[188] | 1 | /*
|
---|
| 2 | * The Apache Software License, Version 1.1
|
---|
| 3 | *
|
---|
| 4 | * Copyright (c) 2002 The Apache Software Foundation. All rights
|
---|
| 5 | * reserved.
|
---|
| 6 | *
|
---|
| 7 | * Redistribution and use in source and binary forms, with or without
|
---|
| 8 | * modification, are permitted provided that the following conditions
|
---|
| 9 | * are met:
|
---|
| 10 | *
|
---|
| 11 | * 1. Redistributions of source code must retain the above copyright
|
---|
| 12 | * notice, this list of conditions and the following disclaimer.
|
---|
| 13 | *
|
---|
| 14 | * 2. Redistributions in binary form must reproduce the above copyright
|
---|
| 15 | * notice, this list of conditions and the following disclaimer in
|
---|
| 16 | * the documentation and/or other materials provided with the
|
---|
| 17 | * distribution.
|
---|
| 18 | *
|
---|
| 19 | * 3. The end-user documentation included with the redistribution,
|
---|
| 20 | * if any, must include the following acknowledgment:
|
---|
| 21 | * "This product includes software developed by the
|
---|
| 22 | * Apache Software Foundation (http://www.apache.org/)."
|
---|
| 23 | * Alternately, this acknowledgment may appear in the software itself,
|
---|
| 24 | * if and wherever such third-party acknowledgments normally appear.
|
---|
| 25 | *
|
---|
| 26 | * 4. The names "Xerces" and "Apache Software Foundation" must
|
---|
| 27 | * not be used to endorse or promote products derived from this
|
---|
| 28 | * software without prior written permission. For written
|
---|
| 29 | * permission, please contact apache\@apache.org.
|
---|
| 30 | *
|
---|
| 31 | * 5. Products derived from this software may not be called "Apache",
|
---|
| 32 | * nor may "Apache" appear in their name, without prior written
|
---|
| 33 | * permission of the Apache Software Foundation.
|
---|
| 34 | *
|
---|
| 35 | * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
|
---|
| 36 | * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
---|
| 37 | * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
|
---|
| 38 | * DISCLAIMED. IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
|
---|
| 39 | * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
---|
| 40 | * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
---|
| 41 | * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
|
---|
| 42 | * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
|
---|
| 43 | * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
|
---|
| 44 | * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
|
---|
| 45 | * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
---|
| 46 | * SUCH DAMAGE.
|
---|
| 47 | * ====================================================================
|
---|
| 48 | *
|
---|
| 49 | * This software consists of voluntary contributions made by many
|
---|
| 50 | * individuals on behalf of the Apache Software Foundation, and was
|
---|
| 51 | * originally based on software copyright (c) 1999, International
|
---|
| 52 | * Business Machines, Inc., http://www.ibm.com . For more information
|
---|
| 53 | * on the Apache Software Foundation, please see
|
---|
| 54 | * <http://www.apache.org/>.
|
---|
| 55 | */
|
---|
| 56 |
|
---|
| 57 | /*
|
---|
| 58 | * $Log: DGXMLScanner.hpp,v $
|
---|
| 59 | * Revision 1.12 2004/01/29 11:46:30 cargilld
|
---|
| 60 | * Code cleanup changes to get rid of various compiler diagnostic messages.
|
---|
| 61 | *
|
---|
| 62 | * Revision 1.11 2003/11/24 05:09:39 neilg
|
---|
| 63 | * implement new, statless, method for detecting duplicate attributes
|
---|
| 64 | *
|
---|
| 65 | * Revision 1.10 2003/10/22 20:22:30 knoaman
|
---|
| 66 | * Prepare for annotation support.
|
---|
| 67 | *
|
---|
| 68 | * Revision 1.9 2003/09/22 19:51:41 neilg
|
---|
| 69 | * scanners should maintain their own pools of undeclared elements, rather than requiring grammars to do this. This makes grammar objects stateless with regard to validation.
|
---|
| 70 | *
|
---|
| 71 | * Revision 1.8 2003/07/24 09:19:09 gareth
|
---|
| 72 | * Patch for bug #20530 - Attributes which have the same expanded name are not considered duplicates. Patch by cargilld.
|
---|
| 73 | *
|
---|
| 74 | * Revision 1.7 2003/07/10 19:47:23 peiyongz
|
---|
| 75 | * Stateless Grammar: Initialize scanner with grammarResolver,
|
---|
| 76 | * creating grammar through grammarPool
|
---|
| 77 | *
|
---|
| 78 | * Revision 1.6 2003/05/22 02:10:51 knoaman
|
---|
| 79 | * Default the memory manager.
|
---|
| 80 | *
|
---|
| 81 | * Revision 1.5 2003/05/15 18:26:29 knoaman
|
---|
| 82 | * Partial implementation of the configurable memory manager.
|
---|
| 83 | *
|
---|
| 84 | * Revision 1.4 2003/03/07 18:08:58 tng
|
---|
| 85 | * Return a reference instead of void for operator=
|
---|
| 86 | *
|
---|
| 87 | * Revision 1.3 2003/01/15 15:49:49 knoaman
|
---|
| 88 | * Change constant declaration name to match its value.
|
---|
| 89 | *
|
---|
| 90 | * Revision 1.2 2003/01/13 18:29:41 knoaman
|
---|
| 91 | * Return proper scanner name.
|
---|
| 92 | *
|
---|
| 93 | * Revision 1.1 2002/12/09 15:45:13 knoaman
|
---|
| 94 | * Initial check-in.
|
---|
| 95 | *
|
---|
| 96 | */
|
---|
| 97 |
|
---|
| 98 |
|
---|
| 99 | #if !defined(DGXMLSCANNER_HPP)
|
---|
| 100 | #define DGXMLSCANNER_HPP
|
---|
| 101 |
|
---|
| 102 | #include <xercesc/internal/XMLScanner.hpp>
|
---|
| 103 | #include <xercesc/util/ValueVectorOf.hpp>
|
---|
| 104 | #include <xercesc/util/NameIdPool.hpp>
|
---|
| 105 | #include <xercesc/validators/common/Grammar.hpp>
|
---|
| 106 |
|
---|
| 107 | XERCES_CPP_NAMESPACE_BEGIN
|
---|
| 108 |
|
---|
| 109 | class DTDElementDecl;
|
---|
| 110 | class DTDGrammar;
|
---|
| 111 | class DTDValidator;
|
---|
| 112 |
|
---|
| 113 | // This is an integrated scanner class, which does DTD/XML Schema grammar
|
---|
| 114 | // processing.
|
---|
| 115 | class XMLPARSER_EXPORT DGXMLScanner : public XMLScanner
|
---|
| 116 | {
|
---|
| 117 | public :
|
---|
| 118 | // -----------------------------------------------------------------------
|
---|
| 119 | // Constructors and Destructor
|
---|
| 120 | // -----------------------------------------------------------------------
|
---|
| 121 | DGXMLScanner
|
---|
| 122 | (
|
---|
| 123 | XMLValidator* const valToAdopt
|
---|
| 124 | , GrammarResolver* const grammarResolver
|
---|
| 125 | , MemoryManager* const manager = XMLPlatformUtils::fgMemoryManager
|
---|
| 126 | );
|
---|
| 127 | DGXMLScanner
|
---|
| 128 | (
|
---|
| 129 | XMLDocumentHandler* const docHandler
|
---|
| 130 | , DocTypeHandler* const docTypeHandler
|
---|
| 131 | , XMLEntityHandler* const entityHandler
|
---|
| 132 | , XMLErrorReporter* const errReporter
|
---|
| 133 | , XMLValidator* const valToAdopt
|
---|
| 134 | , GrammarResolver* const grammarResolver
|
---|
| 135 | , MemoryManager* const manager = XMLPlatformUtils::fgMemoryManager
|
---|
| 136 | );
|
---|
| 137 | virtual ~DGXMLScanner();
|
---|
| 138 |
|
---|
| 139 | // -----------------------------------------------------------------------
|
---|
| 140 | // XMLScanner public virtual methods
|
---|
| 141 | // -----------------------------------------------------------------------
|
---|
| 142 | virtual const XMLCh* getName() const;
|
---|
| 143 | virtual NameIdPool<DTDEntityDecl>* getEntityDeclPool();
|
---|
| 144 | virtual const NameIdPool<DTDEntityDecl>* getEntityDeclPool() const;
|
---|
| 145 | virtual unsigned int resolveQName
|
---|
| 146 | (
|
---|
| 147 | const XMLCh* const qName
|
---|
| 148 | , XMLBuffer& prefixBufToFill
|
---|
| 149 | , const short mode
|
---|
| 150 | , int& prefixColonPos
|
---|
| 151 | );
|
---|
| 152 | virtual void scanDocument
|
---|
| 153 | (
|
---|
| 154 | const InputSource& src
|
---|
| 155 | );
|
---|
| 156 | virtual bool scanNext(XMLPScanToken& toFill);
|
---|
| 157 | virtual Grammar* loadGrammar
|
---|
| 158 | (
|
---|
| 159 | const InputSource& src
|
---|
| 160 | , const short grammarType
|
---|
| 161 | , const bool toCache = false
|
---|
| 162 | );
|
---|
| 163 |
|
---|
| 164 | private :
|
---|
| 165 | // -----------------------------------------------------------------------
|
---|
| 166 | // Unimplemented constructors and operators
|
---|
| 167 | // -----------------------------------------------------------------------
|
---|
| 168 | DGXMLScanner();
|
---|
| 169 | DGXMLScanner(const DGXMLScanner&);
|
---|
| 170 | DGXMLScanner& operator=(const DGXMLScanner&);
|
---|
| 171 |
|
---|
| 172 | // -----------------------------------------------------------------------
|
---|
| 173 | // XMLScanner virtual methods
|
---|
| 174 | // -----------------------------------------------------------------------
|
---|
| 175 | virtual void scanCDSection();
|
---|
| 176 | virtual void scanCharData(XMLBuffer& toToUse);
|
---|
| 177 | virtual EntityExpRes scanEntityRef
|
---|
| 178 | (
|
---|
| 179 | const bool inAttVal
|
---|
| 180 | , XMLCh& firstCh
|
---|
| 181 | , XMLCh& secondCh
|
---|
| 182 | , bool& escaped
|
---|
| 183 | );
|
---|
| 184 | virtual void scanDocTypeDecl();
|
---|
| 185 | virtual void scanReset(const InputSource& src);
|
---|
| 186 | virtual void sendCharData(XMLBuffer& toSend);
|
---|
| 187 |
|
---|
| 188 | // -----------------------------------------------------------------------
|
---|
| 189 | // Private helper methods
|
---|
| 190 | // -----------------------------------------------------------------------
|
---|
| 191 | void commonInit();
|
---|
| 192 | void cleanUp();
|
---|
| 193 | InputSource* resolveSystemId(const XMLCh* const sysId); // return owned by caller
|
---|
| 194 | unsigned int buildAttList
|
---|
| 195 | (
|
---|
| 196 | const unsigned int attCount
|
---|
| 197 | , XMLElementDecl* elemDecl
|
---|
| 198 | , RefVectorOf<XMLAttr>& toFill
|
---|
| 199 | );
|
---|
| 200 | unsigned int resolvePrefix
|
---|
| 201 | (
|
---|
| 202 | const XMLCh* const prefix
|
---|
| 203 | , const ElemStack::MapModes mode
|
---|
| 204 | );
|
---|
| 205 | void updateNSMap
|
---|
| 206 | (
|
---|
| 207 | const XMLCh* const attrPrefix
|
---|
| 208 | , const XMLCh* const attrLocalName
|
---|
| 209 | , const XMLCh* const attrValue
|
---|
| 210 | );
|
---|
| 211 | void scanAttrListforNameSpaces(RefVectorOf<XMLAttr>* theAttrList, int attCount, XMLElementDecl* elemDecl);
|
---|
| 212 |
|
---|
| 213 | // -----------------------------------------------------------------------
|
---|
| 214 | // Private scanning methods
|
---|
| 215 | // -----------------------------------------------------------------------
|
---|
| 216 | bool scanAttValue
|
---|
| 217 | (
|
---|
| 218 | const XMLAttDef* const attDef
|
---|
| 219 | , const XMLCh *const attrName
|
---|
| 220 | , XMLBuffer& toFill
|
---|
| 221 | );
|
---|
| 222 | bool scanContent();
|
---|
| 223 | void scanEndTag(bool& gotData);
|
---|
| 224 | bool scanStartTag(bool& gotData);
|
---|
| 225 | bool scanStartTagNS(bool& gotData);
|
---|
| 226 |
|
---|
| 227 | // -----------------------------------------------------------------------
|
---|
| 228 | // Grammar preparsing methods
|
---|
| 229 | // -----------------------------------------------------------------------
|
---|
| 230 | Grammar* loadDTDGrammar(const InputSource& src, const bool toCache = false);
|
---|
| 231 |
|
---|
| 232 | // -----------------------------------------------------------------------
|
---|
| 233 | // Data members
|
---|
| 234 | //
|
---|
| 235 | // fRawAttrList
|
---|
| 236 | // During the initial scan of the attributes we can only do a raw
|
---|
| 237 | // scan for key/value pairs. So this vector is used to store them
|
---|
| 238 | // until they can be processed (and put into fAttrList.)
|
---|
| 239 | //
|
---|
| 240 | // fDTDValidator
|
---|
| 241 | // The DTD validator instance.
|
---|
| 242 | //
|
---|
| 243 | // fElemState
|
---|
| 244 | // fElemStateSize
|
---|
| 245 | // Stores an element next state from DFA content model - used for
|
---|
| 246 | // wildcard validation
|
---|
| 247 | //
|
---|
| 248 | // fDTDElemNonDeclPool
|
---|
| 249 | // registry of "faulted-in" DTD element decls
|
---|
| 250 | // fElemCount
|
---|
| 251 | // count of the number of start tags seen so far (starts at 1).
|
---|
| 252 | // Used for duplicate attribute detection/processing of required/defaulted attributes
|
---|
| 253 | // fAttDefRegistry
|
---|
| 254 | // mapping from XMLAttDef instances to the count of the last
|
---|
| 255 | // start tag where they were utilized.
|
---|
| 256 | // fUndeclaredAttrRegistry
|
---|
| 257 | // mapping of attr QNames to the count of the last start tag in which they occurred
|
---|
| 258 | //
|
---|
| 259 | // -----------------------------------------------------------------------
|
---|
| 260 | ValueVectorOf<XMLAttr*>* fAttrNSList;
|
---|
| 261 | DTDValidator* fDTDValidator;
|
---|
| 262 | DTDGrammar* fDTDGrammar;
|
---|
| 263 | NameIdPool<DTDElementDecl>* fDTDElemNonDeclPool;
|
---|
| 264 | unsigned int fElemCount;
|
---|
| 265 | RefHashTableOf<unsigned int>* fAttDefRegistry;
|
---|
| 266 | RefHashTableOf<unsigned int>* fUndeclaredAttrRegistry;
|
---|
| 267 | };
|
---|
| 268 |
|
---|
| 269 | inline const XMLCh* DGXMLScanner::getName() const
|
---|
| 270 | {
|
---|
| 271 | return XMLUni::fgDGXMLScanner;
|
---|
| 272 | }
|
---|
| 273 |
|
---|
| 274 |
|
---|
| 275 | XERCES_CPP_NAMESPACE_END
|
---|
| 276 |
|
---|
| 277 | #endif
|
---|