL;
            char *d=(char*)malloc(i+1);
            WideCharToMultiByte(codePage,  // code page
                0,                       // performance and mapping flags
                s,                       // wide-character string
                -1,                       // number of chars in string
                d,                       // buffer for new string
                i,                       // size of buffer
                NULL,                    // default for unmappable chars
                NULL                     // set when default char used
                );
            d[i]=0;
            return d;
        }
        static inline FILE *xfopen(XMLCSTR filename,XMLCSTR mode) { return fopen(filename,mode); }
        static inline int xstrlen(XMLCSTR c)   { return (int)strlen(c); }
        static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return _strnicmp(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 _stricmp(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
    #ifdef __BORLANDC__
    static inline int _strnicmp(char *c1, char *c2, int l){ return strnicmp(c1,c2,l);}
    #endif
#else
// for gcc and CC
    #ifdef XML_NO_WIDE_CHAR
        char *myWideCharToMultiByte(const wchar_t *s) { return NULL; }
    #else
        char *myWideCharToMultiByte(const wchar_t *s)
        {
            const wchar_t *ss=s;
            int i=(int)wcsrtombs(NULL,&ss,0,NULL);
            if (i<0) return NULL;
            char *d=(char *)malloc(i+1);
            wcsrtombs(d,&s,i,NULL);
            d[i]=0;
            return d;
        }
    #endif
    #ifdef _XMLWIDECHAR
        wchar_t *myMultiByteToWideChar(const char *s)
        {
            const char *ss=s;
            int i=(int)mbsrtowcs(NULL,&ss,0,NULL);
            if (i<0) return NULL;
            wchar_t *d=(wchar_t *)malloc((i+1)*sizeof(wchar_t));
            mbsrtowcs(d,&s,i,NULL);
            d[i]=0;
            return d;
        }
        int xstrlen(XMLCSTR c)   { return wcslen(c); }
        #ifdef sun
        // for CC
           #include <widec.h>
           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]==_X('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 "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).