xmlParser.h revision 10152
1/****************************************************************************/ 2/*! \mainpage XMLParser library 3 * \section intro_sec Introduction 4 * 5 * This is a basic XML parser written in ANSI C++ for portability. 6 * It works by using recursion and a node tree for breaking 7 * down the elements of an XML document. 8 * 9 * @version V2.41 10 * @author Frank Vanden Berghen 11 * 12 * The following license terms for the "XMLParser library from Business-Insight" apply to projects 13 * that are in some way related to 14 * the "mcpat project", including applications 15 * using "mcpat project" and tools developed 16 * for enhancing "mcpat project". All other projects 17 * (not related to "mcpat project") have to use the "XMLParser library from Business-Insight" 18 * code under the Aladdin Free Public License (AFPL) 19 * See the file "AFPL-license.txt" for more informations about the AFPL license. 20 * (see http://www.artifex.com/downloads/doc/Public.htm for detailed AFPL terms) 21 * 22 * Redistribution and use of the "XMLParser library from Business-Insight" in source and binary forms, with or without 23 * modification, are permitted provided that the following conditions are met: 24 * * Redistributions of source code must retain the above copyright 25 * notice, this list of conditions and the following disclaimer. 26 * * Redistributions in binary form must reproduce the above copyright 27 * notice, this list of conditions and the following disclaimer in the 28 * documentation and/or other materials provided with the distribution. 29 * * Neither the name of Frank Vanden Berghen nor the 30 * names of its contributors may be used to endorse or promote products 31 * derived from this software without specific prior written permission. 32 * 33 * THIS SOFTWARE IS PROVIDED BY Business-Insight ``AS IS'' AND ANY 34 * EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 35 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 36 * DISCLAIMED. IN NO EVENT SHALL Business-Insight BE LIABLE FOR ANY 37 * DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES 38 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; 39 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND 40 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 41 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 42 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 43 * 44 * Copyright (c) 2002, Business-Insight 45 * <a href="http://www.Business-Insight.com">Business-Insight</a> 46 * All rights reserved. 47 * 48 * \section tutorial First Tutorial 49 * You can follow a simple <a href="../../xmlParser.html">Tutorial</a> to know the basics... 50 * 51 * \section usage General usage: How to include the XMLParser library inside your project. 52 * 53 * The library is composed of two files: <a href="../../xmlParser.cpp">xmlParser.cpp</a> and 54 * <a href="../../xmlParser.h">xmlParser.h</a>. These are the ONLY 2 files that you need when 55 * using the library inside your own projects. 56 * 57 * All the functions of the library are documented inside the comments of the file 58 * <a href="../../xmlParser.h">xmlParser.h</a>. These comments can be transformed in 59 * full-fledged HTML documentation using the DOXYGEN software: simply type: "doxygen doxy.cfg" 60 * 61 * By default, the XMLParser library uses (char*) for string representation.To use the (wchar_t*) 62 * version of the library, you need to define the "_UNICODE" preprocessor definition variable 63 * (this is usually done inside your project definition file) (This is done automatically for you 64 * when using Visual Studio). 65 * 66 * \section example Advanced Tutorial and Many Examples of usage. 67 * 68 * Some very small introductory examples are described inside the Tutorial file 69 * <a href="../../xmlParser.html">xmlParser.html</a> 70 * 71 * Some additional small examples are also inside the file <a href="../../xmlTest.cpp">xmlTest.cpp</a> 72 * (for the "char*" version of the library) and inside the file 73 * <a href="../../xmlTestUnicode.cpp">xmlTestUnicode.cpp</a> (for the "wchar_t*" 74 * version of the library). If you have a question, please review these additionnal examples 75 * before sending an e-mail to the author. 76 * 77 * To build the examples: 78 * - linux/unix: type "make" 79 * - solaris: type "make -f makefile.solaris" 80 * - windows: Visual Studio: double-click on xmlParser.dsw 81 * (under Visual Studio .NET, the .dsp and .dsw files will be automatically converted to .vcproj and .sln files) 82 * 83 * In order to build the examples you need some additional files: 84 * - linux/unix: makefile 85 * - solaris: makefile.solaris 86 * - windows: Visual Studio: *.dsp, xmlParser.dsw and also xmlParser.lib and xmlParser.dll 87 * 88 * \section debugging Debugging with the XMLParser library 89 * 90 * \subsection debugwin Debugging under WINDOWS 91 * 92 * Inside Visual C++, the "debug versions" of the memory allocation functions are 93 * very slow: Do not forget to compile in "release mode" to get maximum speed. 94 * When I had to debug a software that was using the XMLParser Library, it was usually 95 * a nightmare because the library was sooOOOoooo slow in debug mode (because of the 96 * slow memory allocations in Debug mode). To solve this 97 * problem, during all the debugging session, I am now using a very fast DLL version of the 98 * XMLParser Library (the DLL is compiled in release mode). Using the DLL version of 99 * the XMLParser Library allows me to have lightening XML parsing speed even in debug! 100 * Other than that, the DLL version is useless: In the release version of my tool, 101 * I always use the normal, ".cpp"-based, XMLParser Library (I simply include the 102 * <a href="../../xmlParser.cpp">xmlParser.cpp</a> and 103 * <a href="../../xmlParser.h">xmlParser.h</a> files into the project). 104 * 105 * The file <a href="../../XMLNodeAutoexp.txt">XMLNodeAutoexp.txt</a> contains some 106 * "tweaks" that improve substancially the display of the content of the XMLNode objects 107 * inside the Visual Studio Debugger. Believe me, once you have seen inside the debugger 108 * the "smooth" display of the XMLNode objects, you cannot live without it anymore! 109 * 110 * \subsection debuglinux Debugging under LINUX/UNIX 111 * 112 * The speed of the debug version of the XMLParser library is tolerable so no extra 113 * work.has been done. 114 * 115 ****************************************************************************/ 116 117#ifndef __INCLUDE_XML_NODE__ 118#define __INCLUDE_XML_NODE__ 119 120#include <stdlib.h> 121 122#ifdef _UNICODE 123// If you comment the next "define" line then the library will never "switch to" _UNICODE (wchar_t*) mode (16/32 bits per characters). 124// This is useful when you get error messages like: 125// 'XMLNode::openFileHelper' : cannot convert parameter 2 from 'const char [5]' to 'const wchar_t *' 126// The _XMLWIDECHAR preprocessor variable force the XMLParser library into either utf16/32-mode (the proprocessor variable 127// must be defined) or utf8-mode(the pre-processor variable must be undefined). 128#define _XMLWIDECHAR 129#endif 130 131#if defined(WIN32) || defined(UNDER_CE) || defined(_WIN32) || defined(WIN64) || defined(__BORLANDC__) 132// comment the next line if you are under windows and the compiler is not Microsoft Visual Studio (6.0 or .NET) or Borland 133#define _XMLWINDOWS 134#endif 135 136#ifdef XMLDLLENTRY 137#undef XMLDLLENTRY 138#endif 139#ifdef _USE_XMLPARSER_DLL 140#ifdef _DLL_EXPORTS_ 141#define XMLDLLENTRY __declspec(dllexport) 142#else 143#define XMLDLLENTRY __declspec(dllimport) 144#endif 145#else 146#define XMLDLLENTRY 147#endif 148 149// uncomment the next line if you want no support for wchar_t* (no need for the <wchar.h> or <tchar.h> libraries anymore to compile) 150//#define XML_NO_WIDE_CHAR 151 152#ifdef XML_NO_WIDE_CHAR 153#undef _XMLWINDOWS 154#undef _XMLWIDECHAR 155#endif 156 157#ifdef _XMLWINDOWS 158#include <tchar.h> 159#else 160#define XMLDLLENTRY 161#ifndef XML_NO_WIDE_CHAR 162#include <wchar.h> // to have 'wcsrtombs' for ANSI version 163 // to have 'mbsrtowcs' for WIDECHAR version 164#endif 165#endif 166 167// Some common types for char set portable code 168#ifdef _XMLWIDECHAR 169 #define _CXML(c) L ## c 170 #define XMLCSTR const wchar_t * 171 #define XMLSTR wchar_t * 172 #define XMLCHAR wchar_t 173#else 174 #define _CXML(c) c 175 #define XMLCSTR const char * 176 #define XMLSTR char * 177 #define XMLCHAR char 178#endif 179#ifndef FALSE 180 #define FALSE 0 181#endif /* FALSE */ 182#ifndef TRUE 183 #define TRUE 1 184#endif /* TRUE */ 185 186 187/// Enumeration for XML parse errors. 188typedef enum XMLError 189{ 190 eXMLErrorNone = 0, 191 eXMLErrorMissingEndTag, 192 eXMLErrorNoXMLTagFound, 193 eXMLErrorEmpty, 194 eXMLErrorMissingTagName, 195 eXMLErrorMissingEndTagName, 196 eXMLErrorUnmatchedEndTag, 197 eXMLErrorUnmatchedEndClearTag, 198 eXMLErrorUnexpectedToken, 199 eXMLErrorNoElements, 200 eXMLErrorFileNotFound, 201 eXMLErrorFirstTagNotFound, 202 eXMLErrorUnknownCharacterEntity, 203 eXMLErrorCharacterCodeAbove255, 204 eXMLErrorCharConversionError, 205 eXMLErrorCannotOpenWriteFile, 206 eXMLErrorCannotWriteFile, 207 208 eXMLErrorBase64DataSizeIsNotMultipleOf4, 209 eXMLErrorBase64DecodeIllegalCharacter, 210 eXMLErrorBase64DecodeTruncatedData, 211 eXMLErrorBase64DecodeBufferTooSmall 212} XMLError; 213 214 215/// Enumeration used to manage type of data. Use in conjunction with structure XMLNodeContents 216typedef enum XMLElementType 217{ 218 eNodeChild=0, 219 eNodeAttribute=1, 220 eNodeText=2, 221 eNodeClear=3, 222 eNodeNULL=4 223} XMLElementType; 224 225/// Structure used to obtain error details if the parse fails. 226typedef struct XMLResults 227{ 228 enum XMLError error; 229 int nLine,nColumn; 230} XMLResults; 231 232/// Structure for XML clear (unformatted) node (usually comments) 233typedef struct XMLClear { 234 XMLCSTR lpszValue; XMLCSTR lpszOpenTag; XMLCSTR lpszCloseTag; 235} XMLClear; 236 237/// Structure for XML attribute. 238typedef struct XMLAttribute { 239 XMLCSTR lpszName; XMLCSTR lpszValue; 240} XMLAttribute; 241 242/// XMLElementPosition are not interchangeable with simple indexes 243typedef int XMLElementPosition; 244 245struct XMLNodeContents; 246 247/** @defgroup XMLParserGeneral The XML parser */ 248 249/// Main Class representing a XML node 250/** 251 * All operations are performed using this class. 252 * \note The constructors of the XMLNode class are protected, so use instead one of these four methods to get your first instance of XMLNode: 253 * <ul> 254 * <li> XMLNode::parseString </li> 255 * <li> XMLNode::parseFile </li> 256 * <li> XMLNode::openFileHelper </li> 257 * <li> XMLNode::createXMLTopNode (or XMLNode::createXMLTopNode_WOSD)</li> 258 * </ul> */ 259typedef struct XMLDLLENTRY XMLNode 260{ 261 private: 262 263 struct XMLNodeDataTag; 264 265 /// Constructors are protected, so use instead one of: XMLNode::parseString, XMLNode::parseFile, XMLNode::openFileHelper, XMLNode::createXMLTopNode 266 XMLNode(struct XMLNodeDataTag *pParent, XMLSTR lpszName, char isDeclaration); 267 /// Constructors are protected, so use instead one of: XMLNode::parseString, XMLNode::parseFile, XMLNode::openFileHelper, XMLNode::createXMLTopNode 268 XMLNode(struct XMLNodeDataTag *p); 269 270 public: 271 static XMLCSTR getVersion();///< Return the XMLParser library version number 272 273 /** @defgroup conversions Parsing XML files/strings to an XMLNode structure and Rendering XMLNode's to files/string. 274 * @ingroup XMLParserGeneral 275 * @{ */ 276 277 /// Parse an XML string and return the root of a XMLNode tree representing the string. 278 static XMLNode parseString (XMLCSTR lpXMLString, XMLCSTR tag=NULL, XMLResults *pResults=NULL); 279 /**< The "parseString" function parse an XML string and return the root of a XMLNode tree. The "opposite" of this function is 280 * the function "createXMLString" that re-creates an XML string from an XMLNode tree. If the XML document is corrupted, the 281 * "parseString" method will initialize the "pResults" variable with some information that can be used to trace the error. 282 * If you still want to parse the file, you can use the APPROXIMATE_PARSING option as explained inside the note at the 283 * beginning of the "xmlParser.cpp" file. 284 * 285 * @param lpXMLString the XML string to parse 286 * @param tag the name of the first tag inside the XML file. If the tag parameter is omitted, this function returns a node that represents the head of the xml document including the declaration term (<? ... ?>). 287 * @param pResults a pointer to a XMLResults variable that will contain some information that can be used to trace the XML parsing error. You can have a user-friendly explanation of the parsing error with the "getError" function. 288 */ 289 290 /// Parse an XML file and return the root of a XMLNode tree representing the file. 291 static XMLNode parseFile (XMLCSTR filename, XMLCSTR tag=NULL, XMLResults *pResults=NULL); 292 /**< The "parseFile" function parse an XML file and return the root of a XMLNode tree. The "opposite" of this function is 293 * the function "writeToFile" that re-creates an XML file from an XMLNode tree. If the XML document is corrupted, the 294 * "parseFile" method will initialize the "pResults" variable with some information that can be used to trace the error. 295 * If you still want to parse the file, you can use the APPROXIMATE_PARSING option as explained inside the note at the 296 * beginning of the "xmlParser.cpp" file. 297 * 298 * @param filename the path to the XML file to parse 299 * @param tag the name of the first tag inside the XML file. If the tag parameter is omitted, this function returns a node that represents the head of the xml document including the declaration term (<? ... ?>). 300 * @param pResults a pointer to a XMLResults variable that will contain some information that can be used to trace the XML parsing error. You can have a user-friendly explanation of the parsing error with the "getError" function. 301 */ 302 303 /// Parse an XML file and return the root of a XMLNode tree representing the file. A very crude error checking is made. An attempt to guess the Char Encoding used in the file is made. 304 static XMLNode openFileHelper(XMLCSTR filename, XMLCSTR tag=NULL); 305 /**< The "openFileHelper" function reports to the screen all the warnings and errors that occurred during parsing of the XML file. 306 * This function also tries to guess char Encoding (UTF-8, ASCII or SHIT-JIS) based on the first 200 bytes of the file. Since each 307 * application has its own way to report and deal with errors, you should rather use the "parseFile" function to parse XML files 308 * and program yourself thereafter an "error reporting" tailored for your needs (instead of using the very crude "error reporting" 309 * mechanism included inside the "openFileHelper" function). 310 * 311 * If the XML document is corrupted, the "openFileHelper" method will: 312 * - display an error message on the console (or inside a messageBox for windows). 313 * - stop execution (exit). 314 * 315 * I strongly suggest that you write your own "openFileHelper" method tailored to your needs. If you still want to parse 316 * the file, you can use the APPROXIMATE_PARSING option as explained inside the note at the beginning of the "xmlParser.cpp" file. 317 * 318 * @param filename the path of the XML file to parse. 319 * @param tag the name of the first tag inside the XML file. If the tag parameter is omitted, this function returns a node that represents the head of the xml document including the declaration term (<? ... ?>). 320 */ 321 322 static XMLCSTR getError(XMLError error); ///< this gives you a user-friendly explanation of the parsing error 323 324 /// Create an XML string starting from the current XMLNode. 325 XMLSTR createXMLString(int nFormat=1, int *pnSize=NULL) const; 326 /**< The returned string should be free'd using the "freeXMLString" function. 327 * 328 * If nFormat==0, no formatting is required otherwise this returns an user friendly XML string from a given element 329 * with appropriate white spaces and carriage returns. if pnSize is given it returns the size in character of the string. */ 330 331 /// Save the content of an xmlNode inside a file 332 XMLError writeToFile(XMLCSTR filename, 333 const char *encoding=NULL, 334 char nFormat=1) const; 335 /**< If nFormat==0, no formatting is required otherwise this returns an user friendly XML string from a given element with appropriate white spaces and carriage returns. 336 * If the global parameter "characterEncoding==encoding_UTF8", then the "encoding" parameter is ignored and always set to "utf-8". 337 * If the global parameter "characterEncoding==encoding_ShiftJIS", then the "encoding" parameter is ignored and always set to "SHIFT-JIS". 338 * If "_XMLWIDECHAR=1", then the "encoding" parameter is ignored and always set to "utf-16". 339 * If no "encoding" parameter is given the "ISO-8859-1" encoding is used. */ 340 /** @} */ 341 342 /** @defgroup navigate Navigate the XMLNode structure 343 * @ingroup XMLParserGeneral 344 * @{ */ 345 XMLCSTR getName() const; ///< name of the node 346 XMLCSTR getText(int i=0) const; ///< return ith text field 347 int nText() const; ///< nbr of text field 348 XMLNode getParentNode() const; ///< return the parent node 349 XMLNode getChildNode(int i=0) const; ///< return ith child node 350 XMLNode getChildNode(XMLCSTR name, int i) const; ///< return ith child node with specific name (return an empty node if failing). If i==-1, this returns the last XMLNode with the given name. 351 XMLNode getChildNode(XMLCSTR name, int *i=NULL) const; ///< return next child node with specific name (return an empty node if failing) 352 XMLNode getChildNodeWithAttribute(XMLCSTR tagName, 353 XMLCSTR attributeName, 354 XMLCSTR attributeValue=NULL, 355 int *i=NULL) const; ///< return child node with specific name/attribute (return an empty node if failing) 356 XMLNode getChildNodeByPath(XMLCSTR path, char createNodeIfMissing=0, XMLCHAR sep='/'); 357 ///< return the first child node with specific path 358 XMLNode getChildNodeByPathNonConst(XMLSTR path, char createNodeIfMissing=0, XMLCHAR sep='/'); 359 ///< return the first child node with specific path. 360 361 int nChildNode(XMLCSTR name) const; ///< return the number of child node with specific name 362 int nChildNode() const; ///< nbr of child node 363 XMLAttribute getAttribute(int i=0) const; ///< return ith attribute 364 XMLCSTR getAttributeName(int i=0) const; ///< return ith attribute name 365 XMLCSTR getAttributeValue(int i=0) const; ///< return ith attribute value 366 char isAttributeSet(XMLCSTR name) const; ///< test if an attribute with a specific name is given 367 XMLCSTR getAttribute(XMLCSTR name, int i) const; ///< return ith attribute content with specific name (return a NULL if failing) 368 XMLCSTR getAttribute(XMLCSTR name, int *i=NULL) const; ///< return next attribute content with specific name (return a NULL if failing) 369 int nAttribute() const; ///< nbr of attribute 370 XMLClear getClear(int i=0) const; ///< return ith clear field (comments) 371 int nClear() const; ///< nbr of clear field 372 XMLNodeContents enumContents(XMLElementPosition i) const; ///< enumerate all the different contents (attribute,child,text, clear) of the current XMLNode. The order is reflecting the order of the original file/string. NOTE: 0 <= i < nElement(); 373 int nElement() const; ///< nbr of different contents for current node 374 char isEmpty() const; ///< is this node Empty? 375 char isDeclaration() const; ///< is this node a declaration <? .... ?> 376 XMLNode deepCopy() const; ///< deep copy (duplicate/clone) a XMLNode 377 static XMLNode emptyNode(); ///< return XMLNode::emptyXMLNode; 378 /** @} */ 379 380 ~XMLNode(); 381 XMLNode(const XMLNode &A); ///< to allow shallow/fast copy: 382 XMLNode& operator=( const XMLNode& A ); ///< to allow shallow/fast copy: 383 384 XMLNode(): d(NULL){}; 385 static XMLNode emptyXMLNode; 386 static XMLClear emptyXMLClear; 387 static XMLAttribute emptyXMLAttribute; 388 389 /** @defgroup xmlModify Create or Update the XMLNode structure 390 * @ingroup XMLParserGeneral 391 * The functions in this group allows you to create from scratch (or update) a XMLNode structure. Start by creating your top 392 * node with the "createXMLTopNode" function and then add new nodes with the "addChild" function. The parameter 'pos' gives 393 * the position where the childNode, the text or the XMLClearTag will be inserted. The default value (pos=-1) inserts at the 394 * end. The value (pos=0) insert at the beginning (Insertion at the beginning is slower than at the end). <br> 395 * 396 * REMARK: 0 <= pos < nChild()+nText()+nClear() <br> 397 */ 398 399 /** @defgroup creation Creating from scratch a XMLNode structure 400 * @ingroup xmlModify 401 * @{ */ 402 static XMLNode createXMLTopNode(XMLCSTR lpszName, char isDeclaration=FALSE); ///< Create the top node of an XMLNode structure 403 XMLNode addChild(XMLCSTR lpszName, char isDeclaration=FALSE, XMLElementPosition pos=-1); ///< Add a new child node 404 XMLNode addChild(XMLNode nodeToAdd, XMLElementPosition pos=-1); ///< If the "nodeToAdd" has some parents, it will be detached from it's parents before being attached to the current XMLNode 405 XMLAttribute *addAttribute(XMLCSTR lpszName, XMLCSTR lpszValuev); ///< Add a new attribute 406 XMLCSTR addText(XMLCSTR lpszValue, XMLElementPosition pos=-1); ///< Add a new text content 407 XMLClear *addClear(XMLCSTR lpszValue, XMLCSTR lpszOpen=NULL, XMLCSTR lpszClose=NULL, XMLElementPosition pos=-1); 408 /**< Add a new clear tag 409 * @param lpszOpen default value "<![CDATA[" 410 * @param lpszClose default value "]]>" 411 */ 412 /** @} */ 413 414 /** @defgroup xmlUpdate Updating Nodes 415 * @ingroup xmlModify 416 * Some update functions: 417 * @{ 418 */ 419 XMLCSTR updateName(XMLCSTR lpszName); ///< change node's name 420 XMLAttribute *updateAttribute(XMLAttribute *newAttribute, XMLAttribute *oldAttribute); ///< if the attribute to update is missing, a new one will be added 421 XMLAttribute *updateAttribute(XMLCSTR lpszNewValue, XMLCSTR lpszNewName=NULL,int i=0); ///< if the attribute to update is missing, a new one will be added 422 XMLAttribute *updateAttribute(XMLCSTR lpszNewValue, XMLCSTR lpszNewName,XMLCSTR lpszOldName);///< set lpszNewName=NULL if you don't want to change the name of the attribute if the attribute to update is missing, a new one will be added 423 XMLCSTR updateText(XMLCSTR lpszNewValue, int i=0); ///< if the text to update is missing, a new one will be added 424 XMLCSTR updateText(XMLCSTR lpszNewValue, XMLCSTR lpszOldValue); ///< if the text to update is missing, a new one will be added 425 XMLClear *updateClear(XMLCSTR lpszNewContent, int i=0); ///< if the clearTag to update is missing, a new one will be added 426 XMLClear *updateClear(XMLClear *newP,XMLClear *oldP); ///< if the clearTag to update is missing, a new one will be added 427 XMLClear *updateClear(XMLCSTR lpszNewValue, XMLCSTR lpszOldValue); ///< if the clearTag to update is missing, a new one will be added 428 /** @} */ 429 430 /** @defgroup xmlDelete Deleting Nodes or Attributes 431 * @ingroup xmlModify 432 * Some deletion functions: 433 * @{ 434 */ 435 /// The "deleteNodeContent" function forces the deletion of the content of this XMLNode and the subtree. 436 void deleteNodeContent(); 437 /**< \note The XMLNode instances that are referring to the part of the subtree that has been deleted CANNOT be used anymore!!. Unexpected results will occur if you continue using them. */ 438 void deleteAttribute(int i=0); ///< Delete the ith attribute of the current XMLNode 439 void deleteAttribute(XMLCSTR lpszName); ///< Delete the attribute with the given name (the "strcmp" function is used to find the right attribute) 440 void deleteAttribute(XMLAttribute *anAttribute); ///< Delete the attribute with the name "anAttribute->lpszName" (the "strcmp" function is used to find the right attribute) 441 void deleteText(int i=0); ///< Delete the Ith text content of the current XMLNode 442 void deleteText(XMLCSTR lpszValue); ///< Delete the text content "lpszValue" inside the current XMLNode (direct "pointer-to-pointer" comparison is used to find the right text) 443 void deleteClear(int i=0); ///< Delete the Ith clear tag inside the current XMLNode 444 void deleteClear(XMLCSTR lpszValue); ///< Delete the clear tag "lpszValue" inside the current XMLNode (direct "pointer-to-pointer" comparison is used to find the clear tag) 445 void deleteClear(XMLClear *p); ///< Delete the clear tag "p" inside the current XMLNode (direct "pointer-to-pointer" comparison on the lpszName of the clear tag is used to find the clear tag) 446 /** @} */ 447 448 /** @defgroup xmlWOSD ???_WOSD functions. 449 * @ingroup xmlModify 450 * The strings given as parameters for the "add" and "update" methods that have a name with 451 * the postfix "_WOSD" (that means "WithOut String Duplication")(for example "addText_WOSD") 452 * will be free'd by the XMLNode class. For example, it means that this is incorrect: 453 * \code 454 * xNode.addText_WOSD("foo"); 455 * xNode.updateAttribute_WOSD("#newcolor" ,NULL,"color"); 456 * \endcode 457 * In opposition, this is correct: 458 * \code 459 * xNode.addText("foo"); 460 * xNode.addText_WOSD(stringDup("foo")); 461 * xNode.updateAttribute("#newcolor" ,NULL,"color"); 462 * xNode.updateAttribute_WOSD(stringDup("#newcolor"),NULL,"color"); 463 * \endcode 464 * Typically, you will never do: 465 * \code 466 * char *b=(char*)malloc(...); 467 * xNode.addText(b); 468 * free(b); 469 * \endcode 470 * ... but rather: 471 * \code 472 * char *b=(char*)malloc(...); 473 * xNode.addText_WOSD(b); 474 * \endcode 475 * ('free(b)' is performed by the XMLNode class) 476 * @{ */ 477 static XMLNode createXMLTopNode_WOSD(XMLSTR lpszName, char isDeclaration=FALSE); ///< Create the top node of an XMLNode structure 478 XMLNode addChild_WOSD(XMLSTR lpszName, char isDeclaration=FALSE, XMLElementPosition pos=-1); ///< Add a new child node 479 XMLAttribute *addAttribute_WOSD(XMLSTR lpszName, XMLSTR lpszValue); ///< Add a new attribute 480 XMLCSTR addText_WOSD(XMLSTR lpszValue, XMLElementPosition pos=-1); ///< Add a new text content 481 XMLClear *addClear_WOSD(XMLSTR lpszValue, XMLCSTR lpszOpen=NULL, XMLCSTR lpszClose=NULL, XMLElementPosition pos=-1); ///< Add a new clear Tag 482 483 XMLCSTR updateName_WOSD(XMLSTR lpszName); ///< change node's name 484 XMLAttribute *updateAttribute_WOSD(XMLAttribute *newAttribute, XMLAttribute *oldAttribute); ///< if the attribute to update is missing, a new one will be added 485 XMLAttribute *updateAttribute_WOSD(XMLSTR lpszNewValue, XMLSTR lpszNewName=NULL,int i=0); ///< if the attribute to update is missing, a new one will be added 486 XMLAttribute *updateAttribute_WOSD(XMLSTR lpszNewValue, XMLSTR lpszNewName,XMLCSTR lpszOldName); ///< set lpszNewName=NULL if you don't want to change the name of the attribute if the attribute to update is missing, a new one will be added 487 XMLCSTR updateText_WOSD(XMLSTR lpszNewValue, int i=0); ///< if the text to update is missing, a new one will be added 488 XMLCSTR updateText_WOSD(XMLSTR lpszNewValue, XMLCSTR lpszOldValue); ///< if the text to update is missing, a new one will be added 489 XMLClear *updateClear_WOSD(XMLSTR lpszNewContent, int i=0); ///< if the clearTag to update is missing, a new one will be added 490 XMLClear *updateClear_WOSD(XMLClear *newP,XMLClear *oldP); ///< if the clearTag to update is missing, a new one will be added 491 XMLClear *updateClear_WOSD(XMLSTR lpszNewValue, XMLCSTR lpszOldValue); ///< if the clearTag to update is missing, a new one will be added 492 /** @} */ 493 494 /** @defgroup xmlPosition Position helper functions (use in conjunction with the update&add functions 495 * @ingroup xmlModify 496 * These are some useful functions when you want to insert a childNode, a text or a XMLClearTag in the 497 * middle (at a specified position) of a XMLNode tree already constructed. The value returned by these 498 * methods is to be used as last parameter (parameter 'pos') of addChild, addText or addClear. 499 * @{ */ 500 XMLElementPosition positionOfText(int i=0) const; 501 XMLElementPosition positionOfText(XMLCSTR lpszValue) const; 502 XMLElementPosition positionOfClear(int i=0) const; 503 XMLElementPosition positionOfClear(XMLCSTR lpszValue) const; 504 XMLElementPosition positionOfClear(XMLClear *a) const; 505 XMLElementPosition positionOfChildNode(int i=0) const; 506 XMLElementPosition positionOfChildNode(XMLNode x) const; 507 XMLElementPosition positionOfChildNode(XMLCSTR name, int i=0) const; ///< return the position of the ith childNode with the specified name if (name==NULL) return the position of the ith childNode 508 /** @} */ 509 510 /// Enumeration for XML character encoding. 511 typedef enum XMLCharEncoding 512 { 513 char_encoding_error=0, 514 char_encoding_UTF8=1, 515 char_encoding_legacy=2, 516 char_encoding_ShiftJIS=3, 517 char_encoding_GB2312=4, 518 char_encoding_Big5=5, 519 char_encoding_GBK=6 // this is actually the same as Big5 520 } XMLCharEncoding; 521 522 /** \addtogroup conversions 523 * @{ */ 524 525 /// Sets the global options for the conversions 526 static char setGlobalOptions(XMLCharEncoding characterEncoding=XMLNode::char_encoding_UTF8, char guessWideCharChars=1, 527 char dropWhiteSpace=1, char removeCommentsInMiddleOfText=1); 528 /**< The "setGlobalOptions" function allows you to change four global parameters that affect string & file 529 * parsing. First of all, you most-probably will never have to change these 3 global parameters. 530 * 531 * @param guessWideCharChars If "guessWideCharChars"=1 and if this library is compiled in WideChar mode, then the 532 * XMLNode::parseFile and XMLNode::openFileHelper functions will test if the file contains ASCII 533 * characters. If this is the case, then the file will be loaded and converted in memory to 534 * WideChar before being parsed. If 0, no conversion will be performed. 535 * 536 * @param guessWideCharChars If "guessWideCharChars"=1 and if this library is compiled in ASCII/UTF8/char* mode, then the 537 * XMLNode::parseFile and XMLNode::openFileHelper functions will test if the file contains WideChar 538 * characters. If this is the case, then the file will be loaded and converted in memory to 539 * ASCII/UTF8/char* before being parsed. If 0, no conversion will be performed. 540 * 541 * @param characterEncoding This parameter is only meaningful when compiling in char* mode (multibyte character mode). 542 * In wchar_t* (wide char mode), this parameter is ignored. This parameter should be one of the 543 * three currently recognized encodings: XMLNode::encoding_UTF8, XMLNode::encoding_ascii, 544 * XMLNode::encoding_ShiftJIS. 545 * 546 * @param dropWhiteSpace In most situations, text fields containing only white spaces (and carriage returns) 547 * are useless. Even more, these "empty" text fields are annoying because they increase the 548 * complexity of the user's code for parsing. So, 99% of the time, it's better to drop 549 * the "empty" text fields. However The XML specification indicates that no white spaces 550 * should be lost when parsing the file. So to be perfectly XML-compliant, you should set 551 * dropWhiteSpace=0. A note of caution: if you set "dropWhiteSpace=0", the parser will be 552 * slower and your code will be more complex. 553 * 554 * @param removeCommentsInMiddleOfText To explain this parameter, let's consider this code: 555 * \code 556 * XMLNode x=XMLNode::parseString("<a>foo<!-- hello -->bar<!DOCTYPE world >chu</a>","a"); 557 * \endcode 558 * If removeCommentsInMiddleOfText=0, then we will have: 559 * \code 560 * x.getText(0) -> "foo" 561 * x.getText(1) -> "bar" 562 * x.getText(2) -> "chu" 563 * x.getClear(0) --> "<!-- hello -->" 564 * x.getClear(1) --> "<!DOCTYPE world >" 565 * \endcode 566 * If removeCommentsInMiddleOfText=1, then we will have: 567 * \code 568 * x.getText(0) -> "foobar" 569 * x.getText(1) -> "chu" 570 * x.getClear(0) --> "<!DOCTYPE world >" 571 * \endcode 572 * 573 * \return "0" when there are no errors. If you try to set an unrecognized encoding then the return value will be "1" to signal an error. 574 * 575 * \note Sometime, it's useful to set "guessWideCharChars=0" to disable any conversion 576 * because the test to detect the file-type (ASCII/UTF8/char* or WideChar) may fail (rarely). */ 577 578 /// Guess the character encoding of the string (ascii, utf8 or shift-JIS) 579 static XMLCharEncoding guessCharEncoding(void *buffer, int bufLen, char useXMLEncodingAttribute=1); 580 /**< The "guessCharEncoding" function try to guess the character encoding. You most-probably will never 581 * have to use this function. It then returns the appropriate value of the global parameter 582 * "characterEncoding" described in the XMLNode::setGlobalOptions. The guess is based on the content of a buffer of length 583 * "bufLen" bytes that contains the first bytes (minimum 25 bytes; 200 bytes is a good value) of the 584 * file to be parsed. The XMLNode::openFileHelper function is using this function to automatically compute 585 * the value of the "characterEncoding" global parameter. There are several heuristics used to do the 586 * guess. One of the heuristic is based on the "encoding" attribute. The original XML specifications 587 * forbids to use this attribute to do the guess but you can still use it if you set 588 * "useXMLEncodingAttribute" to 1 (this is the default behavior and the behavior of most parsers). 589 * If an inconsistency in the encoding is detected, then the return value is "0". */ 590 /** @} */ 591 592 private: 593 // these are functions and structures used internally by the XMLNode class (don't bother about them): 594 595 typedef struct XMLNodeDataTag // to allow shallow copy and "intelligent/smart" pointers (automatic delete): 596 { 597 XMLCSTR lpszName; // Element name (=NULL if root) 598 int nChild, // Number of child nodes 599 nText, // Number of text fields 600 nClear, // Number of Clear fields (comments) 601 nAttribute; // Number of attributes 602 char isDeclaration; // Whether node is an XML declaration - '<?xml ?>' 603 struct XMLNodeDataTag *pParent; // Pointer to parent element (=NULL if root) 604 XMLNode *pChild; // Array of child nodes 605 XMLCSTR *pText; // Array of text fields 606 XMLClear *pClear; // Array of clear fields 607 XMLAttribute *pAttribute; // Array of attributes 608 int *pOrder; // order of the child_nodes,text_fields,clear_fields 609 int ref_count; // for garbage collection (smart pointers) 610 } XMLNodeData; 611 XMLNodeData *d; 612 613 char parseClearTag(void *px, void *pa); 614 char maybeAddTxT(void *pa, XMLCSTR tokenPStr); 615 int ParseXMLElement(void *pXML); 616 void *addToOrder(int memInc, int *_pos, int nc, void *p, int size, XMLElementType xtype); 617 int indexText(XMLCSTR lpszValue) const; 618 int indexClear(XMLCSTR lpszValue) const; 619 XMLNode addChild_priv(int,XMLSTR,char,int); 620 XMLAttribute *addAttribute_priv(int,XMLSTR,XMLSTR); 621 XMLCSTR addText_priv(int,XMLSTR,int); 622 XMLClear *addClear_priv(int,XMLSTR,XMLCSTR,XMLCSTR,int); 623 void emptyTheNode(char force); 624 static inline XMLElementPosition findPosition(XMLNodeData *d, int index, XMLElementType xtype); 625 static int CreateXMLStringR(XMLNodeData *pEntry, XMLSTR lpszMarker, int nFormat); 626 static int removeOrderElement(XMLNodeData *d, XMLElementType t, int index); 627 static void exactMemory(XMLNodeData *d); 628 static int detachFromParent(XMLNodeData *d); 629} XMLNode; 630 631/// This structure is given by the function XMLNode::enumContents. 632typedef struct XMLNodeContents 633{ 634 /// This dictates what's the content of the XMLNodeContent 635 enum XMLElementType etype; 636 /**< should be an union to access the appropriate data. Compiler does not allow union of object with constructor... too bad. */ 637 XMLNode child; 638 XMLAttribute attrib; 639 XMLCSTR text; 640 XMLClear clear; 641 642} XMLNodeContents; 643 644/** @defgroup StringAlloc String Allocation/Free functions 645 * @ingroup xmlModify 646 * @{ */ 647/// Duplicate (copy in a new allocated buffer) the source string. 648XMLDLLENTRY XMLSTR stringDup(XMLCSTR source, int cbData=-1); 649/**< This is 650 * a very handy function when used with all the "XMLNode::*_WOSD" functions (\link xmlWOSD \endlink). 651 * @param cbData If !=0 then cbData is the number of chars to duplicate. New strings allocated with 652 * this function should be free'd using the "freeXMLString" function. */ 653 654/// to free the string allocated inside the "stringDup" function or the "createXMLString" function. 655XMLDLLENTRY void freeXMLString(XMLSTR t); // {free(t);} 656/** @} */ 657 658/** @defgroup atoX ato? like functions 659 * @ingroup XMLParserGeneral 660 * The "xmlto?" functions are equivalents to the atoi, atol, atof functions. 661 * The only difference is: If the variable "xmlString" is NULL, than the return value 662 * is "defautValue". These 6 functions are only here as "convenience" functions for the 663 * user (they are not used inside the XMLparser). If you don't need them, you can 664 * delete them without any trouble. 665 * 666 * @{ */ 667XMLDLLENTRY char xmltob(XMLCSTR xmlString,char defautValue=0); 668XMLDLLENTRY int xmltoi(XMLCSTR xmlString,int defautValue=0); 669XMLDLLENTRY long xmltol(XMLCSTR xmlString,long defautValue=0); 670XMLDLLENTRY double xmltof(XMLCSTR xmlString,double defautValue=.0); 671XMLDLLENTRY XMLCSTR xmltoa(XMLCSTR xmlString,XMLCSTR defautValue=_CXML("")); 672XMLDLLENTRY XMLCHAR xmltoc(XMLCSTR xmlString,XMLCHAR defautValue=_CXML('\0')); 673/** @} */ 674 675/** @defgroup ToXMLStringTool Helper class to create XML files using "printf", "fprintf", "cout",... functions. 676 * @ingroup XMLParserGeneral 677 * @{ */ 678/// Helper class to create XML files using "printf", "fprintf", "cout",... functions. 679/** The ToXMLStringTool class helps you creating XML files using "printf", "fprintf", "cout",... functions. 680 * The "ToXMLStringTool" class is processing strings so that all the characters 681 * &,",',<,> are replaced by their XML equivalent: 682 * \verbatim &, ", ', <, > \endverbatim 683 * Using the "ToXMLStringTool class" and the "fprintf function" is THE most efficient 684 * way to produce VERY large XML documents VERY fast. 685 * \note If you are creating from scratch an XML file using the provided XMLNode class 686 * you must not use the "ToXMLStringTool" class (because the "XMLNode" class does the 687 * processing job for you during rendering).*/ 688typedef struct XMLDLLENTRY ToXMLStringTool 689{ 690public: 691 ToXMLStringTool(): buf(NULL),buflen(0){} 692 ~ToXMLStringTool(); 693 void freeBuffer();///<call this function when you have finished using this object to release memory used by the internal buffer. 694 695 XMLSTR toXML(XMLCSTR source);///< returns a pointer to an internal buffer that contains a XML-encoded string based on the "source" parameter. 696 697 /** The "toXMLUnSafe" function is deprecated because there is a possibility of 698 * "destination-buffer-overflow". It converts the string 699 * "source" to the string "dest". */ 700 static XMLSTR toXMLUnSafe(XMLSTR dest,XMLCSTR source); ///< deprecated: use "toXML" instead 701 static int lengthXMLString(XMLCSTR source); ///< deprecated: use "toXML" instead 702 703private: 704 XMLSTR buf; 705 int buflen; 706} ToXMLStringTool; 707/** @} */ 708 709/** @defgroup XMLParserBase64Tool Helper class to include binary data inside XML strings using "Base64 encoding". 710 * @ingroup XMLParserGeneral 711 * @{ */ 712/// Helper class to include binary data inside XML strings using "Base64 encoding". 713/** The "XMLParserBase64Tool" class allows you to include any binary data (images, sounds,...) 714 * into an XML document using "Base64 encoding". This class is completely 715 * separated from the rest of the xmlParser library and can be removed without any problem. 716 * To include some binary data into an XML file, you must convert the binary data into 717 * standard text (using "encode"). To retrieve the original binary data from the 718 * b64-encoded text included inside the XML file, use "decode". Alternatively, these 719 * functions can also be used to "encrypt/decrypt" some critical data contained inside 720 * the XML (it's not a strong encryption at all, but sometimes it can be useful). */ 721typedef struct XMLDLLENTRY XMLParserBase64Tool 722{ 723public: 724 XMLParserBase64Tool(): buf(NULL),buflen(0){} 725 ~XMLParserBase64Tool(); 726 void freeBuffer();///< Call this function when you have finished using this object to release memory used by the internal buffer. 727 728 /** 729 * @param formatted If "formatted"=true, some space will be reserved for a carriage-return every 72 chars. */ 730 static int encodeLength(int inBufLen, char formatted=0); ///< return the length of the base64 string that encodes a data buffer of size inBufLen bytes. 731 732 /** 733 * The "base64Encode" function returns a string containing the base64 encoding of "inByteLen" bytes 734 * from "inByteBuf". If "formatted" parameter is true, then there will be a carriage-return every 72 chars. 735 * The string will be free'd when the XMLParserBase64Tool object is deleted. 736 * All returned strings are sharing the same memory space. */ 737 XMLSTR encode(unsigned char *inByteBuf, unsigned int inByteLen, char formatted=0); ///< returns a pointer to an internal buffer containing the base64 string containing the binary data encoded from "inByteBuf" 738 739 /// returns the number of bytes which will be decoded from "inString". 740 static unsigned int decodeSize(XMLCSTR inString, XMLError *xe=NULL); 741 742 /** 743 * The "decode" function returns a pointer to a buffer containing the binary data decoded from "inString" 744 * The output buffer will be free'd when the XMLParserBase64Tool object is deleted. 745 * All output buffer are sharing the same memory space. 746 * @param inString If "instring" is malformed, NULL will be returned */ 747 unsigned char* decode(XMLCSTR inString, int *outByteLen=NULL, XMLError *xe=NULL); ///< returns a pointer to an internal buffer containing the binary data decoded from "inString" 748 749 /** 750 * decodes data from "inString" to "outByteBuf". You need to provide the size (in byte) of "outByteBuf" 751 * in "inMaxByteOutBuflen". If "outByteBuf" is not large enough or if data is malformed, then "FALSE" 752 * will be returned; otherwise "TRUE". */ 753 static unsigned char decode(XMLCSTR inString, unsigned char *outByteBuf, int inMaxByteOutBuflen, XMLError *xe=NULL); ///< deprecated. 754 755private: 756 void *buf; 757 int buflen; 758 void alloc(int newsize); 759}XMLParserBase64Tool; 760/** @} */ 761 762#undef XMLDLLENTRY 763 764#endif 765