Home | History | Annotate | Download | only in parser
      1 /*
      2  * Copyright (C) 2000 Peter Kelly (pmk (at) post.com)
      3  * Copyright (C) 2005, 2006, 2007 Apple Inc. All rights reserved.
      4  * Copyright (C) 2007 Samuel Weinig (sam (at) webkit.org)
      5  * Copyright (C) 2008 Nokia Corporation and/or its subsidiary(-ies)
      6  * Copyright (C) 2008, 2009 Torch Mobile Inc. All rights reserved. (http://www.torchmobile.com/)
      7  *
      8  * This library is free software; you can redistribute it and/or
      9  * modify it under the terms of the GNU Library General Public
     10  * License as published by the Free Software Foundation; either
     11  * version 2 of the License, or (at your option) any later version.
     12  *
     13  * This library is distributed in the hope that it will be useful,
     14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
     15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
     16  * Library General Public License for more details.
     17  *
     18  * You should have received a copy of the GNU Library General Public License
     19  * along with this library; see the file COPYING.LIB.  If not, write to
     20  * the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor,
     21  * Boston, MA 02110-1301, USA.
     22  *
     23  */
     24 
     25 #ifndef XMLDocumentParser_h
     26 #define XMLDocumentParser_h
     27 
     28 #include "core/dom/ParserContentPolicy.h"
     29 #include "core/dom/ScriptableDocumentParser.h"
     30 #include "core/loader/cache/ResourceClient.h"
     31 #include "core/loader/cache/ResourcePtr.h"
     32 #include "core/platform/text/SegmentedString.h"
     33 #include "core/xml/XMLErrors.h"
     34 #include <libxml/tree.h>
     35 #include <wtf/HashMap.h>
     36 #include <wtf/OwnPtr.h>
     37 #include <wtf/text/CString.h>
     38 #include <wtf/text/StringHash.h>
     39 
     40 namespace WebCore {
     41 
     42 class ContainerNode;
     43 class ScriptResource;
     44 class ResourceFetcher;
     45 class DocumentFragment;
     46 class Document;
     47 class Element;
     48 class FrameView;
     49 class Text;
     50 
     51     class XMLParserContext : public RefCounted<XMLParserContext> {
     52     public:
     53         static PassRefPtr<XMLParserContext> createMemoryParser(xmlSAXHandlerPtr, void* userData, const CString& chunk);
     54         static PassRefPtr<XMLParserContext> createStringParser(xmlSAXHandlerPtr, void* userData);
     55         ~XMLParserContext();
     56         xmlParserCtxtPtr context() const { return m_context; }
     57 
     58     private:
     59         XMLParserContext(xmlParserCtxtPtr context)
     60             : m_context(context)
     61         {
     62         }
     63         xmlParserCtxtPtr m_context;
     64     };
     65 
     66     class XMLDocumentParser : public ScriptableDocumentParser, public ResourceClient {
     67         WTF_MAKE_FAST_ALLOCATED;
     68     public:
     69         static PassRefPtr<XMLDocumentParser> create(Document* document, FrameView* view)
     70         {
     71             return adoptRef(new XMLDocumentParser(document, view));
     72         }
     73         static PassRefPtr<XMLDocumentParser> create(DocumentFragment* fragment, Element* element, ParserContentPolicy parserContentPolicy)
     74         {
     75             return adoptRef(new XMLDocumentParser(fragment, element, parserContentPolicy));
     76         }
     77 
     78         ~XMLDocumentParser();
     79 
     80         // Exposed for callbacks:
     81         void handleError(XMLErrors::ErrorType, const char* message, TextPosition);
     82 
     83         void setIsXHTMLDocument(bool isXHTML) { m_isXHTMLDocument = isXHTML; }
     84         bool isXHTMLDocument() const { return m_isXHTMLDocument; }
     85 
     86         bool isCurrentlyParsing8BitChunk() { return m_isCurrentlyParsing8BitChunk; }
     87 
     88         static bool parseDocumentFragment(const String&, DocumentFragment*, Element* parent = 0, ParserContentPolicy = AllowScriptingContent);
     89 
     90         // Used by the XMLHttpRequest to check if the responseXML was well formed.
     91         virtual bool wellFormed() const { return !m_sawError; }
     92 
     93         TextPosition textPosition() const;
     94 
     95         static bool supportsXMLVersion(const String&);
     96 
     97         class PendingCallback {
     98         public:
     99             virtual ~PendingCallback() { }
    100             virtual void call(XMLDocumentParser*) = 0;
    101         };
    102 
    103     private:
    104         XMLDocumentParser(Document*, FrameView* = 0);
    105         XMLDocumentParser(DocumentFragment*, Element*, ParserContentPolicy);
    106 
    107         // From DocumentParser
    108         virtual void insert(const SegmentedString&);
    109         virtual void append(PassRefPtr<StringImpl>);
    110         virtual void finish();
    111         virtual bool isWaitingForScripts() const;
    112         virtual void stopParsing();
    113         virtual void detach();
    114         virtual OrdinalNumber lineNumber() const;
    115         OrdinalNumber columnNumber() const;
    116 
    117         // from ResourceClient
    118         virtual void notifyFinished(Resource*);
    119 
    120         void end();
    121 
    122         void pauseParsing();
    123         void resumeParsing();
    124 
    125         bool appendFragmentSource(const String&);
    126 
    127     public:
    128         // callbacks from parser SAX
    129         void error(XMLErrors::ErrorType, const char* message, va_list args) WTF_ATTRIBUTE_PRINTF(3, 0);
    130         void startElementNs(const AtomicString& localName, const AtomicString& prefix, const AtomicString& uri, int nb_namespaces,
    131                             const xmlChar** namespaces, int nb_attributes, int nb_defaulted, const xmlChar** libxmlAttributes);
    132         void endElementNs();
    133         void characters(const xmlChar* chars, int length);
    134         void processingInstruction(const String& target, const String& data);
    135         void cdataBlock(const String&);
    136         void comment(const String&);
    137         void startDocument(const String& version, const String& encoding, int standalone);
    138         void internalSubset(const String& name, const String& externalID, const String& systemID);
    139         void endDocument();
    140 
    141     private:
    142         void initializeParserContext(const CString& chunk = CString());
    143 
    144         void pushCurrentNode(ContainerNode*);
    145         void popCurrentNode();
    146         void clearCurrentNodeStack();
    147 
    148         void insertErrorMessageBlock();
    149 
    150         void enterText();
    151         void exitText();
    152 
    153         void doWrite(const String&);
    154         void doEnd();
    155 
    156         FrameView* m_view;
    157 
    158         SegmentedString m_originalSourceForTransform;
    159 
    160         xmlParserCtxtPtr context() const { return m_context ? m_context->context() : 0; };
    161         RefPtr<XMLParserContext> m_context;
    162         Deque<OwnPtr<PendingCallback> > m_pendingCallbacks;
    163         Vector<xmlChar> m_bufferedText;
    164 
    165         ContainerNode* m_currentNode;
    166         Vector<ContainerNode*> m_currentNodeStack;
    167 
    168         RefPtr<Text> m_leafTextNode;
    169 
    170         bool m_isCurrentlyParsing8BitChunk;
    171         bool m_sawError;
    172         bool m_sawCSS;
    173         bool m_sawXSLTransform;
    174         bool m_sawFirstElement;
    175         bool m_isXHTMLDocument;
    176         bool m_parserPaused;
    177         bool m_requestingScript;
    178         bool m_finishCalled;
    179 
    180         XMLErrors m_xmlErrors;
    181 
    182         ResourcePtr<ScriptResource> m_pendingScript;
    183         RefPtr<Element> m_scriptElement;
    184         TextPosition m_scriptStartPosition;
    185 
    186         bool m_parsingFragment;
    187         AtomicString m_defaultNamespaceURI;
    188 
    189         typedef HashMap<AtomicString, AtomicString> PrefixForNamespaceMap;
    190         PrefixForNamespaceMap m_prefixToNamespaceMap;
    191         SegmentedString m_pendingSrc;
    192     };
    193 
    194 xmlDocPtr xmlDocPtrForString(ResourceFetcher*, const String& source, const String& url);
    195 
    196 HashMap<String, String> parseAttributes(const String&, bool& attrsOK);
    197 
    198 } // namespace WebCore
    199 
    200 #endif // XMLDocumentParser_h
    201