[164] | 1 | /** |
---|
| 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> |
---|
| 6 | * |
---|
| 7 | * @version V2.33 |
---|
| 8 | * @author Frank Vanden Berghen |
---|
| 9 | * |
---|
| 10 | * NOTE: |
---|
| 11 | * |
---|
| 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> |
---|
| 17 | * |
---|
| 18 | * NOTE: |
---|
| 19 | * |
---|
| 20 | * If you add "#define APPROXIMATE_PARSING" on the first line of this file |
---|
| 21 | * the parser will see the following XML-stream: |
---|
| 22 | * <data name="n1"> |
---|
| 23 | * <data name="n2"> |
---|
| 24 | * <data name="n3" /> |
---|
| 25 | * as equivalent to the following XML-stream: |
---|
| 26 | * <data name="n1" /> |
---|
| 27 | * <data name="n2" /> |
---|
| 28 | * <data name="n3" /> |
---|
| 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): |
---|
| 32 | * <data name="n1"> |
---|
| 33 | * <data name="n2"> |
---|
| 34 | * <data name="n3" /> |
---|
| 35 | * </data> |
---|
| 36 | * </data> |
---|
| 37 | * |
---|
| 38 | * NOTE: |
---|
| 39 | * |
---|
| 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. |
---|
| 44 | * |
---|
| 45 | * BSD license: |
---|
| 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: |
---|
| 50 | * |
---|
| 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. |
---|
| 59 | * |
---|
| 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. |
---|
| 70 | * |
---|
| 71 | **************************************************************************** |
---|
| 72 | */ |
---|
| 73 | #ifndef _CRT_SECURE_NO_DEPRECATE |
---|
| 74 | #define _CRT_SECURE_NO_DEPRECATE |
---|
| 75 | #endif |
---|
| 76 | #include "xmlParser.h" |
---|
| 77 | #ifdef _XMLWINDOWS |
---|
| 78 | //#ifdef _DEBUG |
---|
| 79 | //#define _CRTDBG_MAP_ALLOC |
---|
| 80 | //#include <crtdbg.h> |
---|
| 81 | //#endif |
---|
| 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 |
---|
| 85 | #endif |
---|
| 86 | |
---|
| 87 | #include <memory.h> |
---|
| 88 | #include <assert.h> |
---|
| 89 | #include <stdio.h> |
---|
| 90 | #include <string.h> |
---|
| 91 | #include <stdlib.h> |
---|
| 92 | |
---|
| 93 | XMLCSTR XMLNode::getVersion() { return _X("v2.30"); } |
---|
| 94 | void freeXMLString(XMLSTR t){free(t);} |
---|
| 95 | |
---|
| 96 | static XMLNode::XMLCharEncoding characterEncoding=XMLNode::encoding_UTF8; |
---|
| 97 | static char guessWideCharChars=1, dropWhiteSpace=1; |
---|
| 98 | |
---|
| 99 | inline int mmin( const int t1, const int t2 ) { return t1 < t2 ? t1 : t2; } |
---|
| 100 | |
---|
| 101 | // You can modify the initialization of the variable "XMLClearTags" below |
---|
| 102 | // to change the clearTags that are currently recognized by the library. |
---|
| 103 | // The number on the second columns is the length of the string inside the |
---|
| 104 | // first column. The "<!DOCTYPE" declaration must be the second in the list. |
---|
| 105 | typedef struct { XMLCSTR lpszOpen; int openTagLen; XMLCSTR lpszClose;} ALLXMLClearTag; |
---|
| 106 | static ALLXMLClearTag XMLClearTags[] = |
---|
| 107 | { |
---|
| 108 | { _X("<![CDATA["),9, _X("]]>") }, |
---|
| 109 | { _X("<!DOCTYPE"),9, _X(">") }, |
---|
| 110 | { _X("<PRE>") ,5, _X("</PRE>") }, |
---|
| 111 | { _X("<Script>") ,8, _X("</Script>")}, |
---|
| 112 | { _X("<!--") ,4, _X("-->") }, |
---|
| 113 | { NULL ,0, NULL } |
---|
| 114 | }; |
---|
| 115 | |
---|
| 116 | // You can modify the initialization of the variable "XMLEntities" below |
---|
| 117 | // to change the character entities that are currently recognized by the library. |
---|
| 118 | // The number on the second columns is the length of the string inside the |
---|
| 119 | // first column. Additionally, the syntaxes " " and " " are recognized. |
---|
| 120 | typedef struct { XMLCSTR s; int l; XMLCHAR c;} XMLCharacterEntity; |
---|
| 121 | static XMLCharacterEntity XMLEntities[] = |
---|
| 122 | { |
---|
| 123 | { _X("&" ), 5, _X('&' )}, |
---|
| 124 | { _X("<" ), 4, _X('<' )}, |
---|
| 125 | { _X(">" ), 4, _X('>' )}, |
---|
| 126 | { _X("""), 6, _X('\"')}, |
---|
| 127 | { _X("'"), 6, _X('\'')}, |
---|
| 128 | { NULL , 0, '\0' } |
---|
| 129 | }; |
---|
| 130 | |
---|
| 131 | // When rendering the XMLNode to a string (using the "createXMLString" function), |
---|
| 132 | // you can ask for a beautiful formatting. This formatting is using the |
---|
| 133 | // following indentation character: |
---|
| 134 | #define INDENTCHAR _X('\t') |
---|
| 135 | |
---|
| 136 | // The following function parses the XML errors into a user friendly string. |
---|
| 137 | // You can edit this to change the output language of the library to something else. |
---|
| 138 | XMLCSTR XMLNode::getError(XMLError xerror) |
---|
| 139 | { |
---|
| 140 | switch (xerror) |
---|
| 141 | { |
---|
| 142 | case eXMLErrorNone: return _X("No error"); |
---|
| 143 | case eXMLErrorMissingEndTag: return _X("Warning: Unmatched end tag"); |
---|
| 144 | case eXMLErrorNoXMLTagFound: return _X("Warning: No XML tag found"); |
---|
| 145 | case eXMLErrorEmpty: return _X("Error: No XML data"); |
---|
| 146 | case eXMLErrorMissingTagName: return _X("Error: Missing start tag name"); |
---|
| 147 | case eXMLErrorMissingEndTagName: return _X("Error: Missing end tag name"); |
---|
| 148 | case eXMLErrorUnmatchedEndTag: return _X("Error: Unmatched end tag"); |
---|
| 149 | case eXMLErrorUnmatchedEndClearTag: return _X("Error: Unmatched clear tag end"); |
---|
| 150 | case eXMLErrorUnexpectedToken: return _X("Error: Unexpected token found"); |
---|
| 151 | case eXMLErrorNoElements: return _X("Error: No elements found"); |
---|
| 152 | case eXMLErrorFileNotFound: return _X("Error: File not found"); |
---|
| 153 | case eXMLErrorFirstTagNotFound: return _X("Error: First Tag not found"); |
---|
| 154 | case eXMLErrorUnknownCharacterEntity:return _X("Error: Unknown character entity"); |
---|
| 155 | case eXMLErrorCharConversionError: return _X("Error: unable to convert between WideChar and MultiByte chars"); |
---|
| 156 | case eXMLErrorCannotOpenWriteFile: return _X("Error: unable to open file for writing"); |
---|
| 157 | case eXMLErrorCannotWriteFile: return _X("Error: cannot write into file"); |
---|
| 158 | |
---|
| 159 | case eXMLErrorBase64DataSizeIsNotMultipleOf4: return _X("Warning: Base64-string length is not a multiple of 4"); |
---|
| 160 | case eXMLErrorBase64DecodeTruncatedData: return _X("Warning: Base64-string is truncated"); |
---|
| 161 | case eXMLErrorBase64DecodeIllegalCharacter: return _X("Error: Base64-string contains an illegal character"); |
---|
| 162 | case eXMLErrorBase64DecodeBufferTooSmall: return _X("Error: Base64 decode output buffer is too small"); |
---|
| 163 | }; |
---|
| 164 | return _X("Unknown"); |
---|
| 165 | } |
---|
| 166 | |
---|
| 167 | ///////////////////////////////////////////////////////////////////////// |
---|
| 168 | // Here start the abstraction layer to be OS-independent // |
---|
| 169 | ///////////////////////////////////////////////////////////////////////// |
---|
| 170 | |
---|
| 171 | // Here is an abstraction layer to access some common string manipulation functions. |
---|
| 172 | // The abstraction layer is currently working for gcc, Microsoft Visual Studio 6.0, |
---|
| 173 | // Microsoft Visual Studio .NET, CC (sun compiler) and Borland C++. |
---|
| 174 | // If you plan to "port" the library to a new system/compiler, all you have to do is |
---|
| 175 | // to edit the following lines. |
---|
| 176 | #ifdef XML_NO_WIDE_CHAR |
---|
| 177 | char myIsTextWideChar(const void *b, int len) { return FALSE; } |
---|
| 178 | #else |
---|
| 179 | #if defined (UNDER_CE) || !defined(_XMLWINDOWS) |
---|
| 180 | char myIsTextWideChar(const void *b, int len) // inspired by the Wine API: RtlIsTextUnicode |
---|
| 181 | { |
---|
| 182 | #ifdef sun |
---|
| 183 | // for SPARC processors: wchar_t* buffers must always be alligned, otherwise it's a char* buffer. |
---|
| 184 | if ((((unsigned long)b)%sizeof(wchar_t))!=0) return FALSE; |
---|
| 185 | #endif |
---|
| 186 | const wchar_t *s=(const wchar_t*)b; |
---|
| 187 | |
---|
| 188 | // buffer too small: |
---|
| 189 | if (len<(int)sizeof(wchar_t)) return FALSE; |
---|
| 190 | |
---|
| 191 | // odd length test |
---|
| 192 | if (len&1) return FALSE; |
---|
| 193 | |
---|
| 194 | /* only checks the first 256 characters */ |
---|
| 195 | len=mmin(256,len/sizeof(wchar_t)); |
---|
| 196 | |
---|
| 197 | // Check for the special byte order: |
---|
| 198 | if (*((unsigned short*)s) == 0xFFFE) return TRUE; // IS_TEXT_UNICODE_REVERSE_SIGNATURE; |
---|
| 199 | if (*((unsigned short*)s) == 0xFEFF) return TRUE; // IS_TEXT_UNICODE_SIGNATURE |
---|
| 200 | |
---|
| 201 | // checks for ASCII characters in the UNICODE stream |
---|
| 202 | int i,stats=0; |
---|
| 203 | for (i=0; i<len; i++) if (s[i]<=(unsigned short)255) stats++; |
---|
| 204 | if (stats>len/2) return TRUE; |
---|
| 205 | |
---|
| 206 | // Check for UNICODE NULL chars |
---|
| 207 | for (i=0; i<len; i++) if (!s[i]) return TRUE; |
---|
| 208 | |
---|
| 209 | return FALSE; |
---|
| 210 | } |
---|
| 211 | #else |
---|
| 212 | char myIsTextWideChar(const void *b,int l) { return (char)IsTextUnicode((CONST LPVOID)b,l,NULL); }; |
---|
| 213 | #endif |
---|
| 214 | #endif |
---|
| 215 | |
---|
| 216 | #ifdef _XMLWINDOWS |
---|
| 217 | // for Microsoft Visual Studio 6.0 and Microsoft Visual Studio .NET, |
---|
| 218 | #ifdef _XMLWIDECHAR |
---|
| 219 | wchar_t *myMultiByteToWideChar(const char *s) |
---|
| 220 | { |
---|
| 221 | int i; |
---|
| 222 | if (characterEncoding==XMLNode::encoding_UTF8) i=(int)MultiByteToWideChar(CP_UTF8,0 ,s,-1,NULL,0); |
---|
| 223 | else i=(int)MultiByteToWideChar(CP_ACP ,MB_PRECOMPOSED,s,-1,NULL,0); |
---|
| 224 | if (i<0) return NULL; |
---|
| 225 | wchar_t *d=(wchar_t *)malloc((i+1)*sizeof(XMLCHAR)); |
---|
| 226 | if (characterEncoding==XMLNode::encoding_UTF8) i=(int)MultiByteToWideChar(CP_UTF8,0 ,s,-1,d,i); |
---|
| 227 | else i=(int)MultiByteToWideChar(CP_ACP ,MB_PRECOMPOSED,s,-1,d,i); |
---|
| 228 | d[i]=0; |
---|
| 229 | return d; |
---|
| 230 | } |
---|
| 231 | static inline FILE *xfopen(XMLCSTR filename,XMLCSTR mode) { return _wfopen(filename,mode); } |
---|
| 232 | static inline int xstrlen(XMLCSTR c) { return (int)wcslen(c); } |
---|
| 233 | static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return _wcsnicmp(c1,c2,l);} |
---|
| 234 | static inline int xstrncmp(XMLCSTR c1, XMLCSTR c2, int l) { return wcsncmp(c1,c2,l);} |
---|
| 235 | static inline int xstricmp(XMLCSTR c1, XMLCSTR c2) { return _wcsicmp(c1,c2); } |
---|
| 236 | static inline XMLSTR xstrstr(XMLCSTR c1, XMLCSTR c2) { return (XMLSTR)wcsstr(c1,c2); } |
---|
| 237 | static inline XMLSTR xstrcpy(XMLSTR c1, XMLCSTR c2) { return (XMLSTR)wcscpy(c1,c2); } |
---|
| 238 | #else |
---|
| 239 | char *myWideCharToMultiByte(const wchar_t *s) |
---|
| 240 | { |
---|
| 241 | UINT codePage=CP_ACP; if (characterEncoding==XMLNode::encoding_UTF8) codePage=CP_UTF8; |
---|
| 242 | int i=(int)WideCharToMultiByte(codePage, // code page |
---|
| 243 | 0, // performance and mapping flags |
---|
| 244 | s, // wide-character string |
---|
| 245 | -1, // number of chars in string |
---|
| 246 | NULL, // buffer for new string |
---|
| 247 | 0, // size of buffer |
---|
| 248 | NULL, // default for unmappable chars |
---|
| 249 | NULL // set when default char used |
---|
| 250 | ); |
---|
| 251 | if (i<0) return NULL; |
---|
| 252 | char *d=(char*)malloc(i+1); |
---|
| 253 | WideCharToMultiByte(codePage, // code page |
---|
| 254 | 0, // performance and mapping flags |
---|
| 255 | s, // wide-character string |
---|
| 256 | -1, // number of chars in string |
---|
| 257 | d, // buffer for new string |
---|
| 258 | i, // size of buffer |
---|
| 259 | NULL, // default for unmappable chars |
---|
| 260 | NULL // set when default char used |
---|
| 261 | ); |
---|
| 262 | d[i]=0; |
---|
| 263 | return d; |
---|
| 264 | } |
---|
| 265 | static inline FILE *xfopen(XMLCSTR filename,XMLCSTR mode) { return fopen(filename,mode); } |
---|
| 266 | static inline int xstrlen(XMLCSTR c) { return (int)strlen(c); } |
---|
| 267 | static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return _strnicmp(c1,c2,l);} |
---|
| 268 | static inline int xstrncmp(XMLCSTR c1, XMLCSTR c2, int l) { return strncmp(c1,c2,l);} |
---|
| 269 | static inline int xstricmp(XMLCSTR c1, XMLCSTR c2) { return _stricmp(c1,c2); } |
---|
| 270 | static inline XMLSTR xstrstr(XMLCSTR c1, XMLCSTR c2) { return (XMLSTR)strstr(c1,c2); } |
---|
| 271 | static inline XMLSTR xstrcpy(XMLSTR c1, XMLCSTR c2) { return (XMLSTR)strcpy(c1,c2); } |
---|
| 272 | #endif |
---|
| 273 | #ifdef __BORLANDC__ |
---|
| 274 | static inline int _strnicmp(char *c1, char *c2, int l){ return strnicmp(c1,c2,l);} |
---|
| 275 | #endif |
---|
| 276 | #else |
---|
| 277 | // for gcc and CC |
---|
| 278 | #ifdef XML_NO_WIDE_CHAR |
---|
| 279 | char *myWideCharToMultiByte(const wchar_t *s) { return NULL; } |
---|
| 280 | #else |
---|
| 281 | char *myWideCharToMultiByte(const wchar_t *s) |
---|
| 282 | { |
---|
| 283 | const wchar_t *ss=s; |
---|
| 284 | int i=(int)wcsrtombs(NULL,&ss,0,NULL); |
---|
| 285 | if (i<0) return NULL; |
---|
| 286 | char *d=(char *)malloc(i+1); |
---|
| 287 | wcsrtombs(d,&s,i,NULL); |
---|
| 288 | d[i]=0; |
---|
| 289 | return d; |
---|
| 290 | } |
---|
| 291 | #endif |
---|
| 292 | #ifdef _XMLWIDECHAR |
---|
| 293 | wchar_t *myMultiByteToWideChar(const char *s) |
---|
| 294 | { |
---|
| 295 | const char *ss=s; |
---|
| 296 | int i=(int)mbsrtowcs(NULL,&ss,0,NULL); |
---|
| 297 | if (i<0) return NULL; |
---|
| 298 | wchar_t *d=(wchar_t *)malloc((i+1)*sizeof(wchar_t)); |
---|
| 299 | mbsrtowcs(d,&s,i,NULL); |
---|
| 300 | d[i]=0; |
---|
| 301 | return d; |
---|
| 302 | } |
---|
| 303 | int xstrlen(XMLCSTR c) { return wcslen(c); } |
---|
| 304 | #ifdef sun |
---|
| 305 | // for CC |
---|
| 306 | #include <widec.h> |
---|
| 307 | static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return wsncasecmp(c1,c2,l);} |
---|
| 308 | static inline int xstrncmp(XMLCSTR c1, XMLCSTR c2, int l) { return wsncmp(c1,c2,l);} |
---|
| 309 | static inline int xstricmp(XMLCSTR c1, XMLCSTR c2) { return wscasecmp(c1,c2); } |
---|
| 310 | #else |
---|
| 311 | // for gcc |
---|
| 312 | static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return wcsncasecmp(c1,c2,l);} |
---|
| 313 | static inline int xstrncmp(XMLCSTR c1, XMLCSTR c2, int l) { return wcsncmp(c1,c2,l);} |
---|
| 314 | static inline int xstricmp(XMLCSTR c1, XMLCSTR c2) { return wcscasecmp(c1,c2); } |
---|
| 315 | #endif |
---|
| 316 | static inline XMLSTR xstrstr(XMLCSTR c1, XMLCSTR c2) { return (XMLSTR)wcsstr(c1,c2); } |
---|
| 317 | static inline XMLSTR xstrcpy(XMLSTR c1, XMLCSTR c2) { return (XMLSTR)wcscpy(c1,c2); } |
---|
| 318 | static inline FILE *xfopen(XMLCSTR filename,XMLCSTR mode) |
---|
| 319 | { |
---|
| 320 | char *filenameAscii=myWideCharToMultiByte(filename); |
---|
| 321 | FILE *f; |
---|
| 322 | if (mode[0]==_X('r')) f=fopen(filenameAscii,"rb"); |
---|
| 323 | else f=fopen(filenameAscii,"wb"); |
---|
| 324 | free(filenameAscii); |
---|
| 325 | return f; |
---|
| 326 | } |
---|
| 327 | #else |
---|
| 328 | static inline FILE *xfopen(XMLCSTR filename,XMLCSTR mode) { return fopen(filename,mode); } |
---|
| 329 | static inline int xstrlen(XMLCSTR c) { return strlen(c); } |
---|
| 330 | static inline int xstrnicmp(XMLCSTR c1, XMLCSTR c2, int l) { return strncasecmp(c1,c2,l);} |
---|
| 331 | static inline int xstrncmp(XMLCSTR c1, XMLCSTR c2, int l) { return strncmp(c1,c2,l);} |
---|
| 332 | static inline int xstricmp(XMLCSTR c1, XMLCSTR c2) { return strcasecmp(c1,c2); } |
---|
| 333 | static inline XMLSTR xstrstr(XMLCSTR c1, XMLCSTR c2) { return (XMLSTR)strstr(c1,c2); } |
---|
| 334 | static inline XMLSTR xstrcpy(XMLSTR c1, XMLCSTR c2) { return (XMLSTR)strcpy(c1,c2); } |
---|
| 335 | #endif |
---|
| 336 | static inline int _strnicmp(const char *c1,const char *c2, int l) { return strncasecmp(c1,c2,l);} |
---|
| 337 | #endif |
---|
| 338 | |
---|
| 339 | ///////////////////////////////////////////////////////////////////////// |
---|
| 340 | // the "openFileHelper" function // |
---|
| 341 | ///////////////////////////////////////////////////////////////////////// |
---|
| 342 | |
---|
| 343 | // Since each application has its own way to report and deal with errors, you should modify & rewrite |
---|
| 344 | // the following "openFileHelper" function to get an "error reporting mechanism" tailored to your needs. |
---|
| 345 | XMLNode XMLNode::openFileHelper(XMLCSTR filename, XMLCSTR tag) |
---|
| 346 | { |
---|
| 347 | // guess the value of the global parameter "characterEncoding" |
---|
| 348 | // (the guess is based on the first 200 bytes of the file). |
---|
| 349 | FILE *f=xfopen(filename,_X("rb")); |
---|
| 350 | if (f) |
---|
| 351 | { |
---|
| 352 | char bb[205]; |
---|
| 353 | int l=(int)fread(bb,1,200,f); |
---|
| 354 | setGlobalOptions(guessCharEncoding(bb,l),guessWideCharChars,dropWhiteSpace); |
---|
| 355 | fclose(f); |
---|
| 356 | } |
---|
| 357 | |
---|
| 358 | // parse the file |
---|
| 359 | XMLResults pResults; |
---|
| 360 | XMLNode xnode=XMLNode::parseFile(filename,tag,&pResults); |
---|
| 361 | |
---|
| 362 | // display error message (if any) |
---|
| 363 | if (pResults.error != eXMLErrorNone) |
---|
| 364 | { |
---|
| 365 | // create message |
---|
| 366 | char message[2000],*s1=(char*)"",*s3=(char*)""; XMLCSTR s2=_X(""); |
---|
| 367 | if (pResults.error==eXMLErrorFirstTagNotFound) { s1=(char*)"First Tag should be '"; s2=tag; s3=(char*)"'.\n"; } |
---|
| 368 | sprintf(message, |
---|
| 369 | #ifdef _XMLWIDECHAR |
---|
| 370 | "XML Parsing error inside file '%S'.\n%S\nAt line %i, column %i.\n%s%S%s" |
---|
| 371 | #else |
---|
| 372 | "XML Parsing error inside file '%s'.\n%s\nAt line %i, column %i.\n%s%s%s" |
---|
| 373 | #endif |
---|
| 374 | ,filename,XMLNode::getError(pResults.error),pResults.nLine,pResults.nColumn,s1,s2,s3); |
---|
| 375 | |
---|
| 376 | // display message |
---|
| 377 | #if defined(_XMLWINDOWS) && !defined(UNDER_CE) && !defined(_XMLPARSER_NO_MESSAGEBOX_) |
---|
| 378 | MessageBoxA(NULL,message,"XML Parsing error",MB_OK|MB_ICONERROR|MB_TOPMOST); |
---|
| 379 | #else |
---|
| 380 | printf("%s",message); |
---|
| 381 | #endif |
---|
| 382 | exit(255); |
---|
| 383 | } |
---|
| 384 | return xnode; |
---|
| 385 | } |
---|
| 386 | |
---|
| 387 | ///////////////////////////////////////////////////////////////////////// |
---|
| 388 | // Here start the core implementation of the XMLParser library // |
---|
| 389 | ///////////////////////////////////////////////////////////////////////// |
---|
| 390 | |
---|
| 391 | // You should normally not change anything below this point. |
---|
| 392 | |
---|
| 393 | #ifndef _XMLWIDECHAR |
---|
| 394 | // If "characterEncoding=ascii" then we assume that all characters have the same length of 1 byte. |
---|
| 395 | // If "characterEncoding=UTF8" then the characters have different lengths (from 1 byte to 4 bytes). |
---|
| 396 | // If "characterEncoding=ShiftJIS" then the characters have different lengths (from 1 byte to 2 bytes). |
---|
| 397 | // This table is used as lookup-table to know the length of a character (in byte) based on the |
---|
| 398 | // content of the first byte of the character. |
---|
| 399 | // (note: if you modify this, you must always have XML_utf8ByteTable[0]=0 ). |
---|
| 400 | static const char XML_utf8ByteTable[256] = |
---|
| 401 | { |
---|
| 402 | // 0 1 2 3 4 5 6 7 8 9 a b c d e f |
---|
| 403 | 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x00 |
---|
| 404 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x10 |
---|
| 405 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x20 |
---|
| 406 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x30 |
---|
| 407 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x40 |
---|
| 408 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x50 |
---|
| 409 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x60 |
---|
| 410 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x70End of ASCII range |
---|
| 411 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x80 0x80 to 0xc1 invalid |
---|
| 412 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x90 |
---|
| 413 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xa0 |
---|
| 414 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xb0 |
---|
| 415 | 1,1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xc0 0xc2 to 0xdf 2 byte |
---|
| 416 | 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xd0 |
---|
| 417 | 3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,3,// 0xe0 0xe0 to 0xef 3 byte |
---|
| 418 | 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 |
---|
| 419 | }; |
---|
| 420 | static const char XML_asciiByteTable[256] = |
---|
| 421 | { |
---|
| 422 | 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, |
---|
| 423 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
---|
| 424 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
---|
| 425 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
---|
| 426 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1, |
---|
| 427 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1 |
---|
| 428 | }; |
---|
| 429 | static const char XML_sjisByteTable[256] = |
---|
| 430 | { |
---|
| 431 | // 0 1 2 3 4 5 6 7 8 9 a b c d e f |
---|
| 432 | 0,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x00 |
---|
| 433 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x10 |
---|
| 434 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x20 |
---|
| 435 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x30 |
---|
| 436 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x40 |
---|
| 437 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x50 |
---|
| 438 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x60 |
---|
| 439 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0x70 End of ASCII range |
---|
| 440 | 1,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0x80 0x81 to 0x9F 2 bytes |
---|
| 441 | 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0x90 |
---|
| 442 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xa0 |
---|
| 443 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xb0 |
---|
| 444 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xc0 |
---|
| 445 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,// 0xd0 |
---|
| 446 | 2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,// 0xe0 0xe0 to 0xef 2 bytes |
---|
| 447 | 1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1 // 0xf0 |
---|
| 448 | }; |
---|
| 449 | static const char *XML_ByteTable=(const char *)XML_utf8ByteTable; // the default is "characterEncoding=XMLNode::encoding_UTF8" |
---|
| 450 | #endif |
---|
| 451 | |
---|
| 452 | |
---|
| 453 | XMLNode XMLNode::emptyXMLNode; |
---|
| 454 | XMLClear XMLNode::emptyXMLClear={ NULL, NULL, NULL}; |
---|
| 455 | XMLAttribute XMLNode::emptyXMLAttribute={ NULL, NULL}; |
---|
| 456 | |
---|
| 457 | // Enumeration used to decipher what type a token is |
---|
| 458 | typedef enum XMLTokenTypeTag |
---|
| 459 | { |
---|
| 460 | eTokenText = 0, |
---|
| 461 | eTokenQuotedText, |
---|
| 462 | eTokenTagStart, /* "<" */ |
---|
| 463 | eTokenTagEnd, /* "</" */ |
---|
| 464 | eTokenCloseTag, /* ">" */ |
---|
| 465 | eTokenEquals, /* "=" */ |
---|
| 466 | eTokenDeclaration, /* "<?" */ |
---|
| 467 | eTokenShortHandClose, /* "/>" */ |
---|
| 468 | eTokenClear, |
---|
| 469 | eTokenError |
---|
| 470 | } XMLTokenType; |
---|
| 471 | |
---|
| 472 | // Main structure used for parsing XML |
---|
| 473 | typedef struct XML |
---|
| 474 | { |
---|
| 475 | XMLCSTR lpXML; |
---|
| 476 | XMLCSTR lpszText; |
---|
| 477 | int nIndex,nIndexMissigEndTag; |
---|
| 478 | enum XMLError error; |
---|
| 479 | XMLCSTR lpEndTag; |
---|
| 480 | int cbEndTag; |
---|
| 481 | XMLCSTR lpNewElement; |
---|
| 482 | int cbNewElement; |
---|
| 483 | int nFirst; |
---|
| 484 | } XML; |
---|
| 485 | |
---|
| 486 | typedef struct |
---|
| 487 | { |
---|
| 488 | ALLXMLClearTag *pClr; |
---|
| 489 | XMLCSTR pStr; |
---|
| 490 | } NextToken; |
---|
| 491 | |
---|
| 492 | // Enumeration used when parsing attributes |
---|
| 493 | typedef enum Attrib |
---|
| 494 | { |
---|
| 495 | eAttribName = 0, |
---|
| 496 | eAttribEquals, |
---|
| 497 | eAttribValue |
---|
| 498 | } Attrib; |
---|
| 499 | |
---|
| 500 | // Enumeration used when parsing elements to dictate whether we are currently |
---|
| 501 | // inside a tag |
---|
| 502 | typedef enum Status |
---|
| 503 | { |
---|
| 504 | eInsideTag = 0, |
---|
| 505 | eOutsideTag |
---|
| 506 | } Status; |
---|
| 507 | |
---|
| 508 | XMLError XMLNode::writeToFile(XMLCSTR filename, const char *encoding, char nFormat) const |
---|
| 509 | { |
---|
| 510 | if (!d) return eXMLErrorNone; |
---|
| 511 | FILE *f=xfopen(filename,_X("wb")); |
---|
| 512 | if (!f) return eXMLErrorCannotOpenWriteFile; |
---|
| 513 | #ifdef _XMLWIDECHAR |
---|
| 514 | unsigned char h[2]={ 0xFF, 0xFE }; |
---|
| 515 | if (!fwrite(h,2,1,f)) return eXMLErrorCannotWriteFile; |
---|
| 516 | if ((!isDeclaration())&&((d->lpszName)||(!getChildNode().isDeclaration()))) |
---|
| 517 | { |
---|
| 518 | if (!fwrite(_X("<?xml version=\"1.0\" encoding=\"utf-16\"?>\n"),sizeof(wchar_t)*40,1,f)) |
---|
| 519 | return eXMLErrorCannotWriteFile; |
---|
| 520 | } |
---|
| 521 | #else |
---|
| 522 | if ((!isDeclaration())&&((d->lpszName)||(!getChildNode().isDeclaration()))) |
---|
| 523 | { |
---|
| 524 | if (characterEncoding==encoding_UTF8) |
---|
| 525 | { |
---|
| 526 | // header so that windows recognize the file as UTF-8: |
---|
| 527 | unsigned char h[3]={0xEF,0xBB,0xBF}; if (!fwrite(h,3,1,f)) return eXMLErrorCannotWriteFile; |
---|
| 528 | encoding="utf-8"; |
---|
| 529 | } else if (characterEncoding==encoding_ShiftJIS) encoding="SHIFT-JIS"; |
---|
| 530 | |
---|
| 531 | if (!encoding) encoding="ISO-8859-1"; |
---|
| 532 | if (fprintf(f,"<?xml version=\"1.0\" encoding=\"%s\"?>\n",encoding)<0) return eXMLErrorCannotWriteFile; |
---|
| 533 | } else |
---|
| 534 | { |
---|
| 535 | if (characterEncoding==encoding_UTF8) |
---|
| 536 | { |
---|
| 537 | unsigned char h[3]={0xEF,0xBB,0xBF}; if (!fwrite(h,3,1,f)) return eXMLErrorCannotWriteFile; |
---|
| 538 | } |
---|
| 539 | } |
---|
| 540 | #endif |
---|
| 541 | int i; |
---|
| 542 | XMLSTR t=createXMLString(nFormat,&i); |
---|
| 543 | if (!fwrite(t,sizeof(XMLCHAR)*i,1,f)) return eXMLErrorCannotWriteFile; |
---|
| 544 | if (fclose(f)!=0) return eXMLErrorCannotWriteFile; |
---|
| 545 | free(t); |
---|
| 546 | return eXMLErrorNone; |
---|
| 547 | } |
---|
| 548 | |
---|
| 549 | // Duplicate a given string. |
---|
| 550 | XMLSTR stringDup(XMLCSTR lpszData, int cbData) |
---|
| 551 | { |
---|
| 552 | if (lpszData==NULL) return NULL; |
---|
| 553 | |
---|
| 554 | XMLSTR lpszNew; |
---|
| 555 | if (cbData==0) cbData=(int)xstrlen(lpszData); |
---|
| 556 | lpszNew = (XMLSTR)malloc((cbData+1) * sizeof(XMLCHAR)); |
---|
| 557 | if (lpszNew) |
---|
| 558 | { |
---|
| 559 | memcpy(lpszNew, lpszData, (cbData) * sizeof(XMLCHAR)); |
---|
| 560 | lpszNew[cbData] = (XMLCHAR)NULL; |
---|
| 561 | } |
---|
| 562 | return lpszNew; |
---|
| 563 | } |
---|
| 564 | |
---|
| 565 | XMLSTR toXMLStringUnSafe(XMLSTR dest,XMLCSTR source) |
---|
| 566 | { |
---|
| 567 | XMLSTR dd=dest; |
---|
| 568 | XMLCHAR ch; |
---|
| 569 | XMLCharacterEntity *entity; |
---|
| 570 | while ((ch=*source)) |
---|
| 571 | { |
---|
| 572 | entity=XMLEntities; |
---|
| 573 | do |
---|
| 574 | { |
---|
| 575 | if (ch==entity->c) {xstrcpy(dest,entity->s); dest+=entity->l; source++; goto out_of_loop1; } |
---|
| 576 | entity++; |
---|
| 577 | } while(entity->s); |
---|
| 578 | #ifdef _XMLWIDECHAR |
---|
| 579 | *(dest++)=*(source++); |
---|
| 580 | #else |
---|
| 581 | switch(XML_ByteTable[(unsigned char)ch]) |
---|
| 582 | { |
---|
| 583 | case 4: *(dest++)=*(source++); |
---|
| 584 | case 3: *(dest++)=*(source++); |
---|
| 585 | case 2: *(dest++)=*(source++); |
---|
| 586 | case 1: *(dest++)=*(source++); |
---|
| 587 | } |
---|
| 588 | #endif |
---|
| 589 | out_of_loop1: |
---|
| 590 | ; |
---|
| 591 | } |
---|
| 592 | *dest=0; |
---|
| 593 | return dd; |
---|
| 594 | } |
---|
| 595 | |
---|
| 596 | // private (used while rendering): |
---|
| 597 | int lengthXMLString(XMLCSTR source) |
---|
| 598 | { |
---|
| 599 | int r=0; |
---|
| 600 | XMLCharacterEntity *entity; |
---|
| 601 | XMLCHAR ch; |
---|
| 602 | while ((ch=*source)) |
---|
| 603 | { |
---|
| 604 | entity=XMLEntities; |
---|
| 605 | do |
---|
| 606 | { |
---|
| 607 | if (ch==entity->c) { r+=entity->l; source++; goto out_of_loop1; } |
---|
| 608 | entity++; |
---|
| 609 | } while(entity->s); |
---|
| 610 | #ifdef _XMLWIDECHAR |
---|
| 611 | r++; source++; |
---|
| 612 | #else |
---|
| 613 | ch=XML_ByteTable[(unsigned char)ch]; r+=ch; source+=ch; |
---|
| 614 | #endif |
---|
| 615 | out_of_loop1: |
---|
| 616 | ; |
---|
| 617 | } |
---|
| 618 | return r; |
---|
| 619 | } |
---|
| 620 | |
---|
| 621 | ToXMLStringTool::~ToXMLStringTool(){ freeBuffer(); } |
---|
| 622 | void ToXMLStringTool::freeBuffer(){ if (buf) free(buf); buf=NULL; buflen=0; } |
---|
| 623 | XMLSTR ToXMLStringTool::toXML(XMLCSTR source) |
---|
| 624 | { |
---|
| 625 | int l=lengthXMLString(source)+1; |
---|
| 626 | if (l>buflen) { buflen=l; buf=(XMLSTR)realloc(buf,l*sizeof(XMLCHAR)); } |
---|
| 627 | return toXMLStringUnSafe(buf,source); |
---|
| 628 | } |
---|
| 629 | |
---|
| 630 | // private: |
---|
| 631 | XMLSTR fromXMLString(XMLCSTR s, int lo, XML *pXML) |
---|
| 632 | { |
---|
| 633 | // This function is the opposite of the function "toXMLString". It decodes the escape |
---|
| 634 | // sequences &, ", ', <, > and replace them by the characters |
---|
| 635 | // &,",',<,>. This function is used internally by the XML Parser. All the calls to |
---|
| 636 | // the XML library will always gives you back "decoded" strings. |
---|
| 637 | // |
---|
| 638 | // in: string (s) and length (lo) of string |
---|
| 639 | // out: new allocated string converted from xml |
---|
| 640 | if (!s) return NULL; |
---|
| 641 | |
---|
| 642 | int ll=0,j; |
---|
| 643 | XMLSTR d; |
---|
| 644 | XMLCSTR ss=s; |
---|
| 645 | XMLCharacterEntity *entity; |
---|
| 646 | while ((lo>0)&&(*s)) |
---|
| 647 | { |
---|
| 648 | if (*s==_X('&')) |
---|
| 649 | { |
---|
| 650 | if ((lo>2)&&(s[1]==_X('#'))) |
---|
| 651 | { |
---|
| 652 | s+=2; lo-=2; |
---|
| 653 | if ((*s==_X('X'))||(*s==_X('x'))) { s++; lo--; } |
---|
| 654 | while ((*s)&&(*s!=_X(';'))&&((lo--)>0)) s++; |
---|
| 655 | if (*s!=_X(';')) |
---|
| 656 | { |
---|
| 657 | pXML->error=eXMLErrorUnknownCharacterEntity; |
---|
| 658 | return NULL; |
---|
| 659 | } |
---|
| 660 | s++; lo--; |
---|
| 661 | } else |
---|
| 662 | { |
---|
| 663 | entity=XMLEntities; |
---|
| 664 | do |
---|
| 665 | { |
---|
| 666 | if ((lo>=entity->l)&&(xstrnicmp(s,entity->s,entity->l)==0)) { s+=entity->l; lo-=entity->l; break; } |
---|
| 667 | entity++; |
---|
| 668 | } while(entity->s); |
---|
| 669 | if (!entity->s) |
---|
| 670 | { |
---|
| 671 | pXML->error=eXMLErrorUnknownCharacterEntity; |
---|
| 672 | return NULL; |
---|
| 673 | } |
---|
| 674 | } |
---|
| 675 | } else |
---|
| 676 | { |
---|
| 677 | #ifdef _XMLWIDECHAR |
---|
| 678 | s++; lo--; |
---|
| 679 | #else |
---|
| 680 | j=XML_ByteTable[(unsigned char)*s]; s+=j; lo-=j; ll+=j-1; |
---|
| 681 | #endif |
---|
| 682 | } |
---|
| 683 | ll++; |
---|
| 684 | } |
---|
| 685 | |
---|
| 686 | d=(XMLSTR)malloc((ll+1)*sizeof(XMLCHAR)); |
---|
| 687 | s=d; |
---|
| 688 | while (ll-->0) |
---|
| 689 | { |
---|
| 690 | if (*ss==_X('&')) |
---|
| 691 | { |
---|
| 692 | if (ss[1]==_X('#')) |
---|
| 693 | { |
---|
| 694 | ss+=2; j=0; |
---|
| 695 | if ((*ss==_X('X'))||(*ss==_X('x'))) |
---|
| 696 | { |
---|
| 697 | ss++; |
---|
| 698 | while (*ss!=_X(';')) |
---|
| 699 | { |
---|
| 700 | if ((*ss>=_X('0'))&&(*ss<=_X('9'))) j=(j<<4)+*ss-_X('0'); |
---|
| 701 | else if ((*ss>=_X('A'))&&(*ss<=_X('F'))) j=(j<<4)+*ss-_X('A')+10; |
---|
| 702 | else if ((*ss>=_X('a'))&&(*ss<=_X('f'))) j=(j<<4)+*ss-_X('a')+10; |
---|
| 703 | else { free((void*)s); pXML->error=eXMLErrorUnknownCharacterEntity;return NULL;} |
---|
| 704 | ss++; |
---|
| 705 | } |
---|
| 706 | } else |
---|
| 707 | { |
---|
| 708 | while (*ss!=_X(';')) |
---|
| 709 | { |
---|
| 710 | if ((*ss>=_X('0'))&&(*ss<=_X('9'))) j=(j*10)+*ss-_X('0'); |
---|
| 711 | else { free((void*)s); pXML->error=eXMLErrorUnknownCharacterEntity;return NULL;} |
---|
| 712 | ss++; |
---|
| 713 | } |
---|
| 714 | } |
---|
| 715 | (*d++)=(XMLCHAR)j; ss++; |
---|
| 716 | } else |
---|
| 717 | { |
---|
| 718 | entity=XMLEntities; |
---|
| 719 | do |
---|
| 720 | { |
---|
| 721 | if (xstrnicmp(ss,entity->s,entity->l)==0) { *(d++)=entity->c; ss+=entity->l; break; } |
---|
| 722 | entity++; |
---|
| 723 | } while(entity->s); |
---|
| 724 | } |
---|
| 725 | } else |
---|
| 726 | { |
---|
| 727 | #ifdef _XMLWIDECHAR |
---|
| 728 | *(d++)=*(ss++); |
---|
| 729 | #else |
---|
| 730 | switch(XML_ByteTable[(unsigned char)*ss]) |
---|
| 731 | { |
---|
| 732 | case 4: *(d++)=*(ss++); ll--; |
---|
| 733 | case 3: *(d++)=*(ss++); ll--; |
---|
| 734 | case 2: *(d++)=*(ss++); ll--; |
---|
| 735 | case 1: *(d++)=*(ss++); |
---|
| 736 | } |
---|
| 737 | #endif |
---|
| 738 | } |
---|
| 739 | } |
---|
| 740 | *d=0; |
---|
| 741 | return (XMLSTR)s; |
---|
| 742 | } |
---|
| 743 | |
---|
| 744 | #define XML_isSPACECHAR(ch) ((ch==_X('\n'))||(ch==_X(' '))||(ch== _X('\t'))||(ch==_X('\r'))) |
---|
| 745 | |
---|
| 746 | // private: |
---|
| 747 | char myTagCompare(XMLCSTR cclose, XMLCSTR copen) |
---|
| 748 | // !!!! WARNING strange convention&: |
---|
| 749 | // return 0 if equals |
---|
| 750 | // return 1 if different |
---|
| 751 | { |
---|
| 752 | if (!cclose) return 1; |
---|
| 753 | int l=(int)xstrlen(cclose); |
---|
| 754 | if (xstrnicmp(cclose, copen, l)!=0) return 1; |
---|
| 755 | const XMLCHAR c=copen[l]; |
---|
| 756 | if (XML_isSPACECHAR(c)|| |
---|
| 757 | (c==_X('/' ))|| |
---|
| 758 | (c==_X('<' ))|| |
---|
| 759 | (c==_X('>' ))|| |
---|
| 760 | (c==_X('=' ))) return 0; |
---|
| 761 | return 1; |
---|
| 762 | } |
---|
| 763 | |
---|
| 764 | // Obtain the next character from the string. |
---|
| 765 | static inline XMLCHAR getNextChar(XML *pXML) |
---|
| 766 | { |
---|
| 767 | XMLCHAR ch = pXML->lpXML[pXML->nIndex]; |
---|
| 768 | #ifdef _XMLWIDECHAR |
---|
| 769 | if (ch!=0) pXML->nIndex++; |
---|
| 770 | #else |
---|
| 771 | pXML->nIndex+=XML_ByteTable[(unsigned char)ch]; |
---|
| 772 | #endif |
---|
| 773 | return ch; |
---|
| 774 | } |
---|
| 775 | |
---|
| 776 | // Find the next token in a string. |
---|
| 777 | // pcbToken contains the number of characters that have been read. |
---|
| 778 | static NextToken GetNextToken(XML *pXML, int *pcbToken, enum XMLTokenTypeTag *pType) |
---|
| 779 | { |
---|
| 780 | NextToken result; |
---|
| 781 | XMLCHAR ch; |
---|
| 782 | XMLCHAR chTemp; |
---|
| 783 | int indexStart,nFoundMatch,nIsText=FALSE; |
---|
| 784 | result.pClr=NULL; // prevent warning |
---|
| 785 | |
---|
| 786 | // Find next non-white space character |
---|
| 787 | do { indexStart=pXML->nIndex; ch=getNextChar(pXML); } while XML_isSPACECHAR(ch); |
---|
| 788 | |
---|
| 789 | if (ch) |
---|
| 790 | { |
---|
| 791 | // Cache the current string pointer |
---|
| 792 | result.pStr = &pXML->lpXML[indexStart]; |
---|
| 793 | |
---|
| 794 | // First check whether the token is in the clear tag list (meaning it |
---|
| 795 | // does not need formatting). |
---|
| 796 | ALLXMLClearTag *ctag=XMLClearTags; |
---|
| 797 | do |
---|
| 798 | { |
---|
| 799 | if (xstrncmp(ctag->lpszOpen, result.pStr, ctag->openTagLen)==0) |
---|
| 800 | { |
---|
| 801 | result.pClr=ctag; |
---|
| 802 | pXML->nIndex+=ctag->openTagLen-1; |
---|
| 803 | *pType=eTokenClear; |
---|
| 804 | return result; |
---|
| 805 | } |
---|
| 806 | ctag++; |
---|
| 807 | } while(ctag->lpszOpen); |
---|
| 808 | |
---|
| 809 | // If we didn't find a clear tag then check for standard tokens |
---|
| 810 | switch(ch) |
---|
| 811 | { |
---|
| 812 | // Check for quotes |
---|
| 813 | case _X('\''): |
---|
| 814 | case _X('\"'): |
---|
| 815 | // Type of token |
---|
| 816 | *pType = eTokenQuotedText; |
---|
| 817 | chTemp = ch; |
---|
| 818 | |
---|
| 819 | // Set the size |
---|
| 820 | nFoundMatch = FALSE; |
---|
| 821 | |
---|
| 822 | // Search through the string to find a matching quote |
---|
| 823 | while((ch = getNextChar(pXML))) |
---|
| 824 | { |
---|
| 825 | if (ch==chTemp) { nFoundMatch = TRUE; break; } |
---|
| 826 | if (ch==_X('<')) break; |
---|
| 827 | } |
---|
| 828 | |
---|
| 829 | // If we failed to find a matching quote |
---|
| 830 | if (nFoundMatch == FALSE) |
---|
| 831 | { |
---|
| 832 | pXML->nIndex=indexStart+1; |
---|
| 833 | nIsText=TRUE; |
---|
| 834 | break; |
---|
| 835 | } |
---|
| 836 | |
---|
| 837 | // 4.02.2002 |
---|
| 838 | // if (FindNonWhiteSpace(pXML)) pXML->nIndex--; |
---|
| 839 | |
---|
| 840 | break; |
---|
| 841 | |
---|
| 842 | // Equals (used with attribute values) |
---|
| 843 | case _X('='): |
---|
| 844 | *pType = eTokenEquals; |
---|
| 845 | break; |
---|
| 846 | |
---|
| 847 | // Close tag |
---|
| 848 | case _X('>'): |
---|
| 849 | *pType = eTokenCloseTag; |
---|
| 850 | break; |
---|
| 851 | |
---|
| 852 | // Check for tag start and tag end |
---|
| 853 | case _X('<'): |
---|
| 854 | |
---|
| 855 | // Peek at the next character to see if we have an end tag '</', |
---|
| 856 | // or an xml declaration '<?' |
---|
| 857 | chTemp = pXML->lpXML[pXML->nIndex]; |
---|
| 858 | |
---|
| 859 | // If we have a tag end... |
---|
| 860 | if (chTemp == _X('/')) |
---|
| 861 | { |
---|
| 862 | // Set the type and ensure we point at the next character |
---|
| 863 | getNextChar(pXML); |
---|
| 864 | *pType = eTokenTagEnd; |
---|
| 865 | } |
---|
| 866 | |
---|
| 867 | // If we have an XML declaration tag |
---|
| 868 | else if (chTemp == _X('?')) |
---|
| 869 | { |
---|
| 870 | |
---|
| 871 | // Set the type and ensure we point at the next character |
---|
| 872 | getNextChar(pXML); |
---|
| 873 | *pType = eTokenDeclaration; |
---|
| 874 | } |
---|
| 875 | |
---|
| 876 | // Otherwise we must have a start tag |
---|
| 877 | else |
---|
| 878 | { |
---|
| 879 | *pType = eTokenTagStart; |
---|
| 880 | } |
---|
| 881 | break; |
---|
| 882 | |
---|
| 883 | // Check to see if we have a short hand type end tag ('/>'). |
---|
| 884 | case _X('/'): |
---|
| 885 | |
---|
| 886 | // Peek at the next character to see if we have a short end tag '/>' |
---|
| 887 | chTemp = pXML->lpXML[pXML->nIndex]; |
---|
| 888 | |
---|
| 889 | // If we have a short hand end tag... |
---|
| 890 | if (chTemp == _X('>')) |
---|
| 891 | { |
---|
| 892 | // Set the type and ensure we point at the next character |
---|
| 893 | getNextChar(pXML); |
---|
| 894 | *pType = eTokenShortHandClose; |
---|
| 895 | break; |
---|
| 896 | } |
---|
| 897 | |
---|
| 898 | // If we haven't found a short hand closing tag then drop into the |
---|
| 899 | // text process |
---|
| 900 | |
---|
| 901 | // Other characters |
---|
| 902 | default: |
---|
| 903 | nIsText = TRUE; |
---|
| 904 | } |
---|
| 905 | |
---|
| 906 | // If this is a TEXT node |
---|
| 907 | if (nIsText) |
---|
| 908 | { |
---|
| 909 | // Indicate we are dealing with text |
---|
| 910 | *pType = eTokenText; |
---|
| 911 | while((ch = getNextChar(pXML))) |
---|
| 912 | { |
---|
| 913 | if XML_isSPACECHAR(ch) |
---|
| 914 | { |
---|
| 915 | indexStart++; break; |
---|
| 916 | |
---|
| 917 | } else if (ch==_X('/')) |
---|
| 918 | { |
---|
| 919 | // If we find a slash then this maybe text or a short hand end tag |
---|
| 920 | // Peek at the next character to see it we have short hand end tag |
---|
| 921 | ch=pXML->lpXML[pXML->nIndex]; |
---|
| 922 | // If we found a short hand end tag then we need to exit the loop |
---|
| 923 | if (ch==_X('>')) { pXML->nIndex--; break; } |
---|
| 924 | |
---|
| 925 | } else if ((ch==_X('<'))||(ch==_X('>'))||(ch==_X('='))) |
---|
| 926 | { |
---|
| 927 | pXML->nIndex--; break; |
---|
| 928 | } |
---|
| 929 | } |
---|
| 930 | } |
---|
| 931 | *pcbToken = pXML->nIndex-indexStart; |
---|
| 932 | } else |
---|
| 933 | { |
---|
| 934 | // If we failed to obtain a valid character |
---|
| 935 | *pcbToken = 0; |
---|
| 936 | *pType = eTokenError; |
---|
| 937 | result.pStr=NULL; |
---|
| 938 | } |
---|
| 939 | |
---|
| 940 | return result; |
---|
| 941 | } |
---|
| 942 | |
---|
| 943 | XMLCSTR XMLNode::updateName_WOSD(XMLSTR lpszName) |
---|
| 944 | { |
---|
| 945 | if (!d) { free(lpszName); return NULL; } |
---|
| 946 | if (d->lpszName&&(lpszName!=d->lpszName)) free((void*)d->lpszName); |
---|
| 947 | d->lpszName=lpszName; |
---|
| 948 | return lpszName; |
---|
| 949 | } |
---|
| 950 | |
---|
| 951 | // private: |
---|
| 952 | XMLNode::XMLNode(struct XMLNodeDataTag *p){ d=p; (p->ref_count)++; } |
---|
| 953 | XMLNode::XMLNode(XMLNodeData *pParent, XMLSTR lpszName, char isDeclaration) |
---|
| 954 | { |
---|
| 955 | d=(XMLNodeData*)malloc(sizeof(XMLNodeData)); |
---|
| 956 | d->ref_count=1; |
---|
| 957 | |
---|
| 958 | d->lpszName=NULL; |
---|
| 959 | d->nChild= 0; |
---|
| 960 | d->nText = 0; |
---|
| 961 | d->nClear = 0; |
---|
| 962 | d->nAttribute = 0; |
---|
| 963 | |
---|
| 964 | d->isDeclaration = isDeclaration; |
---|
| 965 | |
---|
| 966 | d->pParent = pParent; |
---|
| 967 | d->pChild= NULL; |
---|
| 968 | d->pText= NULL; |
---|
| 969 | d->pClear= NULL; |
---|
| 970 | d->pAttribute= NULL; |
---|
| 971 | d->pOrder= NULL; |
---|
| 972 | |
---|
| 973 | updateName_WOSD(lpszName); |
---|
| 974 | } |
---|
| 975 | |
---|
| 976 | XMLNode XMLNode::createXMLTopNode_WOSD(XMLSTR lpszName, char isDeclaration) { return XMLNode(NULL,lpszName,isDeclaration); } |
---|
| 977 | XMLNode XMLNode::createXMLTopNode(XMLCSTR lpszName, char isDeclaration) { return XMLNode(NULL,stringDup(lpszName),isDeclaration); } |
---|
| 978 | |
---|
| 979 | #define MEMORYINCREASE 50 |
---|
| 980 | |
---|
| 981 | static inline void myFree(void *p) { if (p) free(p); }; |
---|
| 982 | static inline void *myRealloc(void *p, int newsize, int memInc, int sizeofElem) |
---|
| 983 | { |
---|
| 984 | if (p==NULL) { if (memInc) return malloc(memInc*sizeofElem); return malloc(sizeofElem); } |
---|
| 985 | if ((memInc==0)||((newsize%memInc)==0)) p=realloc(p,(newsize+memInc)*sizeofElem); |
---|
| 986 | // if (!p) |
---|
| 987 | // { |
---|
| 988 | // printf("XMLParser Error: Not enough memory! Aborting...\n"); exit(220); |
---|
| 989 | // } |
---|
| 990 | return p; |
---|
| 991 | } |
---|
| 992 | |
---|
| 993 | // private: |
---|
| 994 | XMLElementPosition XMLNode::findPosition(XMLNodeData *d, int index, XMLElementType xxtype) |
---|
| 995 | { |
---|
| 996 | if (index<0) return -1; |
---|
| 997 | int i=0,j=(int)((index<<2)+xxtype),*o=d->pOrder; while (o[i]!=j) i++; return i; |
---|
| 998 | } |
---|
| 999 | |
---|
| 1000 | // private: |
---|
| 1001 | // update "order" information when deleting a content of a XMLNode |
---|
| 1002 | int XMLNode::removeOrderElement(XMLNodeData *d, XMLElementType t, int index) |
---|
| 1003 | { |
---|
| 1004 | int n=d->nChild+d->nText+d->nClear, *o=d->pOrder,i=findPosition(d,index,t); |
---|
| 1005 | memmove(o+i, o+i+1, (n-i)*sizeof(int)); |
---|
| 1006 | for (;i<n;i++) |
---|
| 1007 | if ((o[i]&3)==(int)t) o[i]-=4; |
---|
| 1008 | // We should normally do: |
---|
| 1009 | // d->pOrder=(int)realloc(d->pOrder,n*sizeof(int)); |
---|
| 1010 | // but we skip reallocation because it's too time consuming. |
---|
| 1011 | // Anyway, at the end, it will be free'd completely at once. |
---|
| 1012 | return i; |
---|
| 1013 | } |
---|
| 1014 | |
---|
| 1015 | void *XMLNode::addToOrder(int memoryIncrease,int *_pos, int nc, void *p, int size, XMLElementType xtype) |
---|
| 1016 | { |
---|
| 1017 | // in: *_pos is the position inside d->pOrder ("-1" means "EndOf") |
---|
| 1018 | // out: *_pos is the index inside p |
---|
| 1019 | p=myRealloc(p,(nc+1),memoryIncrease,size); |
---|
| 1020 | int n=d->nChild+d->nText+d->nClear; |
---|
| 1021 | d->pOrder=(int*)myRealloc(d->pOrder,n+1,memoryIncrease*3,sizeof(int)); |
---|
| 1022 | int pos=*_pos,*o=d->pOrder; |
---|
| 1023 | |
---|
| 1024 | if ((pos<0)||(pos>=n)) { *_pos=nc; o[n]=(int)((nc<<2)+xtype); return p; } |
---|
| 1025 | |
---|
| 1026 | int i=pos; |
---|
| 1027 | memmove(o+i+1, o+i, (n-i)*sizeof(int)); |
---|
| 1028 | |
---|
| 1029 | while ((pos<n)&&((o[pos]&3)!=(int)xtype)) pos++; |
---|
| 1030 | if (pos==n) { *_pos=nc; o[n]=(int)((nc<<2)+xtype); return p; } |
---|
| 1031 | |
---|
| 1032 | o[i]=o[pos]; |
---|
| 1033 | for (i=pos+1;i<=n;i++) if ((o[i]&3)==(int)xtype) o[i]+=4; |
---|
| 1034 | |
---|
| 1035 | *_pos=pos=o[pos]>>2; |
---|
| 1036 | memmove(((char*)p)+(pos+1)*size,((char*)p)+pos*size,(nc-pos)*size); |
---|
| 1037 | |
---|
| 1038 | return p; |
---|
| 1039 | } |
---|
| 1040 | |
---|
| 1041 | // Add a child node to the given element. |
---|
| 1042 | XMLNode XMLNode::addChild_priv(int memoryIncrease, XMLSTR lpszName, char isDeclaration, int pos) |
---|
| 1043 | { |
---|
| 1044 | if (!lpszName) return emptyXMLNode; |
---|
| 1045 | d->pChild=(XMLNode*)addToOrder(memoryIncrease,&pos,d->nChild,d->pChild,sizeof(XMLNode),eNodeChild); |
---|
| 1046 | d->pChild[pos].d=NULL; |
---|
| 1047 | d->pChild[pos]=XMLNode(d,lpszName,isDeclaration); |
---|
| 1048 | d->nChild++; |
---|
| 1049 | return d->pChild[pos]; |
---|
| 1050 | } |
---|
| 1051 | |
---|
| 1052 | // Add an attribute to an element. |
---|
| 1053 | XMLAttribute *XMLNode::addAttribute_priv(int memoryIncrease,XMLSTR lpszName, XMLSTR lpszValuev) |
---|
| 1054 | { |
---|
| 1055 | if (!lpszName) return &emptyXMLAttribute; |
---|
| 1056 | if (!d) { myFree(lpszName); myFree(lpszValuev); return &emptyXMLAttribute; } |
---|
| 1057 | int nc=d->nAttribute; |
---|
| 1058 | d->pAttribute=(XMLAttribute*)myRealloc(d->pAttribute,(nc+1),memoryIncrease,sizeof(XMLAttribute)); |
---|
| 1059 | XMLAttribute *pAttr=d->pAttribute+nc; |
---|
| 1060 | pAttr->lpszName = lpszName; |
---|
| 1061 | pAttr->lpszValue = lpszValuev; |
---|
| 1062 | d->nAttribute++; |
---|
| 1063 | return pAttr; |
---|
| 1064 | } |
---|
| 1065 | |
---|
| 1066 | // Add text to the element. |
---|
| 1067 | XMLCSTR XMLNode::addText_priv(int memoryIncrease, XMLSTR lpszValue, int pos) |
---|
| 1068 | { |
---|
| 1069 | if (!lpszValue) return NULL; |
---|
| 1070 | if (!d) { myFree(lpszValue); return NULL; } |
---|
| 1071 | d->pText=(XMLCSTR*)addToOrder(memoryIncrease,&pos,d->nText,d->pText,sizeof(XMLSTR),eNodeText); |
---|
| 1072 | d->pText[pos]=lpszValue; |
---|
| 1073 | d->nText++; |
---|
| 1074 | return lpszValue; |
---|
| 1075 | } |
---|
| 1076 | |
---|
| 1077 | // Add clear (unformatted) text to the element. |
---|
| 1078 | XMLClear *XMLNode::addClear_priv(int memoryIncrease, XMLSTR lpszValue, XMLCSTR lpszOpen, XMLCSTR lpszClose, int pos) |
---|
| 1079 | { |
---|
| 1080 | if (!lpszValue) return &emptyXMLClear; |
---|
| 1081 | if (!d) { myFree(lpszValue); return &emptyXMLClear; } |
---|
| 1082 | d->pClear=(XMLClear *)addToOrder(memoryIncrease,&pos,d->nClear,d->pClear,sizeof(XMLClear),eNodeClear); |
---|
| 1083 | XMLClear *pNewClear=d->pClear+pos; |
---|
| 1084 | pNewClear->lpszValue = lpszValue; |
---|
| 1085 | if (!lpszOpen) lpszOpen=XMLClearTags->lpszOpen; |
---|
| 1086 | if (!lpszClose) lpszClose=XMLClearTags->lpszClose; |
---|
| 1087 | pNewClear->lpszOpenTag = lpszOpen; |
---|
| 1088 | pNewClear->lpszCloseTag = lpszClose; |
---|
| 1089 | d->nClear++; |
---|
| 1090 | return pNewClear; |
---|
| 1091 | } |
---|
| 1092 | |
---|
| 1093 | // private: |
---|
| 1094 | // Parse a clear (unformatted) type node. |
---|
| 1095 | char XMLNode::parseClearTag(void *px, void *_pClear) |
---|
| 1096 | { |
---|
| 1097 | XML *pXML=(XML *)px; |
---|
| 1098 | ALLXMLClearTag pClear=*((ALLXMLClearTag*)_pClear); |
---|
| 1099 | int cbTemp=0; |
---|
| 1100 | XMLCSTR lpszTemp=NULL; |
---|
| 1101 | XMLCSTR lpXML=&pXML->lpXML[pXML->nIndex]; |
---|
| 1102 | static XMLCSTR docTypeEnd=_X("]>"); |
---|
| 1103 | |
---|
| 1104 | // Find the closing tag |
---|
| 1105 | // Seems the <!DOCTYPE need a better treatment so lets handle it |
---|
| 1106 | if (pClear.lpszOpen==XMLClearTags[1].lpszOpen) |
---|
| 1107 | { |
---|
| 1108 | XMLCSTR pCh=lpXML; |
---|
| 1109 | while (*pCh) |
---|
| 1110 | { |
---|
| 1111 | if (*pCh==_X('<')) { pClear.lpszClose=docTypeEnd; lpszTemp=xstrstr(lpXML,docTypeEnd); break; } |
---|
| 1112 | else if (*pCh==_X('>')) { lpszTemp=pCh; break; } |
---|
| 1113 | #ifdef _XMLWIDECHAR |
---|
| 1114 | pCh++; |
---|
| 1115 | #else |
---|
| 1116 | pCh+=XML_ByteTable[(unsigned char)(*pCh)]; |
---|
| 1117 | #endif |
---|
| 1118 | } |
---|
| 1119 | } else lpszTemp=xstrstr(lpXML, pClear.lpszClose); |
---|
| 1120 | |
---|
| 1121 | if (lpszTemp) |
---|
| 1122 | { |
---|
| 1123 | // Cache the size and increment the index |
---|
| 1124 | cbTemp = (int)(lpszTemp - lpXML); |
---|
| 1125 | |
---|
| 1126 | pXML->nIndex += cbTemp+(int)xstrlen(pClear.lpszClose); |
---|
| 1127 | |
---|
| 1128 | // Add the clear node to the current element |
---|
| 1129 | addClear_priv(MEMORYINCREASE,stringDup(lpXML,cbTemp), pClear.lpszOpen, pClear.lpszClose,-1); |
---|
| 1130 | return 0; |
---|
| 1131 | } |
---|
| 1132 | |
---|
| 1133 | // If we failed to find the end tag |
---|
| 1134 | pXML->error = eXMLErrorUnmatchedEndClearTag; |
---|
| 1135 | return 1; |
---|
| 1136 | } |
---|
| 1137 | |
---|
| 1138 | void XMLNode::exactMemory(XMLNodeData *d) |
---|
| 1139 | { |
---|
| 1140 | if (d->pOrder) d->pOrder=(int*)realloc(d->pOrder,(d->nChild+d->nText+d->nClear)*sizeof(int)); |
---|
| 1141 | if (d->pChild) d->pChild=(XMLNode*)realloc(d->pChild,d->nChild*sizeof(XMLNode)); |
---|
| 1142 | if (d->pAttribute) d->pAttribute=(XMLAttribute*)realloc(d->pAttribute,d->nAttribute*sizeof(XMLAttribute)); |
---|
| 1143 | if (d->pText) d->pText=(XMLCSTR*)realloc(d->pText,d->nText*sizeof(XMLSTR)); |
---|
| 1144 | if (d->pClear) d->pClear=(XMLClear *)realloc(d->pClear,d->nClear*sizeof(XMLClear)); |
---|
| 1145 | } |
---|
| 1146 | |
---|
| 1147 | char XMLNode::maybeAddTxT(void *pa, XMLCSTR tokenPStr) |
---|
| 1148 | { |
---|
| 1149 | XML *pXML=(XML *)pa; |
---|
| 1150 | XMLCSTR lpszText=pXML->lpszText; |
---|
| 1151 | if (!lpszText) return 0; |
---|
| 1152 | if (dropWhiteSpace) while (XML_isSPACECHAR(*lpszText)&&(lpszText!=tokenPStr)) lpszText++; |
---|
| 1153 | int cbText = (int)(tokenPStr - lpszText); |
---|
| 1154 | if (!cbText) { pXML->lpszText=NULL; return 0; } |
---|
| 1155 | if (dropWhiteSpace) { cbText--; while ((cbText)&&XML_isSPACECHAR(lpszText[cbText])) cbText--; cbText++; } |
---|
| 1156 | if (!cbText) { pXML->lpszText=NULL; return 0; } |
---|
| 1157 | XMLSTR lpt=fromXMLString(lpszText,cbText,pXML); |
---|
| 1158 | if (!lpt) return 1; |
---|
| 1159 | addText_priv(MEMORYINCREASE,lpt,-1); |
---|
| 1160 | pXML->lpszText=NULL; |
---|
| 1161 | return 0; |
---|
| 1162 | } |
---|
| 1163 | // private: |
---|
| 1164 | // Recursively parse an XML element. |
---|
| 1165 | int XMLNode::ParseXMLElement(void *pa) |
---|
| 1166 | { |
---|
| 1167 | XML *pXML=(XML *)pa; |
---|
| 1168 | int cbToken; |
---|
| 1169 | enum XMLTokenTypeTag xtype; |
---|
| 1170 | NextToken token; |
---|
| 1171 | XMLCSTR lpszTemp=NULL; |
---|
| 1172 | int cbTemp=0; |
---|
| 1173 | char nDeclaration; |
---|
| 1174 | XMLNode pNew; |
---|
| 1175 | enum Status status; // inside or outside a tag |
---|
| 1176 | enum Attrib attrib = eAttribName; |
---|
| 1177 | |
---|
| 1178 | assert(pXML); |
---|
| 1179 | |
---|
| 1180 | // If this is the first call to the function |
---|
| 1181 | if (pXML->nFirst) |
---|
| 1182 | { |
---|
| 1183 | // Assume we are outside of a tag definition |
---|
| 1184 | pXML->nFirst = FALSE; |
---|
| 1185 | status = eOutsideTag; |
---|
| 1186 | } else |
---|
| 1187 | { |
---|
| 1188 | // If this is not the first call then we should only be called when inside a tag. |
---|
| 1189 | status = eInsideTag; |
---|
| 1190 | } |
---|
| 1191 | |
---|
| 1192 | // Iterate through the tokens in the document |
---|
| 1193 | for(;;) |
---|
| 1194 | { |
---|
| 1195 | // Obtain the next token |
---|
| 1196 | token = GetNextToken(pXML, &cbToken, &xtype); |
---|
| 1197 | |
---|
| 1198 | if (xtype != eTokenError) |
---|
| 1199 | { |
---|
| 1200 | // Check the current status |
---|
| 1201 | switch(status) |
---|
| 1202 | { |
---|
| 1203 | |
---|
| 1204 | // If we are outside of a tag definition |
---|
| 1205 | case eOutsideTag: |
---|
| 1206 | |
---|
| 1207 | // Check what type of token we obtained |
---|
| 1208 | switch(xtype) |
---|
| 1209 | { |
---|
| 1210 | // If we have found text or quoted text |
---|
| 1211 | case eTokenText: |
---|
| 1212 | case eTokenCloseTag: /* '>' */ |
---|
| 1213 | case eTokenShortHandClose: /* '/>' */ |
---|
| 1214 | case eTokenQuotedText: |
---|
| 1215 | case eTokenEquals: |
---|
| 1216 | break; |
---|
| 1217 | |
---|
| 1218 | // If we found a start tag '<' and declarations '<?' |
---|
| 1219 | case eTokenTagStart: |
---|
| 1220 | case eTokenDeclaration: |
---|
| 1221 | |
---|
| 1222 | // Cache whether this new element is a declaration or not |
---|
| 1223 | nDeclaration = (xtype == eTokenDeclaration); |
---|
| 1224 | |
---|
| 1225 | // If we have node text then add this to the element |
---|
| 1226 | if (maybeAddTxT(pXML,token.pStr)) return FALSE; |
---|
| 1227 | |
---|
| 1228 | // Find the name of the tag |
---|
| 1229 | token = GetNextToken(pXML, &cbToken, &xtype); |
---|
| 1230 | |
---|
| 1231 | // Return an error if we couldn't obtain the next token or |
---|
| 1232 | // it wasnt text |
---|
| 1233 | if (xtype != eTokenText) |
---|
| 1234 | { |
---|
| 1235 | pXML->error = eXMLErrorMissingTagName; |
---|
| 1236 | return FALSE; |
---|
| 1237 | } |
---|
| 1238 | |
---|
| 1239 | // If we found a new element which is the same as this |
---|
| 1240 | // element then we need to pass this back to the caller.. |
---|
| 1241 | |
---|
| 1242 | #ifdef APPROXIMATE_PARSING |
---|
| 1243 | if (d->lpszName && |
---|
| 1244 | myTagCompare(d->lpszName, token.pStr) == 0) |
---|
| 1245 | { |
---|
| 1246 | // Indicate to the caller that it needs to create a |
---|
| 1247 | // new element. |
---|
| 1248 | pXML->lpNewElement = token.pStr; |
---|
| 1249 | pXML->cbNewElement = cbToken; |
---|
| 1250 | return TRUE; |
---|
| 1251 | } else |
---|
| 1252 | #endif |
---|
| 1253 | { |
---|
| 1254 | // If the name of the new element differs from the name of |
---|
| 1255 | // the current element we need to add the new element to |
---|
| 1256 | // the current one and recurse |
---|
| 1257 | pNew = addChild_priv(MEMORYINCREASE,stringDup(token.pStr,cbToken), nDeclaration,-1); |
---|
| 1258 | |
---|
| 1259 | while (!pNew.isEmpty()) |
---|
| 1260 | { |
---|
| 1261 | // Callself to process the new node. If we return |
---|
| 1262 | // FALSE this means we dont have any more |
---|
| 1263 | // processing to do... |
---|
| 1264 | |
---|
| 1265 | if (!pNew.ParseXMLElement(pXML)) return FALSE; |
---|
| 1266 | else |
---|
| 1267 | { |
---|
| 1268 | // If the call to recurse this function |
---|
| 1269 | // evented in a end tag specified in XML then |
---|
| 1270 | // we need to unwind the calls to this |
---|
| 1271 | // function until we find the appropriate node |
---|
| 1272 | // (the element name and end tag name must |
---|
| 1273 | // match) |
---|
| 1274 | if (pXML->cbEndTag) |
---|
| 1275 | { |
---|
| 1276 | // If we are back at the root node then we |
---|
| 1277 | // have an unmatched end tag |
---|
| 1278 | if (!d->lpszName) |
---|
| 1279 | { |
---|
| 1280 | pXML->error=eXMLErrorUnmatchedEndTag; |
---|
| 1281 | return FALSE; |
---|
| 1282 | } |
---|
| 1283 | |
---|
| 1284 | // If the end tag matches the name of this |
---|
| 1285 | // element then we only need to unwind |
---|
| 1286 | // once more... |
---|
| 1287 | |
---|
| 1288 | if (myTagCompare(d->lpszName, pXML->lpEndTag)==0) |
---|
| 1289 | { |
---|
| 1290 | pXML->cbEndTag = 0; |
---|
| 1291 | } |
---|
| 1292 | |
---|
| 1293 | return TRUE; |
---|
| 1294 | } else |
---|
| 1295 | if (pXML->cbNewElement) |
---|
| 1296 | { |
---|
| 1297 | // If the call indicated a new element is to |
---|
| 1298 | // be created on THIS element. |
---|
| 1299 | |
---|
| 1300 | // If the name of this element matches the |
---|
| 1301 | // name of the element we need to create |
---|
| 1302 | // then we need to return to the caller |
---|
| 1303 | // and let it process the element. |
---|
| 1304 | |
---|
| 1305 | if (myTagCompare(d->lpszName, pXML->lpNewElement)==0) |
---|
| 1306 | { |
---|
| 1307 | return TRUE; |
---|
| 1308 | } |
---|
| 1309 | |
---|
| 1310 | // Add the new element and recurse |
---|
| 1311 | pNew = addChild_priv(MEMORYINCREASE,stringDup(pXML->lpNewElement,pXML->cbNewElement),0,-1); |
---|
| 1312 | pXML->cbNewElement = 0; |
---|
| 1313 | } |
---|
| 1314 | else |
---|
| 1315 | { |
---|
| 1316 | // If we didn't have a new element to create |
---|
| 1317 | pNew = emptyXMLNode; |
---|
| 1318 | |
---|
| 1319 | } |
---|
| 1320 | } |
---|
| 1321 | } |
---|
| 1322 | } |
---|
| 1323 | break; |
---|
| 1324 | |
---|
| 1325 | // If we found an end tag |
---|
| 1326 | case eTokenTagEnd: |
---|
| 1327 | |
---|
| 1328 | // If we have node text then add this to the element |
---|
| 1329 | if (maybeAddTxT(pXML,token.pStr)) return FALSE; |
---|
| 1330 | |
---|
| 1331 | // Find the name of the end tag |
---|
| 1332 | token = GetNextToken(pXML, &cbTemp, &xtype); |
---|
| 1333 | |
---|
| 1334 | // The end tag should be text |
---|
| 1335 | if (xtype != eTokenText) |
---|
| 1336 | { |
---|
| 1337 | pXML->error = eXMLErrorMissingEndTagName; |
---|
| 1338 | return FALSE; |
---|
| 1339 | } |
---|
| 1340 | lpszTemp = token.pStr; |
---|
| 1341 | |
---|
| 1342 | // After the end tag we should find a closing tag |
---|
| 1343 | token = GetNextToken(pXML, &cbToken, &xtype); |
---|
| 1344 | if (xtype != eTokenCloseTag) |
---|
| 1345 | { |
---|
| 1346 | pXML->error = eXMLErrorMissingEndTagName; |
---|
| 1347 | return FALSE; |
---|
| 1348 | } |
---|
| 1349 | pXML->lpszText=pXML->lpXML+pXML->nIndex; |
---|
| 1350 | |
---|
| 1351 | // We need to return to the previous caller. If the name |
---|
| 1352 | // of the tag cannot be found we need to keep returning to |
---|
| 1353 | // caller until we find a match |
---|
| 1354 | if (myTagCompare(d->lpszName, lpszTemp) != 0) |
---|
| 1355 | #ifdef STRICT_PARSING |
---|
| 1356 | { |
---|
| 1357 | pXML->error=eXMLErrorUnmatchedEndTag; |
---|
| 1358 | pXML->nIndexMissigEndTag=pXML->nIndex; |
---|
| 1359 | return FALSE; |
---|
| 1360 | } |
---|
| 1361 | #else |
---|
| 1362 | { |
---|
| 1363 | pXML->error=eXMLErrorMissingEndTag; |
---|
| 1364 | pXML->nIndexMissigEndTag=pXML->nIndex; |
---|
| 1365 | pXML->lpEndTag = lpszTemp; |
---|
| 1366 | pXML->cbEndTag = cbTemp; |
---|
| 1367 | } |
---|
| 1368 | #endif |
---|
| 1369 | |
---|
| 1370 | // Return to the caller |
---|
| 1371 | exactMemory(d); |
---|
| 1372 | return TRUE; |
---|
| 1373 | |
---|
| 1374 | // If we found a clear (unformatted) token |
---|
| 1375 | case eTokenClear: |
---|
| 1376 | // If we have node text then add this to the element |
---|
| 1377 | if (maybeAddTxT(pXML,token.pStr)) return FALSE; |
---|
| 1378 | if (parseClearTag(pXML, token.pClr)) return FALSE; |
---|
| 1379 | pXML->lpszText=pXML->lpXML+pXML->nIndex; |
---|
| 1380 | break; |
---|
| 1381 | |
---|
| 1382 | default: |
---|
| 1383 | break; |
---|
| 1384 | } |
---|
| 1385 | break; |
---|
| 1386 | |
---|
| 1387 | // If we are inside a tag definition we need to search for attributes |
---|
| 1388 | case eInsideTag: |
---|
| 1389 | |
---|
| 1390 | // Check what part of the attribute (name, equals, value) we |
---|
| 1391 | // are looking for. |
---|
| 1392 | switch(attrib) |
---|
| 1393 | { |
---|
| 1394 | // If we are looking for a new attribute |
---|
| 1395 | case eAttribName: |
---|
| 1396 | |
---|
| 1397 | // Check what the current token type is |
---|
| 1398 | switch(xtype) |
---|
| 1399 | { |
---|
| 1400 | // If the current type is text... |
---|
| 1401 | // Eg. 'attribute' |
---|
| 1402 | case eTokenText: |
---|
| 1403 | // Cache the token then indicate that we are next to |
---|
| 1404 | // look for the equals |
---|
| 1405 | lpszTemp = token.pStr; |
---|
| 1406 | cbTemp = cbToken; |
---|
| 1407 | attrib = eAttribEquals; |
---|
| 1408 | break; |
---|
| 1409 | |
---|
| 1410 | // If we found a closing tag... |
---|
| 1411 | // Eg. '>' |
---|
| 1412 | case eTokenCloseTag: |
---|
| 1413 | // We are now outside the tag |
---|
| 1414 | status = eOutsideTag; |
---|
| 1415 | pXML->lpszText=pXML->lpXML+pXML->nIndex; |
---|
| 1416 | break; |
---|
| 1417 | |
---|
| 1418 | // If we found a short hand '/>' closing tag then we can |
---|
| 1419 | // return to the caller |
---|
| 1420 | case eTokenShortHandClose: |
---|
| 1421 | exactMemory(d); |
---|
| 1422 | pXML->lpszText=pXML->lpXML+pXML->nIndex; |
---|
| 1423 | return TRUE; |
---|
| 1424 | |
---|
| 1425 | // Errors... |
---|
| 1426 | case eTokenQuotedText: /* '"SomeText"' */ |
---|
| 1427 | case eTokenTagStart: /* '<' */ |
---|
| 1428 | case eTokenTagEnd: /* '</' */ |
---|
| 1429 | case eTokenEquals: /* '=' */ |
---|
| 1430 | case eTokenDeclaration: /* '<?' */ |
---|
| 1431 | case eTokenClear: |
---|
| 1432 | pXML->error = eXMLErrorUnexpectedToken; |
---|
| 1433 | return FALSE; |
---|
| 1434 | default: break; |
---|
| 1435 | } |
---|
| 1436 | break; |
---|
| 1437 | |
---|
| 1438 | // If we are looking for an equals |
---|
| 1439 | case eAttribEquals: |
---|
| 1440 | // Check what the current token type is |
---|
| 1441 | switch(xtype) |
---|
| 1442 | { |
---|
| 1443 | // If the current type is text... |
---|
| 1444 | // Eg. 'Attribute AnotherAttribute' |
---|
| 1445 | case eTokenText: |
---|
| 1446 | // Add the unvalued attribute to the list |
---|
| 1447 | addAttribute_priv(MEMORYINCREASE,stringDup(lpszTemp,cbTemp), NULL); |
---|
| 1448 | // Cache the token then indicate. We are next to |
---|
| 1449 | // look for the equals attribute |
---|
| 1450 | lpszTemp = token.pStr; |
---|
| 1451 | cbTemp = cbToken; |
---|
| 1452 | break; |
---|
| 1453 | |
---|
| 1454 | // If we found a closing tag 'Attribute >' or a short hand |
---|
| 1455 | // closing tag 'Attribute />' |
---|
| 1456 | case eTokenShortHandClose: |
---|
| 1457 | case eTokenCloseTag: |
---|
| 1458 | // If we are a declaration element '<?' then we need |
---|
| 1459 | // to remove extra closing '?' if it exists |
---|
| 1460 | pXML->lpszText=pXML->lpXML+pXML->nIndex; |
---|
| 1461 | |
---|
| 1462 | if (d->isDeclaration && |
---|
| 1463 | (lpszTemp[cbTemp-1]) == _X('?')) |
---|
| 1464 | { |
---|
| 1465 | cbTemp--; |
---|
| 1466 | } |
---|
| 1467 | |
---|
| 1468 | if (cbTemp) |
---|
| 1469 | { |
---|
| 1470 | // Add the unvalued attribute to the list |
---|
| 1471 | addAttribute_priv(MEMORYINCREASE,stringDup(lpszTemp,cbTemp), NULL); |
---|
| 1472 | } |
---|
| 1473 | |
---|
| 1474 | // If this is the end of the tag then return to the caller |
---|
| 1475 | if (xtype == eTokenShortHandClose) |
---|
| 1476 | { |
---|
| 1477 | exactMemory(d); |
---|
| 1478 | return TRUE; |
---|
| 1479 | } |
---|
| 1480 | |
---|
| 1481 | // We are now outside the tag |
---|
| 1482 | status = eOutsideTag; |
---|
| 1483 | break; |
---|
| 1484 | |
---|
| 1485 | // If we found the equals token... |
---|
| 1486 | // Eg. 'Attribute =' |
---|
| 1487 | case eTokenEquals: |
---|
| 1488 | // Indicate that we next need to search for the value |
---|
| 1489 | // for the attribute |
---|
| 1490 | attrib = eAttribValue; |
---|
| 1491 | break; |
---|
| 1492 | |
---|
| 1493 | // Errors... |
---|
| 1494 | case eTokenQuotedText: /* 'Attribute "InvalidAttr"'*/ |
---|
| 1495 | case eTokenTagStart: /* 'Attribute <' */ |
---|
| 1496 | case eTokenTagEnd: /* 'Attribute </' */ |
---|
| 1497 | case eTokenDeclaration: /* 'Attribute <?' */ |
---|
| 1498 | case eTokenClear: |
---|
| 1499 | pXML->error = eXMLErrorUnexpectedToken; |
---|
| 1500 | return FALSE; |
---|
| 1501 | default: break; |
---|
| 1502 | } |
---|
| 1503 | break; |
---|
| 1504 | |
---|
| 1505 | // If we are looking for an attribute value |
---|
| 1506 | case eAttribValue: |
---|
| 1507 | // Check what the current token type is |
---|
| 1508 | switch(xtype) |
---|
| 1509 | { |
---|
| 1510 | // If the current type is text or quoted text... |
---|
| 1511 | // Eg. 'Attribute = "Value"' or 'Attribute = Value' or |
---|
| 1512 | // 'Attribute = 'Value''. |
---|
| 1513 | case eTokenText: |
---|
| 1514 | case eTokenQuotedText: |
---|
| 1515 | // If we are a declaration element '<?' then we need |
---|
| 1516 | // to remove extra closing '?' if it exists |
---|
| 1517 | if (d->isDeclaration && |
---|
| 1518 | (token.pStr[cbToken-1]) == _X('?')) |
---|
| 1519 | { |
---|
| 1520 | cbToken--; |
---|
| 1521 | } |
---|
| 1522 | |
---|
| 1523 | if (cbTemp) |
---|
| 1524 | { |
---|
| 1525 | // Add the valued attribute to the list |
---|
| 1526 | if (xtype==eTokenQuotedText) { token.pStr++; cbToken-=2; } |
---|
| 1527 | XMLSTR attrVal=(XMLSTR)token.pStr; |
---|
| 1528 | if (attrVal) |
---|
| 1529 | { |
---|
| 1530 | attrVal=fromXMLString(attrVal,cbToken,pXML); |
---|
| 1531 | if (!attrVal) return FALSE; |
---|
| 1532 | } |
---|
| 1533 | addAttribute_priv(MEMORYINCREASE,stringDup(lpszTemp,cbTemp),attrVal); |
---|
| 1534 | } |
---|
| 1535 | |
---|
| 1536 | // Indicate we are searching for a new attribute |
---|
| 1537 | attrib = eAttribName; |
---|
| 1538 | break; |
---|
| 1539 | |
---|
| 1540 | // Errors... |
---|
| 1541 | case eTokenTagStart: /* 'Attr = <' */ |
---|
| 1542 | case eTokenTagEnd: /* 'Attr = </' */ |
---|
| 1543 | case eTokenCloseTag: /* 'Attr = >' */ |
---|
| 1544 | case eTokenShortHandClose: /* "Attr = />" */ |
---|
| 1545 | case eTokenEquals: /* 'Attr = =' */ |
---|
| 1546 | case eTokenDeclaration: /* 'Attr = <?' */ |
---|
| 1547 | case eTokenClear: |
---|
| 1548 | pXML->error = eXMLErrorUnexpectedToken; |
---|
| 1549 | return FALSE; |
---|
| 1550 | break; |
---|
| 1551 | default: break; |
---|
| 1552 | } |
---|
| 1553 | } |
---|
| 1554 | } |
---|
| 1555 | } |
---|
| 1556 | // If we failed to obtain the next token |
---|
| 1557 | else |
---|
| 1558 | { |
---|
| 1559 | if ((!d->isDeclaration)&&(d->pParent)) |
---|
| 1560 | { |
---|
| 1561 | #ifdef STRICT_PARSING |
---|
| 1562 | pXML->error=eXMLErrorUnmatchedEndTag; |
---|
| 1563 | #else |
---|
| 1564 | pXML->error=eXMLErrorMissingEndTag; |
---|
| 1565 | #endif |
---|
| 1566 | pXML->nIndexMissigEndTag=pXML->nIndex; |
---|
| 1567 | } |
---|
| 1568 | maybeAddTxT(pXML,pXML->lpXML+pXML->nIndex); |
---|
| 1569 | return FALSE; |
---|
| 1570 | } |
---|
| 1571 | } |
---|
| 1572 | } |
---|
| 1573 | |
---|
| 1574 | // Count the number of lines and columns in an XML string. |
---|
| 1575 | static void CountLinesAndColumns(XMLCSTR lpXML, int nUpto, XMLResults *pResults) |
---|
| 1576 | { |
---|
| 1577 | XMLCHAR ch; |
---|
| 1578 | assert(lpXML); |
---|
| 1579 | assert(pResults); |
---|
| 1580 | |
---|
| 1581 | struct XML xml={ lpXML,lpXML, 0, 0, eXMLErrorNone, NULL, 0, NULL, 0, TRUE }; |
---|
| 1582 | |
---|
| 1583 | pResults->nLine = 1; |
---|
| 1584 | pResults->nColumn = 1; |
---|
| 1585 | while (xml.nIndex<nUpto) |
---|
| 1586 | { |
---|
| 1587 | ch = getNextChar(&xml); |
---|
| 1588 | if (ch != _X('\n')) pResults->nColumn++; |
---|
| 1589 | else |
---|
| 1590 | { |
---|
| 1591 | pResults->nLine++; |
---|
| 1592 | pResults->nColumn=1; |
---|
| 1593 | } |
---|
| 1594 | } |
---|
| 1595 | } |
---|
| 1596 | |
---|
| 1597 | // Parse XML and return the root element. |
---|
| 1598 | XMLNode XMLNode::parseString(XMLCSTR lpszXML, XMLCSTR tag, XMLResults *pResults) |
---|
| 1599 | { |
---|
| 1600 | if (!lpszXML) |
---|
| 1601 | { |
---|
| 1602 | if (pResults) |
---|
| 1603 | { |
---|
| 1604 | pResults->error=eXMLErrorNoElements; |
---|
| 1605 | pResults->nLine=0; |
---|
| 1606 | pResults->nColumn=0; |
---|
| 1607 | } |
---|
| 1608 | return emptyXMLNode; |
---|
| 1609 | } |
---|
| 1610 | |
---|
| 1611 | XMLNode xnode(NULL,NULL,FALSE); |
---|
| 1612 | struct XML xml={ lpszXML, lpszXML, 0, 0, eXMLErrorNone, NULL, 0, NULL, 0, TRUE }; |
---|
| 1613 | |
---|
| 1614 | // Create header element |
---|
| 1615 | xnode.ParseXMLElement(&xml); |
---|
| 1616 | enum XMLError error = xml.error; |
---|
| 1617 | if (!xnode.nChildNode()) error=eXMLErrorNoXMLTagFound; |
---|
| 1618 | if ((xnode.nChildNode()==1)&&(xnode.nElement()==1)) xnode=xnode.getChildNode(); // skip the empty node |
---|
| 1619 | |
---|
| 1620 | // If no error occurred |
---|
| 1621 | if ((error==eXMLErrorNone)||(error==eXMLErrorMissingEndTag)||(error==eXMLErrorNoXMLTagFound)) |
---|
| 1622 | { |
---|
| 1623 | XMLCSTR name=xnode.getName(); |
---|
| 1624 | if (tag&&xstrlen(tag)&&((!name)||(xstricmp(xnode.getName(),tag)))) |
---|
| 1625 | { |
---|
| 1626 | XMLNode nodeTmp; |
---|
| 1627 | int i=0; |
---|
| 1628 | while (i<xnode.nChildNode()) |
---|
| 1629 | { |
---|
| 1630 | nodeTmp=xnode.getChildNode(i); |
---|
| 1631 | if (xstricmp(nodeTmp.getName(),tag)==0) break; |
---|
| 1632 | if (nodeTmp.isDeclaration()) { xnode=nodeTmp; i=0; } else i++; |
---|
| 1633 | } |
---|
| 1634 | if (i>=xnode.nChildNode()) |
---|
| 1635 | { |
---|
| 1636 | if (pResults) |
---|
| 1637 | { |
---|
| 1638 | pResults->error=eXMLErrorFirstTagNotFound; |
---|
| 1639 | pResults->nLine=0; |
---|
| 1640 | pResults->nColumn=0; |
---|
| 1641 | } |
---|
| 1642 | return emptyXMLNode; |
---|
| 1643 | } |
---|
| 1644 | xnode=nodeTmp; |
---|
| 1645 | } |
---|
| 1646 | } else |
---|
| 1647 | { |
---|
| 1648 | // Cleanup: this will destroy all the nodes |
---|
| 1649 | xnode = emptyXMLNode; |
---|
| 1650 | } |
---|
| 1651 | |
---|
| 1652 | |
---|
| 1653 | // If we have been given somewhere to place results |
---|
| 1654 | if (pResults) |
---|
| 1655 | { |
---|
| 1656 | pResults->error = error; |
---|
| 1657 | |
---|
| 1658 | // If we have an error |
---|
| 1659 | if (error!=eXMLErrorNone) |
---|
| 1660 | { |
---|
| 1661 | if (error==eXMLErrorMissingEndTag) xml.nIndex=xml.nIndexMissigEndTag; |
---|
| 1662 | // Find which line and column it starts on. |
---|
| 1663 | CountLinesAndColumns(xml.lpXML, xml.nIndex, pResults); |
---|
| 1664 | } |
---|
| 1665 | } |
---|
| 1666 | return xnode; |
---|
| 1667 | } |
---|
| 1668 | |
---|
| 1669 | XMLNode XMLNode::parseFile(XMLCSTR filename, XMLCSTR tag, XMLResults *pResults) |
---|
| 1670 | { |
---|
| 1671 | if (pResults) { pResults->nLine=0; pResults->nColumn=0; } |
---|
| 1672 | FILE *f=xfopen(filename,_X("rb")); |
---|
| 1673 | if (f==NULL) { if (pResults) pResults->error=eXMLErrorFileNotFound; return emptyXMLNode; } |
---|
| 1674 | fseek(f,0,SEEK_END); |
---|
| 1675 | int l=ftell(f),headerSz=0; |
---|
| 1676 | if (!l) { if (pResults) pResults->error=eXMLErrorEmpty; fclose(f); return emptyXMLNode; } |
---|
| 1677 | fseek(f,0,SEEK_SET); |
---|
| 1678 | unsigned char *buf=(unsigned char*)malloc(l+4); |
---|
| 1679 | fread(buf,l,1,f); |
---|
| 1680 | fclose(f); |
---|
| 1681 | buf[l]=0;buf[l+1]=0;buf[l+2]=0;buf[l+3]=0; |
---|
| 1682 | #ifdef _XMLWIDECHAR |
---|
| 1683 | if (guessWideCharChars) |
---|
| 1684 | { |
---|
| 1685 | if (!myIsTextWideChar(buf,l)) |
---|
| 1686 | { |
---|
| 1687 | if ((buf[0]==0xef)&&(buf[1]==0xbb)&&(buf[2]==0xbf)) headerSz=3; |
---|
| 1688 | XMLSTR b2=myMultiByteToWideChar((const char*)(buf+headerSz)); |
---|
| 1689 | free(buf); buf=(unsigned char*)b2; headerSz=0; |
---|
| 1690 | } else |
---|
| 1691 | { |
---|
| 1692 | if ((buf[0]==0xef)&&(buf[1]==0xff)) headerSz=2; |
---|
| 1693 | if ((buf[0]==0xff)&&(buf[1]==0xfe)) headerSz=2; |
---|
| 1694 | } |
---|
| 1695 | } |
---|
| 1696 | #else |
---|
| 1697 | if (guessWideCharChars) |
---|
| 1698 | { |
---|
| 1699 | if (myIsTextWideChar(buf,l)) |
---|
| 1700 | { |
---|
| 1701 | l/=sizeof(wchar_t); |
---|
| 1702 | if ((buf[0]==0xef)&&(buf[1]==0xff)) headerSz=2; |
---|
| 1703 | if ((buf[0]==0xff)&&(buf[1]==0xfe)) headerSz=2; |
---|
| 1704 | char *b2=myWideCharToMultiByte((const wchar_t*)(buf+headerSz)); |
---|
| 1705 | free(buf); buf=(unsigned char*)b2; headerSz=0; |
---|
| 1706 | } else |
---|
| 1707 | { |
---|
| 1708 | if ((buf[0]==0xef)&&(buf[1]==0xbb)&&(buf[2]==0xbf)) headerSz=3; |
---|
| 1709 | } |
---|
| 1710 | } |
---|
| 1711 | #endif |
---|
| 1712 | |
---|
| 1713 | if (!buf) { if (pResults) pResults->error=eXMLErrorCharConversionError; return emptyXMLNode; } |
---|
| 1714 | XMLNode x=parseString((XMLSTR)(buf+headerSz),tag,pResults); |
---|
| 1715 | free(buf); |
---|
| 1716 | return x; |
---|
| 1717 | } |
---|
| 1718 | |
---|
| 1719 | static inline void charmemset(XMLSTR dest,XMLCHAR c,int l) { while (l--) *(dest++)=c; } |
---|
| 1720 | // private: |
---|
| 1721 | // Creates an user friendly XML string from a given element with |
---|
| 1722 | // appropriate white space and carriage returns. |
---|
| 1723 | // |
---|
| 1724 | // This recurses through all subnodes then adds contents of the nodes to the |
---|
| 1725 | // string. |
---|
| 1726 | int XMLNode::CreateXMLStringR(XMLNodeData *pEntry, XMLSTR lpszMarker, int nFormat) |
---|
| 1727 | { |
---|
| 1728 | int nResult = 0; |
---|
| 1729 | int cb; |
---|
| 1730 | int cbElement; |
---|
| 1731 | int nChildFormat=-1; |
---|
| 1732 | int nElementI=pEntry->nChild+pEntry->nText+pEntry->nClear; |
---|
| 1733 | int i,j; |
---|
| 1734 | |
---|
| 1735 | assert(pEntry); |
---|
| 1736 | |
---|
| 1737 | #define LENSTR(lpsz) (lpsz ? xstrlen(lpsz) : 0) |
---|
| 1738 | |
---|
| 1739 | // If the element has no name then assume this is the head node. |
---|
| 1740 | cbElement = (int)LENSTR(pEntry->lpszName); |
---|
| 1741 | |
---|
| 1742 | if (cbElement) |
---|
| 1743 | { |
---|
| 1744 | // "<elementname " |
---|
| 1745 | cb = nFormat == -1 ? 0 : nFormat; |
---|
| 1746 | |
---|
| 1747 | if (lpszMarker) |
---|
| 1748 | { |
---|
| 1749 | if (cb) charmemset(lpszMarker, INDENTCHAR, sizeof(XMLCHAR)*cb); |
---|
| 1750 | nResult = cb; |
---|
| 1751 | lpszMarker[nResult++]=_X('<'); |
---|
| 1752 | if (pEntry->isDeclaration) lpszMarker[nResult++]=_X('?'); |
---|
| 1753 | xstrcpy(&lpszMarker[nResult], pEntry->lpszName); |
---|
| 1754 | nResult+=cbElement; |
---|
| 1755 | lpszMarker[nResult++]=_X(' '); |
---|
| 1756 | |
---|
| 1757 | } else |
---|
| 1758 | { |
---|
| 1759 | nResult+=cbElement+2+cb; |
---|
| 1760 | if (pEntry->isDeclaration) nResult++; |
---|
| 1761 | } |
---|
| 1762 | |
---|
| 1763 | // Enumerate attributes and add them to the string |
---|
| 1764 | XMLAttribute *pAttr=pEntry->pAttribute; |
---|
| 1765 | for (i=0; i<pEntry->nAttribute; i++) |
---|
| 1766 | { |
---|
| 1767 | // "Attrib |
---|
| 1768 | cb = (int)LENSTR(pAttr->lpszName); |
---|
| 1769 | if (cb) |
---|
| 1770 | { |
---|
| 1771 | if (lpszMarker) xstrcpy(&lpszMarker[nResult], pAttr->lpszName); |
---|
| 1772 | nResult += cb; |
---|
| 1773 | // "Attrib=Value " |
---|
| 1774 | if (pAttr->lpszValue) |
---|
| 1775 | { |
---|
| 1776 | cb=(int)lengthXMLString(pAttr->lpszValue); |
---|
| 1777 | if (lpszMarker) |
---|
| 1778 | { |
---|
| 1779 | lpszMarker[nResult]=_X('='); |
---|
| 1780 | lpszMarker[nResult+1]=_X('"'); |
---|
| 1781 | if (cb) toXMLStringUnSafe(&lpszMarker[nResult+2],pAttr->lpszValue); |
---|
| 1782 | lpszMarker[nResult+cb+2]=_X('"'); |
---|
| 1783 | } |
---|
| 1784 | nResult+=cb+3; |
---|
| 1785 | } |
---|
| 1786 | if (lpszMarker) lpszMarker[nResult] = _X(' '); |
---|
| 1787 | nResult++; |
---|
| 1788 | } |
---|
| 1789 | pAttr++; |
---|
| 1790 | } |
---|
| 1791 | |
---|
| 1792 | if (pEntry->isDeclaration) |
---|
| 1793 | { |
---|
| 1794 | if (lpszMarker) |
---|
| 1795 | { |
---|
| 1796 | lpszMarker[nResult-1]=_X('?'); |
---|
| 1797 | lpszMarker[nResult]=_X('>'); |
---|
| 1798 | } |
---|
| 1799 | nResult++; |
---|
| 1800 | if (nFormat!=-1) |
---|
| 1801 | { |
---|
| 1802 | if (lpszMarker) lpszMarker[nResult]=_X('\n'); |
---|
| 1803 | nResult++; |
---|
| 1804 | } |
---|
| 1805 | } else |
---|
| 1806 | // If there are child nodes we need to terminate the start tag |
---|
| 1807 | if (nElementI) |
---|
| 1808 | { |
---|
| 1809 | if (lpszMarker) lpszMarker[nResult-1]=_X('>'); |
---|
| 1810 | if (nFormat!=-1) |
---|
| 1811 | { |
---|
| 1812 | if (lpszMarker) lpszMarker[nResult]=_X('\n'); |
---|
| 1813 | nResult++; |
---|
| 1814 | } |
---|
| 1815 | } else nResult--; |
---|
| 1816 | } |
---|
| 1817 | |
---|
| 1818 | // Calculate the child format for when we recurse. This is used to |
---|
| 1819 | // determine the number of spaces used for prefixes. |
---|
| 1820 | if (nFormat!=-1) |
---|
| 1821 | { |
---|
| 1822 | if (cbElement&&(!pEntry->isDeclaration)) nChildFormat=nFormat+1; |
---|
| 1823 | else nChildFormat=nFormat; |
---|
| 1824 | } |
---|
| 1825 | |
---|
| 1826 | // Enumerate through remaining children |
---|
| 1827 | for (i=0; i<nElementI; i++) |
---|
| 1828 | { |
---|
| 1829 | j=pEntry->pOrder[i]; |
---|
| 1830 | switch((XMLElementType)(j&3)) |
---|
| 1831 | { |
---|
| 1832 | // Text nodes |
---|
| 1833 | case eNodeText: |
---|
| 1834 | { |
---|
| 1835 | // "Text" |
---|
| 1836 | XMLCSTR pChild=pEntry->pText[j>>2]; |
---|
| 1837 | cb = (int)lengthXMLString(pChild); |
---|
| 1838 | if (cb) |
---|
| 1839 | { |
---|
| 1840 | if (nFormat!=-1) |
---|
| 1841 | { |
---|
| 1842 | if (lpszMarker) |
---|
| 1843 | { |
---|
| 1844 | charmemset(&lpszMarker[nResult],INDENTCHAR,sizeof(XMLCHAR)*(nFormat + 1)); |
---|
| 1845 | toXMLStringUnSafe(&lpszMarker[nResult+nFormat+1],pChild); |
---|
| 1846 | lpszMarker[nResult+nFormat+1+cb]=_X('\n'); |
---|
| 1847 | } |
---|
| 1848 | nResult+=cb+nFormat+2; |
---|
| 1849 | } else |
---|
| 1850 | { |
---|
| 1851 | if (lpszMarker) toXMLStringUnSafe(&lpszMarker[nResult], pChild); |
---|
| 1852 | nResult += cb; |
---|
| 1853 | } |
---|
| 1854 | } |
---|
| 1855 | break; |
---|
| 1856 | } |
---|
| 1857 | |
---|
| 1858 | // Clear type nodes |
---|
| 1859 | case eNodeClear: |
---|
| 1860 | { |
---|
| 1861 | XMLClear *pChild=pEntry->pClear+(j>>2); |
---|
| 1862 | // "OpenTag" |
---|
| 1863 | cb = (int)LENSTR(pChild->lpszOpenTag); |
---|
| 1864 | if (cb) |
---|
| 1865 | { |
---|
| 1866 | if (nFormat!=-1) |
---|
| 1867 | { |
---|
| 1868 | if (lpszMarker) |
---|
| 1869 | { |
---|
| 1870 | charmemset(&lpszMarker[nResult], INDENTCHAR, sizeof(XMLCHAR)*(nFormat + 1)); |
---|
| 1871 | xstrcpy(&lpszMarker[nResult+nFormat+1], pChild->lpszOpenTag); |
---|
| 1872 | } |
---|
| 1873 | nResult+=cb+nFormat+1; |
---|
| 1874 | } |
---|
| 1875 | else |
---|
| 1876 | { |
---|
| 1877 | if (lpszMarker)xstrcpy(&lpszMarker[nResult], pChild->lpszOpenTag); |
---|
| 1878 | nResult += cb; |
---|
| 1879 | } |
---|
| 1880 | } |
---|
| 1881 | |
---|
| 1882 | // "OpenTag Value" |
---|
| 1883 | cb = (int)LENSTR(pChild->lpszValue); |
---|
| 1884 | if (cb) |
---|
| 1885 | { |
---|
| 1886 | if (lpszMarker) xstrcpy(&lpszMarker[nResult], pChild->lpszValue); |
---|
| 1887 | nResult += cb; |
---|
| 1888 | } |
---|
| 1889 | |
---|
| 1890 | // "OpenTag Value CloseTag" |
---|
| 1891 | cb = (int)LENSTR(pChild->lpszCloseTag); |
---|
| 1892 | if (cb) |
---|
| 1893 | { |
---|
| 1894 | if (lpszMarker) xstrcpy(&lpszMarker[nResult], pChild->lpszCloseTag); |
---|
| 1895 | nResult += cb; |
---|
| 1896 | } |
---|
| 1897 | |
---|
| 1898 | if (nFormat!=-1) |
---|
| 1899 | { |
---|
| 1900 | if (lpszMarker) lpszMarker[nResult] = _X('\n'); |
---|
| 1901 | nResult++; |
---|
| 1902 | } |
---|
| 1903 | break; |
---|
| 1904 | } |
---|
| 1905 | |
---|
| 1906 | // Element nodes |
---|
| 1907 | case eNodeChild: |
---|
| 1908 | { |
---|
| 1909 | // Recursively add child nodes |
---|
| 1910 | nResult += CreateXMLStringR(pEntry->pChild[j>>2].d, lpszMarker ? lpszMarker + nResult : 0, nChildFormat); |
---|
| 1911 | break; |
---|
| 1912 | } |
---|
| 1913 | default: break; |
---|
| 1914 | } |
---|
| 1915 | } |
---|
| 1916 | |
---|
| 1917 | if ((cbElement)&&(!pEntry->isDeclaration)) |
---|
| 1918 | { |
---|
| 1919 | // If we have child entries we need to use long XML notation for |
---|
| 1920 | // closing the element - "<elementname>blah blah blah</elementname>" |
---|
| 1921 | if (nElementI) |
---|
| 1922 | { |
---|
| 1923 | // "</elementname>\0" |
---|
| 1924 | if (lpszMarker) |
---|
| 1925 | { |
---|
| 1926 | if (nFormat != -1) |
---|
| 1927 | { |
---|
| 1928 | if (nFormat) |
---|
| 1929 | { |
---|
| 1930 | charmemset(&lpszMarker[nResult], INDENTCHAR,sizeof(XMLCHAR)*nFormat); |
---|
| 1931 | nResult+=nFormat; |
---|
| 1932 | } |
---|
| 1933 | } |
---|
| 1934 | |
---|
| 1935 | xstrcpy(&lpszMarker[nResult], _X("</")); |
---|
| 1936 | nResult += 2; |
---|
| 1937 | xstrcpy(&lpszMarker[nResult], pEntry->lpszName); |
---|
| 1938 | nResult += cbElement; |
---|
| 1939 | |
---|
| 1940 | if (nFormat == -1) |
---|
| 1941 | { |
---|
| 1942 | xstrcpy(&lpszMarker[nResult], _X(">")); |
---|
| 1943 | nResult++; |
---|
| 1944 | } else |
---|
| 1945 | { |
---|
| 1946 | xstrcpy(&lpszMarker[nResult], _X(">\n")); |
---|
| 1947 | nResult+=2; |
---|
| 1948 | } |
---|
| 1949 | } else |
---|
| 1950 | { |
---|
| 1951 | if (nFormat != -1) nResult+=cbElement+4+nFormat; |
---|
| 1952 | else nResult+=cbElement+3; |
---|
| 1953 | } |
---|
| 1954 | } else |
---|
| 1955 | { |
---|
| 1956 | // If there are no children we can use shorthand XML notation - |
---|
| 1957 | // "<elementname/>" |
---|
| 1958 | // "/>\0" |
---|
| 1959 | if (lpszMarker) |
---|
| 1960 | { |
---|
| 1961 | if (nFormat == -1) |
---|
| 1962 | { |
---|
| 1963 | xstrcpy(&lpszMarker[nResult], _X("/>")); |
---|
| 1964 | nResult += 2; |
---|
| 1965 | } |
---|
| 1966 | else |
---|
| 1967 | { |
---|
| 1968 | xstrcpy(&lpszMarker[nResult], _X("/>\n")); |
---|
| 1969 | nResult += 3; |
---|
| 1970 | } |
---|
| 1971 | } |
---|
| 1972 | else |
---|
| 1973 | { |
---|
| 1974 | nResult += nFormat == -1 ? 2 : 3; |
---|
| 1975 | } |
---|
| 1976 | } |
---|
| 1977 | } |
---|
| 1978 | |
---|
| 1979 | return nResult; |
---|
| 1980 | } |
---|
| 1981 | |
---|
| 1982 | #undef LENSTR |
---|
| 1983 | |
---|
| 1984 | // Create an XML string |
---|
| 1985 | // @param int nFormat - 0 if no formatting is required |
---|
| 1986 | // otherwise nonzero for formatted text |
---|
| 1987 | // with carriage returns and indentation. |
---|
| 1988 | // @param int *pnSize - [out] pointer to the size of the |
---|
| 1989 | // returned string not including the |
---|
| 1990 | // NULL terminator. |
---|
| 1991 | // @return XMLSTR - Allocated XML string, you must free |
---|
| 1992 | // this with free(). |
---|
| 1993 | XMLSTR XMLNode::createXMLString(int nFormat, int *pnSize) const |
---|
| 1994 | { |
---|
| 1995 | if (!d) { if (pnSize) *pnSize=0; return NULL; } |
---|
| 1996 | |
---|
| 1997 | XMLSTR lpszResult = NULL; |
---|
| 1998 | int cbStr; |
---|
| 1999 | |
---|
| 2000 | // Recursively Calculate the size of the XML string |
---|
| 2001 | if (!dropWhiteSpace) nFormat=0; |
---|
| 2002 | nFormat = nFormat ? 0 : -1; |
---|
| 2003 | cbStr = CreateXMLStringR(d, 0, nFormat); |
---|
| 2004 | assert(cbStr); |
---|
| 2005 | // Alllocate memory for the XML string + the NULL terminator and |
---|
| 2006 | // create the recursively XML string. |
---|
| 2007 | lpszResult=(XMLSTR)malloc((cbStr+1)*sizeof(XMLCHAR)); |
---|
| 2008 | CreateXMLStringR(d, lpszResult, nFormat); |
---|
| 2009 | if (pnSize) *pnSize = cbStr; |
---|
| 2010 | return lpszResult; |
---|
| 2011 | } |
---|
| 2012 | |
---|
| 2013 | int XMLNode::detachFromParent(XMLNodeData *d) |
---|
| 2014 | { |
---|
| 2015 | XMLNode *pa=d->pParent->pChild; |
---|
| 2016 | int i=0; |
---|
| 2017 | while (((void*)(pa[i].d))!=((void*)d)) i++; |
---|
| 2018 | d->pParent->nChild--; |
---|
| 2019 | if (d->pParent->nChild) memmove(pa+i,pa+i+1,(d->pParent->nChild-i)*sizeof(XMLNode)); |
---|
| 2020 | else { free(pa); d->pParent->pChild=NULL; } |
---|
| 2021 | return removeOrderElement(d->pParent,eNodeChild,i); |
---|
| 2022 | } |
---|
| 2023 | |
---|
| 2024 | XMLNode::~XMLNode() { deleteNodeContent_priv(1,0); } |
---|
| 2025 | void XMLNode::deleteNodeContent(){ deleteNodeContent_priv(0,1); } |
---|
| 2026 | void XMLNode::deleteNodeContent_priv(char isInDestuctor, char force) |
---|
| 2027 | { |
---|
| 2028 | if (!d) return; |
---|
| 2029 | if (isInDestuctor) (d->ref_count)--; |
---|
| 2030 | if ((d->ref_count==0)||force) |
---|
| 2031 | { |
---|
| 2032 | int i; |
---|
| 2033 | if (d->pParent) detachFromParent(d); |
---|
| 2034 | for(i=0; i<d->nChild; i++) { d->pChild[i].d->pParent=NULL; d->pChild[i].deleteNodeContent_priv(1,force); } |
---|
| 2035 | myFree(d->pChild); |
---|
| 2036 | for(i=0; i<d->nText; i++) free((void*)d->pText[i]); |
---|
| 2037 | myFree(d->pText); |
---|
| 2038 | for(i=0; i<d->nClear; i++) free((void*)d->pClear[i].lpszValue); |
---|
| 2039 | myFree(d->pClear); |
---|
| 2040 | for(i=0; i<d->nAttribute; i++) |
---|
| 2041 | { |
---|
| 2042 | free((void*)d->pAttribute[i].lpszName); |
---|
| 2043 | if (d->pAttribute[i].lpszValue) free((void*)d->pAttribute[i].lpszValue); |
---|
| 2044 | } |
---|
| 2045 | myFree(d->pAttribute); |
---|
| 2046 | myFree(d->pOrder); |
---|
| 2047 | myFree((void*)d->lpszName); |
---|
| 2048 | d->nChild=0; d->nText=0; d->nClear=0; d->nAttribute=0; |
---|
| 2049 | d->pChild=NULL; d->pText=NULL; d->pClear=NULL; d->pAttribute=NULL; |
---|
| 2050 | d->pOrder=NULL; d->lpszName=NULL; d->pParent=NULL; |
---|
| 2051 | } |
---|
| 2052 | if (d->ref_count==0) |
---|
| 2053 | { |
---|
| 2054 | free(d); |
---|
| 2055 | d=NULL; |
---|
| 2056 | } |
---|
| 2057 | } |
---|
| 2058 | |
---|
| 2059 | XMLNode XMLNode::deepCopy() const |
---|
| 2060 | { |
---|
| 2061 | if (!d) return XMLNode::emptyXMLNode; |
---|
| 2062 | XMLNode x(NULL,stringDup(d->lpszName),d->isDeclaration); |
---|
| 2063 | XMLNodeData *p=x.d; |
---|
| 2064 | int n=d->nAttribute; |
---|
| 2065 | if (n) |
---|
| 2066 | { |
---|
| 2067 | p->nAttribute=n; p->pAttribute=(XMLAttribute*)malloc(n*sizeof(XMLAttribute)); |
---|
| 2068 | while (n--) |
---|
| 2069 | { |
---|
| 2070 | p->pAttribute[n].lpszName=stringDup(d->pAttribute[n].lpszName); |
---|
| 2071 | p->pAttribute[n].lpszValue=stringDup(d->pAttribute[n].lpszValue); |
---|
| 2072 | } |
---|
| 2073 | } |
---|
| 2074 | if (d->pOrder) |
---|
| 2075 | { |
---|
| 2076 | n=(d->nChild+d->nText+d->nClear)*sizeof(int); p->pOrder=(int*)malloc(n); memcpy(p->pOrder,d->pOrder,n); |
---|
| 2077 | } |
---|
| 2078 | n=d->nText; |
---|
| 2079 | if (n) |
---|
| 2080 | { |
---|
| 2081 | p->nText=n; p->pText=(XMLCSTR*)malloc(n*sizeof(XMLCSTR)); |
---|
| 2082 | while(n--) p->pText[n]=stringDup(d->pText[n]); |
---|
| 2083 | } |
---|
| 2084 | n=d->nClear; |
---|
| 2085 | if (n) |
---|
| 2086 | { |
---|
| 2087 | p->nClear=n; p->pClear=(XMLClear*)malloc(n*sizeof(XMLClear)); |
---|
| 2088 | while (n--) |
---|
| 2089 | { |
---|
| 2090 | p->pClear[n].lpszCloseTag=d->pClear[n].lpszCloseTag; |
---|
| 2091 | p->pClear[n].lpszOpenTag=d->pClear[n].lpszOpenTag; |
---|
| 2092 | p->pClear[n].lpszValue=stringDup(d->pClear[n].lpszValue); |
---|
| 2093 | } |
---|
| 2094 | } |
---|
| 2095 | n=d->nChild; |
---|
| 2096 | if (n) |
---|
| 2097 | { |
---|
| 2098 | p->nChild=n; p->pChild=(XMLNode*)malloc(n*sizeof(XMLNode)); |
---|
| 2099 | while (n--) |
---|
| 2100 | { |
---|
| 2101 | p->pChild[n].d=NULL; |
---|
| 2102 | p->pChild[n]=d->pChild[n].deepCopy(); |
---|
| 2103 | p->pChild[n].d->pParent=p; |
---|
| 2104 | } |
---|
| 2105 | } |
---|
| 2106 | return x; |
---|
| 2107 | } |
---|
| 2108 | |
---|
| 2109 | XMLNode XMLNode::addChild(XMLNode childNode, int pos) |
---|
| 2110 | { |
---|
| 2111 | XMLNodeData *dc=childNode.d; |
---|
| 2112 | if ((!dc)||(!d)) return childNode; |
---|
| 2113 | if (dc->pParent) { if ((detachFromParent(dc)<=pos)&&(dc->pParent==d)) pos--; } else dc->ref_count++; |
---|
| 2114 | dc->pParent=d; |
---|
| 2115 | // int nc=d->nChild; |
---|
| 2116 | // d->pChild=(XMLNode*)myRealloc(d->pChild,(nc+1),memoryIncrease,sizeof(XMLNode)); |
---|
| 2117 | d->pChild=(XMLNode*)addToOrder(0,&pos,d->nChild,d->pChild,sizeof(XMLNode),eNodeChild); |
---|
| 2118 | d->pChild[pos].d=dc; |
---|
| 2119 | d->nChild++; |
---|
| 2120 | return childNode; |
---|
| 2121 | } |
---|
| 2122 | |
---|
| 2123 | void XMLNode::deleteAttribute(int i) |
---|
| 2124 | { |
---|
| 2125 | if ((!d)||(i<0)||(i>=d->nAttribute)) return; |
---|
| 2126 | d->nAttribute--; |
---|
| 2127 | XMLAttribute *p=d->pAttribute+i; |
---|
| 2128 | free((void*)p->lpszName); |
---|
| 2129 | if (p->lpszValue) free((void*)p->lpszValue); |
---|
| 2130 | if (d->nAttribute) memmove(p,p+1,(d->nAttribute-i)*sizeof(XMLAttribute)); else { free(p); d->pAttribute=NULL; } |
---|
| 2131 | } |
---|
| 2132 | |
---|
| 2133 | void XMLNode::deleteAttribute(XMLAttribute *a){ if (a) deleteAttribute(a->lpszName); } |
---|
| 2134 | void XMLNode::deleteAttribute(XMLCSTR lpszName) |
---|
| 2135 | { |
---|
| 2136 | int j=0; |
---|
| 2137 | getAttribute(lpszName,&j); |
---|
| 2138 | if (j) deleteAttribute(j-1); |
---|
| 2139 | } |
---|
| 2140 | |
---|
| 2141 | XMLAttribute *XMLNode::updateAttribute_WOSD(XMLSTR lpszNewValue, XMLSTR lpszNewName,int i) |
---|
| 2142 | { |
---|
| 2143 | if (!d) { if (lpszNewValue) free(lpszNewValue); if (lpszNewName) free(lpszNewName); return NULL; } |
---|
| 2144 | if (i>=d->nAttribute) |
---|
| 2145 | { |
---|
| 2146 | if (lpszNewName) return addAttribute_WOSD(lpszNewName,lpszNewValue); |
---|
| 2147 | return NULL; |
---|
| 2148 | } |
---|
| 2149 | XMLAttribute *p=d->pAttribute+i; |
---|
| 2150 | if (p->lpszValue&&p->lpszValue!=lpszNewValue) free((void*)p->lpszValue); |
---|
| 2151 | p->lpszValue=lpszNewValue; |
---|
| 2152 | if (lpszNewName&&p->lpszName!=lpszNewName) { free((void*)p->lpszName); p->lpszName=lpszNewName; }; |
---|
| 2153 | return p; |
---|
| 2154 | } |
---|
| 2155 | |
---|
| 2156 | XMLAttribute *XMLNode::updateAttribute_WOSD(XMLAttribute *newAttribute, XMLAttribute *oldAttribute) |
---|
| 2157 | { |
---|
| 2158 | if (oldAttribute) return updateAttribute_WOSD((XMLSTR)newAttribute->lpszValue,(XMLSTR)newAttribute->lpszName,oldAttribute->lpszName); |
---|
| 2159 | return addAttribute_WOSD((XMLSTR)newAttribute->lpszName,(XMLSTR)newAttribute->lpszValue); |
---|
| 2160 | } |
---|
| 2161 | |
---|
| 2162 | XMLAttribute *XMLNode::updateAttribute_WOSD(XMLSTR lpszNewValue, XMLSTR lpszNewName,XMLCSTR lpszOldName) |
---|
| 2163 | { |
---|
| 2164 | int j=0; |
---|
| 2165 | getAttribute(lpszOldName,&j); |
---|
| 2166 | if (j) return updateAttribute_WOSD(lpszNewValue,lpszNewName,j-1); |
---|
| 2167 | else |
---|
| 2168 | { |
---|
| 2169 | if (lpszNewName) return addAttribute_WOSD(lpszNewName,lpszNewValue); |
---|
| 2170 | else return addAttribute_WOSD(stringDup(lpszOldName),lpszNewValue); |
---|
| 2171 | } |
---|
| 2172 | } |
---|
| 2173 | |
---|
| 2174 | int XMLNode::indexText(XMLCSTR lpszValue) const |
---|
| 2175 | { |
---|
| 2176 | if (!d) return -1; |
---|
| 2177 | int i,l=d->nText; |
---|
| 2178 | if (!lpszValue) { if (l) return 0; return -1; } |
---|
| 2179 | XMLCSTR *p=d->pText; |
---|
| 2180 | for (i=0; i<l; i++) if (lpszValue==p[i]) return i; |
---|
| 2181 | return -1; |
---|
| 2182 | } |
---|
| 2183 | |
---|
| 2184 | void XMLNode::deleteText(int i) |
---|
| 2185 | { |
---|
| 2186 | if ((!d)||(i<0)||(i>=d->nText)) return; |
---|
| 2187 | d->nText--; |
---|
| 2188 | XMLCSTR *p=d->pText+i; |
---|
| 2189 | free((void*)*p); |
---|
| 2190 | if (d->nText) memmove(p,p+1,(d->nText-i)*sizeof(XMLCSTR)); else { free(p); d->pText=NULL; } |
---|
| 2191 | removeOrderElement(d,eNodeText,i); |
---|
| 2192 | } |
---|
| 2193 | |
---|
| 2194 | void XMLNode::deleteText(XMLCSTR lpszValue) { deleteText(indexText(lpszValue)); } |
---|
| 2195 | |
---|
| 2196 | XMLCSTR XMLNode::updateText_WOSD(XMLSTR lpszNewValue, int i) |
---|
| 2197 | { |
---|
| 2198 | if (!d) { if (lpszNewValue) free(lpszNewValue); return NULL; } |
---|
| 2199 | if (i>=d->nText) return addText_WOSD(lpszNewValue); |
---|
| 2200 | XMLCSTR *p=d->pText+i; |
---|
| 2201 | if (*p!=lpszNewValue) { free((void*)*p); *p=lpszNewValue; } |
---|
| 2202 | return lpszNewValue; |
---|
| 2203 | } |
---|
| 2204 | |
---|
| 2205 | XMLCSTR XMLNode::updateText_WOSD(XMLSTR lpszNewValue, XMLCSTR lpszOldValue) |
---|
| 2206 | { |
---|
| 2207 | if (!d) { if (lpszNewValue) free(lpszNewValue); return NULL; } |
---|
| 2208 | int i=indexText(lpszOldValue); |
---|
| 2209 | if (i>=0) return updateText_WOSD(lpszNewValue,i); |
---|
| 2210 | return addText_WOSD(lpszNewValue); |
---|
| 2211 | } |
---|
| 2212 | |
---|
| 2213 | void XMLNode::deleteClear(int i) |
---|
| 2214 | { |
---|
| 2215 | if ((!d)||(i<0)||(i>=d->nClear)) return; |
---|
| 2216 | d->nClear--; |
---|
| 2217 | XMLClear *p=d->pClear+i; |
---|
| 2218 | free((void*)p->lpszValue); |
---|
| 2219 | if (d->nClear) memmove(p,p+1,(d->nClear-i)*sizeof(XMLClear)); else { free(p); d->pClear=NULL; } |
---|
| 2220 | removeOrderElement(d,eNodeClear,i); |
---|
| 2221 | } |
---|
| 2222 | |
---|
| 2223 | int XMLNode::indexClear(XMLCSTR lpszValue) const |
---|
| 2224 | { |
---|
| 2225 | if (!d) return -1; |
---|
| 2226 | int i,l=d->nClear; |
---|
| 2227 | if (!lpszValue) { if (l) return 0; return -1; } |
---|
| 2228 | XMLClear *p=d->pClear; |
---|
| 2229 | for (i=0; i<l; i++) if (lpszValue==p[i].lpszValue) return i; |
---|
| 2230 | return -1; |
---|
| 2231 | } |
---|
| 2232 | |
---|
| 2233 | void XMLNode::deleteClear(XMLCSTR lpszValue) { deleteClear(indexClear(lpszValue)); } |
---|
| 2234 | void XMLNode::deleteClear(XMLClear *a) { if (a) deleteClear(a->lpszValue); } |
---|
| 2235 | |
---|
| 2236 | XMLClear *XMLNode::updateClear_WOSD(XMLSTR lpszNewContent, int i) |
---|
| 2237 | { |
---|
| 2238 | if (!d) { if (lpszNewContent) free(lpszNewContent); return NULL; } |
---|
| 2239 | if (i>=d->nClear) return addClear_WOSD(lpszNewContent); |
---|
| 2240 | XMLClear *p=d->pClear+i; |
---|
| 2241 | if (lpszNewContent!=p->lpszValue) { free((void*)p->lpszValue); p->lpszValue=lpszNewContent; } |
---|
| 2242 | return p; |
---|
| 2243 | } |
---|
| 2244 | |
---|
| 2245 | XMLClear *XMLNode::updateClear_WOSD(XMLSTR lpszNewContent, XMLCSTR lpszOldValue) |
---|
| 2246 | { |
---|
| 2247 | if (!d) { if (lpszNewContent) free(lpszNewContent); return NULL; } |
---|
| 2248 | int i=indexClear(lpszOldValue); |
---|
| 2249 | if (i>=0) return updateClear_WOSD(lpszNewContent,i); |
---|
| 2250 | return addClear_WOSD(lpszNewContent); |
---|
| 2251 | } |
---|
| 2252 | |
---|
| 2253 | XMLClear *XMLNode::updateClear_WOSD(XMLClear *newP,XMLClear *oldP) |
---|
| 2254 | { |
---|
| 2255 | if (oldP) return updateClear_WOSD((XMLSTR)newP->lpszValue,(XMLSTR)oldP->lpszValue); |
---|
| 2256 | return NULL; |
---|
| 2257 | } |
---|
| 2258 | |
---|
| 2259 | XMLNode& XMLNode::operator=( const XMLNode& A ) |
---|
| 2260 | { |
---|
| 2261 | // shallow copy |
---|
| 2262 | if (this != &A) |
---|
| 2263 | { |
---|
| 2264 | deleteNodeContent_priv(1,0); |
---|
| 2265 | d=A.d; |
---|
| 2266 | if (d) (d->ref_count) ++ ; |
---|
| 2267 | } |
---|
| 2268 | return *this; |
---|
| 2269 | } |
---|
| 2270 | |
---|
| 2271 | XMLNode::XMLNode(const XMLNode &A) |
---|
| 2272 | { |
---|
| 2273 | // shallow copy |
---|
| 2274 | d=A.d; |
---|
| 2275 | if (d) (d->ref_count)++ ; |
---|
| 2276 | } |
---|
| 2277 | |
---|
| 2278 | int XMLNode::nChildNode(XMLCSTR name) const |
---|
| 2279 | { |
---|
| 2280 | if (!d) return 0; |
---|
| 2281 | int i,j=0,n=d->nChild; |
---|
| 2282 | XMLNode *pc=d->pChild; |
---|
| 2283 | for (i=0; i<n; i++) |
---|
| 2284 | { |
---|
| 2285 | if (xstricmp(pc->d->lpszName, name)==0) j++; |
---|
| 2286 | pc++; |
---|
| 2287 | } |
---|
| 2288 | return j; |
---|
| 2289 | } |
---|
| 2290 | |
---|
| 2291 | XMLNode XMLNode::getChildNode(XMLCSTR name, int *j) const |
---|
| 2292 | { |
---|
| 2293 | if (!d) return emptyXMLNode; |
---|
| 2294 | int i=0,n=d->nChild; |
---|
| 2295 | if (j) i=*j; |
---|
| 2296 | XMLNode *pc=d->pChild+i; |
---|
| 2297 | for (; i<n; i++) |
---|
| 2298 | { |
---|
| 2299 | if (xstricmp(pc->d->lpszName, name)==0) |
---|
| 2300 | { |
---|
| 2301 | if (j) *j=i+1; |
---|
| 2302 | return *pc; |
---|
| 2303 | } |
---|
| 2304 | pc++; |
---|
| 2305 | } |
---|
| 2306 | return emptyXMLNode; |
---|
| 2307 | } |
---|
| 2308 | |
---|
| 2309 | XMLNode XMLNode::getChildNode(XMLCSTR name, int j) const |
---|
| 2310 | { |
---|
| 2311 | if (!d) return emptyXMLNode; |
---|
| 2312 | int i=0; |
---|
| 2313 | while (j-->0) getChildNode(name,&i); |
---|
| 2314 | return getChildNode(name,&i); |
---|
| 2315 | } |
---|
| 2316 | |
---|
| 2317 | XMLElementPosition XMLNode::positionOfText (int i) const { if (i>=d->nText ) i=d->nText-1; return findPosition(d,i,eNodeText ); } |
---|
| 2318 | XMLElementPosition XMLNode::positionOfClear (int i) const { if (i>=d->nClear) i=d->nClear-1; return findPosition(d,i,eNodeClear); } |
---|
| 2319 | XMLElementPosition XMLNode::positionOfChildNode(int i) const { if (i>=d->nChild) i=d->nChild-1; return findPosition(d,i,eNodeChild); } |
---|
| 2320 | XMLElementPosition XMLNode::positionOfText (XMLCSTR lpszValue) const { return positionOfText (indexText (lpszValue)); } |
---|
| 2321 | XMLElementPosition XMLNode::positionOfClear(XMLCSTR lpszValue) const { return positionOfClear(indexClear(lpszValue)); } |
---|
| 2322 | XMLElementPosition XMLNode::positionOfClear(XMLClear *a) const { if (a) return positionOfClear(a->lpszValue); return positionOfClear(); } |
---|
| 2323 | XMLElementPosition XMLNode::positionOfChildNode(XMLNode x) const |
---|
| 2324 | { |
---|
| 2325 | if ((!d)||(!x.d)) return -1; |
---|
| 2326 | XMLNodeData *dd=x.d; |
---|
| 2327 | XMLNode *pc=d->pChild; |
---|
| 2328 | int i=d->nChild; |
---|
| 2329 | while (i--) if (pc[i].d==dd) return findPosition(d,i,eNodeChild); |
---|
| 2330 | return -1; |
---|
| 2331 | } |
---|
| 2332 | XMLElementPosition XMLNode::positionOfChildNode(XMLCSTR name, int count) const |
---|
| 2333 | { |
---|
| 2334 | if (!name) return positionOfChildNode(count); |
---|
| 2335 | int j=0; |
---|
| 2336 | do { getChildNode(name,&j); if (j<0) return -1; } while (count--); |
---|
| 2337 | return findPosition(d,j-1,eNodeChild); |
---|
| 2338 | } |
---|
| 2339 | |
---|
| 2340 | XMLNode XMLNode::getChildNodeWithAttribute(XMLCSTR name,XMLCSTR attributeName,XMLCSTR attributeValue, int *k) const |
---|
| 2341 | { |
---|
| 2342 | int i=0,j; |
---|
| 2343 | if (k) i=*k; |
---|
| 2344 | XMLNode x; |
---|
| 2345 | XMLCSTR t; |
---|
| 2346 | do |
---|
| 2347 | { |
---|
| 2348 | x=getChildNode(name,&i); |
---|
| 2349 | if (!x.isEmpty()) |
---|
| 2350 | { |
---|
| 2351 | if (attributeValue) |
---|
| 2352 | { |
---|
| 2353 | j=0; |
---|
| 2354 | do |
---|
| 2355 | { |
---|
| 2356 | t=x.getAttribute(attributeName,&j); |
---|
| 2357 | if (t&&(xstricmp(attributeValue,t)==0)) { if (k) *k=i+1; return x; } |
---|
| 2358 | } while (t); |
---|
| 2359 | } else |
---|
| 2360 | { |
---|
| 2361 | if (x.isAttributeSet(attributeName)) { if (k) *k=i+1; return x; } |
---|
| 2362 | } |
---|
| 2363 | } |
---|
| 2364 | } while (!x.isEmpty()); |
---|
| 2365 | return emptyXMLNode; |
---|
| 2366 | } |
---|
| 2367 | |
---|
| 2368 | // Find an attribute on an node. |
---|
| 2369 | XMLCSTR XMLNode::getAttribute(XMLCSTR lpszAttrib, int *j) const |
---|
| 2370 | { |
---|
| 2371 | if (!d) return NULL; |
---|
| 2372 | int i=0,n=d->nAttribute; |
---|
| 2373 | if (j) i=*j; |
---|
| 2374 | XMLAttribute *pAttr=d->pAttribute+i; |
---|
| 2375 | for (; i<n; i++) |
---|
| 2376 | { |
---|
| 2377 | if (xstricmp(pAttr->lpszName, lpszAttrib)==0) |
---|
| 2378 | { |
---|
| 2379 | if (j) *j=i+1; |
---|
| 2380 | return pAttr->lpszValue; |
---|
| 2381 | } |
---|
| 2382 | pAttr++; |
---|
| 2383 | } |
---|
| 2384 | return NULL; |
---|
| 2385 | } |
---|
| 2386 | |
---|
| 2387 | char XMLNode::isAttributeSet(XMLCSTR lpszAttrib) const |
---|
| 2388 | { |
---|
| 2389 | if (!d) return FALSE; |
---|
| 2390 | int i,n=d->nAttribute; |
---|
| 2391 | XMLAttribute *pAttr=d->pAttribute; |
---|
| 2392 | for (i=0; i<n; i++) |
---|
| 2393 | { |
---|
| 2394 | if (xstricmp(pAttr->lpszName, lpszAttrib)==0) |
---|
| 2395 | { |
---|
| 2396 | return TRUE; |
---|
| 2397 | } |
---|
| 2398 | pAttr++; |
---|
| 2399 | } |
---|
| 2400 | return FALSE; |
---|
| 2401 | } |
---|
| 2402 | |
---|
| 2403 | XMLCSTR XMLNode::getAttribute(XMLCSTR name, int j) const |
---|
| 2404 | { |
---|
| 2405 | if (!d) return NULL; |
---|
| 2406 | int i=0; |
---|
| 2407 | while (j-->0) getAttribute(name,&i); |
---|
| 2408 | return getAttribute(name,&i); |
---|
| 2409 | } |
---|
| 2410 | |
---|
| 2411 | XMLNodeContents XMLNode::enumContents(int i) const |
---|
| 2412 | { |
---|
| 2413 | XMLNodeContents c; |
---|
| 2414 | if (!d) { c.etype=eNodeNULL; return c; } |
---|
| 2415 | if (i<d->nAttribute) |
---|
| 2416 | { |
---|
| 2417 | c.etype=eNodeAttribute; |
---|
| 2418 | c.attrib=d->pAttribute[i]; |
---|
| 2419 | return c; |
---|
| 2420 | } |
---|
| 2421 | i-=d->nAttribute; |
---|
| 2422 | c.etype=(XMLElementType)(d->pOrder[i]&3); |
---|
| 2423 | i=(d->pOrder[i])>>2; |
---|
| 2424 | switch (c.etype) |
---|
| 2425 | { |
---|
| 2426 | case eNodeChild: c.child = d->pChild[i]; break; |
---|
| 2427 | case eNodeText: c.text = d->pText[i]; break; |
---|
| 2428 | case eNodeClear: c.clear = d->pClear[i]; break; |
---|
| 2429 | default: break; |
---|
| 2430 | } |
---|
| 2431 | return c; |
---|
| 2432 | } |
---|
| 2433 | |
---|
| 2434 | XMLCSTR XMLNode::getName() const { if (!d) return NULL; return d->lpszName; } |
---|
| 2435 | int XMLNode::nText() const { if (!d) return 0; return d->nText; } |
---|
| 2436 | int XMLNode::nChildNode() const { if (!d) return 0; return d->nChild; } |
---|
| 2437 | int XMLNode::nAttribute() const { if (!d) return 0; return d->nAttribute; } |
---|
| 2438 | int XMLNode::nClear() const { if (!d) return 0; return d->nClear; } |
---|
| 2439 | int XMLNode::nElement() const { if (!d) return 0; return d->nAttribute+d->nChild+d->nText+d->nClear; } |
---|
| 2440 | XMLClear XMLNode::getClear (int i) const { if ((!d)||(i>=d->nClear )) return emptyXMLClear; return d->pClear[i]; } |
---|
| 2441 | XMLAttribute XMLNode::getAttribute (int i) const { if ((!d)||(i>=d->nAttribute)) return emptyXMLAttribute; return d->pAttribute[i]; } |
---|
| 2442 | XMLCSTR XMLNode::getAttributeName (int i) const { if ((!d)||(i>=d->nAttribute)) return NULL; return d->pAttribute[i].lpszName; } |
---|
| 2443 | XMLCSTR XMLNode::getAttributeValue(int i) const { if ((!d)||(i>=d->nAttribute)) return NULL; return d->pAttribute[i].lpszValue; } |
---|
| 2444 | XMLCSTR XMLNode::getText (int i) const { if ((!d)||(i>=d->nText )) return NULL; return d->pText[i]; } |
---|
| 2445 | XMLNode XMLNode::getChildNode (int i) const { if ((!d)||(i>=d->nChild )) return emptyXMLNode; return d->pChild[i]; } |
---|
| 2446 | XMLNode XMLNode::getParentNode ( ) const { if ((!d)||(!d->pParent )) return emptyXMLNode; return XMLNode(d->pParent); } |
---|
| 2447 | char XMLNode::isDeclaration ( ) const { if (!d) return 0; return d->isDeclaration; } |
---|
| 2448 | char XMLNode::isEmpty ( ) const { return (d==NULL); } |
---|
| 2449 | XMLNode XMLNode::emptyNode ( ) { return XMLNode::emptyXMLNode; } |
---|
| 2450 | |
---|
| 2451 | XMLNode XMLNode::addChild(XMLCSTR lpszName, char isDeclaration, XMLElementPosition pos) |
---|
| 2452 | { return addChild_priv(0,stringDup(lpszName),isDeclaration,pos); } |
---|
| 2453 | XMLNode XMLNode::addChild_WOSD(XMLSTR lpszName, char isDeclaration, XMLElementPosition pos) |
---|
| 2454 | { return addChild_priv(0,lpszName,isDeclaration,pos); } |
---|
| 2455 | XMLAttribute *XMLNode::addAttribute(XMLCSTR lpszName, XMLCSTR lpszValue) |
---|
| 2456 | { return addAttribute_priv(0,stringDup(lpszName),stringDup(lpszValue)); } |
---|
| 2457 | XMLAttribute *XMLNode::addAttribute_WOSD(XMLSTR lpszName, XMLSTR lpszValuev) |
---|
| 2458 | { return addAttribute_priv(0,lpszName,lpszValuev); } |
---|
| 2459 | XMLCSTR XMLNode::addText(XMLCSTR lpszValue, XMLElementPosition pos) |
---|
| 2460 | { return addText_priv(0,stringDup(lpszValue),pos); } |
---|
| 2461 | XMLCSTR XMLNode::addText_WOSD(XMLSTR lpszValue, XMLElementPosition pos) |
---|
| 2462 | { return addText_priv(0,lpszValue,pos); } |
---|
| 2463 | XMLClear *XMLNode::addClear(XMLCSTR lpszValue, XMLCSTR lpszOpen, XMLCSTR lpszClose, XMLElementPosition pos) |
---|
| 2464 | { return addClear_priv(0,stringDup(lpszValue),lpszOpen,lpszClose,pos); } |
---|
| 2465 | XMLClear *XMLNode::addClear_WOSD(XMLSTR lpszValue, XMLCSTR lpszOpen, XMLCSTR lpszClose, XMLElementPosition pos) |
---|
| 2466 | { return addClear_priv(0,lpszValue,lpszOpen,lpszClose,pos); } |
---|
| 2467 | XMLCSTR XMLNode::updateName(XMLCSTR lpszName) |
---|
| 2468 | { return updateName_WOSD(stringDup(lpszName)); } |
---|
| 2469 | XMLAttribute *XMLNode::updateAttribute(XMLAttribute *newAttribute, XMLAttribute *oldAttribute) |
---|
| 2470 | { return updateAttribute_WOSD(stringDup(newAttribute->lpszValue),stringDup(newAttribute->lpszName),oldAttribute->lpszName); } |
---|
| 2471 | XMLAttribute *XMLNode::updateAttribute(XMLCSTR lpszNewValue, XMLCSTR lpszNewName,int i) |
---|
| 2472 | { return updateAttribute_WOSD(stringDup(lpszNewValue),stringDup(lpszNewName),i); } |
---|
| 2473 | XMLAttribute *XMLNode::updateAttribute(XMLCSTR lpszNewValue, XMLCSTR lpszNewName,XMLCSTR lpszOldName) |
---|
| 2474 | { return updateAttribute_WOSD(stringDup(lpszNewValue),stringDup(lpszNewName),lpszOldName); } |
---|
| 2475 | XMLCSTR XMLNode::updateText(XMLCSTR lpszNewValue, int i) |
---|
| 2476 | { return updateText_WOSD(stringDup(lpszNewValue),i); } |
---|
| 2477 | XMLCSTR XMLNode::updateText(XMLCSTR lpszNewValue, XMLCSTR lpszOldValue) |
---|
| 2478 | { return updateText_WOSD(stringDup(lpszNewValue),lpszOldValue); } |
---|
| 2479 | XMLClear *XMLNode::updateClear(XMLCSTR lpszNewContent, int i) |
---|
| 2480 | { return updateClear_WOSD(stringDup(lpszNewContent),i); } |
---|
| 2481 | XMLClear *XMLNode::updateClear(XMLCSTR lpszNewValue, XMLCSTR lpszOldValue) |
---|
| 2482 | { return updateClear_WOSD(stringDup(lpszNewValue),lpszOldValue); } |
---|
| 2483 | XMLClear *XMLNode::updateClear(XMLClear *newP,XMLClear *oldP) |
---|
| 2484 | { return updateClear_WOSD(stringDup(newP->lpszValue),oldP->lpszValue); } |
---|
| 2485 | |
---|
| 2486 | char XMLNode::setGlobalOptions(XMLCharEncoding _characterEncoding, char _guessWideCharChars, char _dropWhiteSpace) |
---|
| 2487 | { |
---|
| 2488 | guessWideCharChars=_guessWideCharChars; dropWhiteSpace=_dropWhiteSpace; |
---|
| 2489 | #ifdef _XMLWIDECHAR |
---|
| 2490 | if (_characterEncoding) characterEncoding=_characterEncoding; |
---|
| 2491 | #else |
---|
| 2492 | switch(_characterEncoding) |
---|
| 2493 | { |
---|
| 2494 | case encoding_UTF8: characterEncoding=_characterEncoding; XML_ByteTable=XML_utf8ByteTable; break; |
---|
| 2495 | case encoding_ascii: characterEncoding=_characterEncoding; XML_ByteTable=XML_asciiByteTable; break; |
---|
| 2496 | case encoding_ShiftJIS: characterEncoding=_characterEncoding; XML_ByteTable=XML_sjisByteTable; break; |
---|
| 2497 | default: return 1; |
---|
| 2498 | } |
---|
| 2499 | #endif |
---|
| 2500 | return 0; |
---|
| 2501 | } |
---|
| 2502 | |
---|
| 2503 | XMLNode::XMLCharEncoding XMLNode::guessCharEncoding(void *buf,int l, char useXMLEncodingAttribute) |
---|
| 2504 | { |
---|
| 2505 | #ifdef _XMLWIDECHAR |
---|
| 2506 | return (XMLCharEncoding)0; |
---|
| 2507 | #else |
---|
| 2508 | if (l<25) return (XMLCharEncoding)0; |
---|
| 2509 | if (guessWideCharChars&&(myIsTextWideChar(buf,l))) return (XMLCharEncoding)0; |
---|
| 2510 | unsigned char *b=(unsigned char*)buf; |
---|
| 2511 | if ((b[0]==0xef)&&(b[1]==0xbb)&&(b[2]==0xbf)) return encoding_UTF8; |
---|
| 2512 | |
---|
| 2513 | // Match utf-8 model ? |
---|
| 2514 | XMLCharEncoding bestGuess=encoding_UTF8; |
---|
| 2515 | int i=0; |
---|
| 2516 | while (i<l) |
---|
| 2517 | switch (XML_utf8ByteTable[b[i]]) |
---|
| 2518 | { |
---|
| 2519 | case 4: i++; if ((i<l)&&(b[i]& 0xC0)!=0x80) { bestGuess=encoding_ascii; i=l; } // 10bbbbbb ? |
---|
| 2520 | case 3: i++; if ((i<l)&&(b[i]& 0xC0)!=0x80) { bestGuess=encoding_ascii; i=l; } // 10bbbbbb ? |
---|
| 2521 | case 2: i++; if ((i<l)&&(b[i]& 0xC0)!=0x80) { bestGuess=encoding_ascii; i=l; } // 10bbbbbb ? |
---|
| 2522 | case 1: i++; break; |
---|
| 2523 | case 0: i=l; |
---|
| 2524 | } |
---|
| 2525 | if (!useXMLEncodingAttribute) return bestGuess; |
---|
| 2526 | // if encoding is specified and different from utf-8 than it's non-utf8 |
---|
| 2527 | // otherwise it's utf-8 |
---|
| 2528 | char bb[201]; |
---|
| 2529 | l=mmin(l,200); |
---|
| 2530 | memcpy(bb,buf,l); // copy buf into bb to be able to do "bb[l]=0" |
---|
| 2531 | bb[l]=0; |
---|
| 2532 | b=(unsigned char*)strstr(bb,"encoding"); |
---|
| 2533 | if (!b) return bestGuess; |
---|
| 2534 | b+=8; while XML_isSPACECHAR(*b) b++; if (*b!='=') return bestGuess; |
---|
| 2535 | b++; while XML_isSPACECHAR(*b) b++; if ((*b!='\'')&&(*b!='"')) return bestGuess; |
---|
| 2536 | b++; while XML_isSPACECHAR(*b) b++; |
---|
| 2537 | |
---|
| 2538 | if ((_strnicmp((char*)b,"utf-8",5)==0)|| |
---|
| 2539 | (_strnicmp((char*)b,"utf8",4)==0)) |
---|
| 2540 | { |
---|
| 2541 | if (bestGuess==encoding_ascii) return (XMLCharEncoding)0; |
---|
| 2542 | return encoding_UTF8; |
---|
| 2543 | } |
---|
| 2544 | |
---|
| 2545 | if ((_strnicmp((char*)b,"shiftjis",8)==0)|| |
---|
| 2546 | (_strnicmp((char*)b,"shift-jis",9)==0)|| |
---|
| 2547 | (_strnicmp((char*)b,"sjis",4)==0)) return encoding_ShiftJIS; |
---|
| 2548 | |
---|
| 2549 | return encoding_ascii; |
---|
| 2550 | #endif |
---|
| 2551 | } |
---|
| 2552 | #undef XML_isSPACECHAR |
---|
| 2553 | |
---|
| 2554 | ////////////////////////////////////////////////////////// |
---|
| 2555 | // Here starts the base64 conversion functions. // |
---|
| 2556 | ////////////////////////////////////////////////////////// |
---|
| 2557 | |
---|
| 2558 | static const char base64Fillchar = _X('='); // used to mark partial words at the end |
---|
| 2559 | |
---|
| 2560 | // this lookup table defines the base64 encoding |
---|
| 2561 | XMLCSTR base64EncodeTable=_X("ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/"); |
---|
| 2562 | |
---|
| 2563 | // Decode Table gives the index of any valid base64 character in the Base64 table] |
---|
| 2564 | // 96: '=' - 97: space char - 98: illegal char - 99: end of string |
---|
| 2565 | const unsigned char base64DecodeTable[] = { |
---|
| 2566 | 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 |
---|
| 2567 | 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 |
---|
| 2568 | 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 |
---|
| 2569 | 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 |
---|
| 2570 | 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 |
---|
| 2571 | 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 |
---|
| 2572 | 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 |
---|
| 2573 | 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 |
---|
| 2574 | 98,98,98,98,98,98,98,98,98,98, 98,98,98,98,98,98 //240 -255 |
---|
| 2575 | }; |
---|
| 2576 | |
---|
| 2577 | XMLParserBase64Tool::~XMLParserBase64Tool(){ freeBuffer(); } |
---|
| 2578 | |
---|
| 2579 | void XMLParserBase64Tool::freeBuffer(){ if (buf) free(buf); buf=NULL; buflen=0; } |
---|
| 2580 | |
---|
| 2581 | int XMLParserBase64Tool::encodeLength(int inlen, char formatted) |
---|
| 2582 | { |
---|
| 2583 | unsigned int i=((inlen-1)/3*4+4+1); |
---|
| 2584 | if (formatted) i+=inlen/54; |
---|
| 2585 | return i; |
---|
| 2586 | } |
---|
| 2587 | |
---|
| 2588 | XMLSTR XMLParserBase64Tool::encode(unsigned char *inbuf, unsigned int inlen, char formatted) |
---|
| 2589 | { |
---|
| 2590 | int i=encodeLength(inlen,formatted),k=17,eLen=inlen/3,j; |
---|
| 2591 | alloc(i*sizeof(XMLCHAR)); |
---|
| 2592 | XMLSTR curr=(XMLSTR)buf; |
---|
| 2593 | for(i=0;i<eLen;i++) |
---|
| 2594 | { |
---|
| 2595 | // Copy next three bytes into lower 24 bits of int, paying attention to sign. |
---|
| 2596 | j=(inbuf[0]<<16)|(inbuf[1]<<8)|inbuf[2]; inbuf+=3; |
---|
| 2597 | // Encode the int into four chars |
---|
| 2598 | *(curr++)=base64EncodeTable[ j>>18 ]; |
---|
| 2599 | *(curr++)=base64EncodeTable[(j>>12)&0x3f]; |
---|
| 2600 | *(curr++)=base64EncodeTable[(j>> 6)&0x3f]; |
---|
| 2601 | *(curr++)=base64EncodeTable[(j )&0x3f]; |
---|
| 2602 | if (formatted) { if (!k) { *(curr++)=_X('\n'); k=18; } k--; } |
---|
| 2603 | } |
---|
| 2604 | eLen=inlen-eLen*3; // 0 - 2. |
---|
| 2605 | if (eLen==1) |
---|
| 2606 | { |
---|
| 2607 | *(curr++)=base64EncodeTable[ inbuf[0]>>2 ]; |
---|
| 2608 | *(curr++)=base64EncodeTable[(inbuf[0]<<4)&0x3F]; |
---|
| 2609 | *(curr++)=base64Fillchar; |
---|
| 2610 | *(curr++)=base64Fillchar; |
---|
| 2611 | } else if (eLen==2) |
---|
| 2612 | { |
---|
| 2613 | j=(inbuf[0]<<8)|inbuf[1]; |
---|
| 2614 | *(curr++)=base64EncodeTable[ j>>10 ]; |
---|
| 2615 | *(curr++)=base64EncodeTable[(j>> 4)&0x3f]; |
---|
| 2616 | *(curr++)=base64EncodeTable[(j<< 2)&0x3f]; |
---|
| 2617 | *(curr++)=base64Fillchar; |
---|
| 2618 | } |
---|
| 2619 | *(curr++)=0; |
---|
| 2620 | return (XMLSTR)buf; |
---|
| 2621 | } |
---|
| 2622 | |
---|
| 2623 | unsigned int XMLParserBase64Tool::decodeSize(XMLCSTR data,XMLError *xe) |
---|
| 2624 | { |
---|
| 2625 | if (xe) *xe=eXMLErrorNone; |
---|
| 2626 | int size=0; |
---|
| 2627 | unsigned char c; |
---|
| 2628 | //skip any extra characters (e.g. newlines or spaces) |
---|
| 2629 | while (*data) |
---|
| 2630 | { |
---|
| 2631 | #ifdef _XMLWIDECHAR |
---|
| 2632 | if (*data>255) { if (xe) *xe=eXMLErrorBase64DecodeIllegalCharacter; return 0; } |
---|
| 2633 | #endif |
---|
| 2634 | c=base64DecodeTable[(unsigned char)(*data)]; |
---|
| 2635 | if (c<97) size++; |
---|
| 2636 | else if (c==98) { if (xe) *xe=eXMLErrorBase64DecodeIllegalCharacter; return 0; } |
---|
| 2637 | data++; |
---|
| 2638 | } |
---|
| 2639 | if (xe&&(size%4!=0)) *xe=eXMLErrorBase64DataSizeIsNotMultipleOf4; |
---|
| 2640 | if (size==0) return 0; |
---|
| 2641 | do { data--; size--; } while(*data==base64Fillchar); size++; |
---|
| 2642 | return (unsigned int)((size*3)/4); |
---|
| 2643 | } |
---|
| 2644 | |
---|
| 2645 | unsigned char XMLParserBase64Tool::decode(XMLCSTR data, unsigned char *buf, int len, XMLError *xe) |
---|
| 2646 | { |
---|
| 2647 | if (xe) *xe=eXMLErrorNone; |
---|
| 2648 | int i=0,p=0; |
---|
| 2649 | unsigned char d,c; |
---|
| 2650 | for(;;) |
---|
| 2651 | { |
---|
| 2652 | |
---|
| 2653 | #ifdef _XMLWIDECHAR |
---|
| 2654 | #define BASE64DECODE_READ_NEXT_CHAR(c) \ |
---|
| 2655 | do { \ |
---|
| 2656 | if (data[i]>255){ c=98; break; } \ |
---|
| 2657 | c=base64DecodeTable[(unsigned char)data[i++]]; \ |
---|
| 2658 | }while (c==97); \ |
---|
| 2659 | if(c==98){ if(xe)*xe=eXMLErrorBase64DecodeIllegalCharacter; return 0; } |
---|
| 2660 | #else |
---|
| 2661 | #define BASE64DECODE_READ_NEXT_CHAR(c) \ |
---|
| 2662 | do { c=base64DecodeTable[(unsigned char)data[i++]]; }while (c==97); \ |
---|
| 2663 | if(c==98){ if(xe)*xe=eXMLErrorBase64DecodeIllegalCharacter; return 0; } |
---|
| 2664 | #endif |
---|
| 2665 | |
---|
| 2666 | BASE64DECODE_READ_NEXT_CHAR(c) |
---|
| 2667 | if (c==99) { return 2; } |
---|
| 2668 | if (c==96) |
---|
| 2669 | { |
---|
| 2670 | if (p==(int)len) return 2; |
---|
| 2671 | if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; |
---|
| 2672 | return 1; |
---|
| 2673 | } |
---|
| 2674 | |
---|
| 2675 | BASE64DECODE_READ_NEXT_CHAR(d) |
---|
| 2676 | if ((d==99)||(d==96)) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; } |
---|
| 2677 | if (p==(int)len) { if (xe) *xe=eXMLErrorBase64DecodeBufferTooSmall; return 0; } |
---|
| 2678 | buf[p++]=(unsigned char)((c<<2)|((d>>4)&0x3)); |
---|
| 2679 | |
---|
| 2680 | BASE64DECODE_READ_NEXT_CHAR(c) |
---|
| 2681 | if (c==99) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; } |
---|
| 2682 | if (p==(int)len) |
---|
| 2683 | { |
---|
| 2684 | if (c==96) return 2; |
---|
| 2685 | if (xe) *xe=eXMLErrorBase64DecodeBufferTooSmall; |
---|
| 2686 | return 0; |
---|
| 2687 | } |
---|
| 2688 | if (c==96) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; } |
---|
| 2689 | buf[p++]=(unsigned char)(((d<<4)&0xf0)|((c>>2)&0xf)); |
---|
| 2690 | |
---|
| 2691 | BASE64DECODE_READ_NEXT_CHAR(d) |
---|
| 2692 | if (d==99 ) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; } |
---|
| 2693 | if (p==(int)len) |
---|
| 2694 | { |
---|
| 2695 | if (d==96) return 2; |
---|
| 2696 | if (xe) *xe=eXMLErrorBase64DecodeBufferTooSmall; |
---|
| 2697 | return 0; |
---|
| 2698 | } |
---|
| 2699 | if (d==96) { if (xe) *xe=eXMLErrorBase64DecodeTruncatedData; return 1; } |
---|
| 2700 | buf[p++]=(unsigned char)(((c<<6)&0xc0)|d); |
---|
| 2701 | } |
---|
| 2702 | } |
---|
| 2703 | #undef BASE64DECODE_READ_NEXT_CHAR |
---|
| 2704 | |
---|
| 2705 | void XMLParserBase64Tool::alloc(int newsize) |
---|
| 2706 | { |
---|
| 2707 | if ((!buf)&&(newsize)) { buf=malloc(newsize); buflen=newsize; return; } |
---|
| 2708 | if (newsize>buflen) { buf=realloc(buf,newsize); buflen=newsize; } |
---|
| 2709 | } |
---|
| 2710 | |
---|
| 2711 | unsigned char *XMLParserBase64Tool::decode(XMLCSTR data, int *outlen, XMLError *xe) |
---|
| 2712 | { |
---|
| 2713 | if (xe) *xe=eXMLErrorNone; |
---|
| 2714 | unsigned int len=decodeSize(data,xe); |
---|
| 2715 | if (outlen) *outlen=len; |
---|
| 2716 | if (!len) return NULL; |
---|
| 2717 | alloc(len+1); |
---|
| 2718 | if(!decode(data,(unsigned char*)buf,len,xe)){ return NULL; } |
---|
| 2719 | return (unsigned char*)buf; |
---|
| 2720 | } |
---|
| 2721 | |
---|