Merge page serializers [11/12]

Source/core/page/PageSerializer.cpp

Index: Source/core/page/PageSerializer.cpp
diff --git a/Source/core/page/PageSerializer.cpp b/Source/core/page/PageSerializer.cpp
index a7c6ffd0d36109194c4d0c279731c0e79b946074..bbbc4c749e5920d61a7d11c513733ef6fe87040f 100644
--- a/Source/core/page/PageSerializer.cpp
+++ b/Source/core/page/PageSerializer.cpp
@@ -87,18 +87,13 @@ static bool isCharsetSpecifyingNode(const Node& node)
     return textEncoding.isValid();
 }
 
-static bool shouldIgnoreElement(const Element& element)
-{
-    return isHTMLScriptElement(element) || isHTMLNoScriptElement(element) || isCharsetSpecifyingNode(element);
-}
-
 static const QualifiedName& frameOwnerURLAttributeName(const HTMLFrameOwnerElement& frameOwner)
 {
     // FIXME: We should support all frame owners including applets.
     return isHTMLObjectElement(frameOwner) ? HTMLNames::dataAttr : HTMLNames::srcAttr;
 }
 
-class SerializerMarkupAccumulator final : public MarkupAccumulator {
+class SerializerMarkupAccumulator : public MarkupAccumulator {
     STACK_ALLOCATED();
 public:
     SerializerMarkupAccumulator(PageSerializer*, const Document&, WillBeHeapVector<RawPtrWillBeMember<Node>>&);
@@ -112,6 +107,11 @@ protected:
     virtual void appendStartTag(Node&, Namespaces* = nullptr) override;
     virtual void appendEndTag(const Element&) override;
 
+    virtual bool shouldIgnoreElement(const Element&) const;
+
+    PageSerializer* pageSerializer();
+    const Document& document();
+
 private:
     PageSerializer* m_serializer;
     RawPtrWillBeMember<const Document> m_document;
@@ -203,8 +203,108 @@ void SerializerMarkupAccumulator::appendEndTag(const Element& element)
         MarkupAccumulator::appendEndTag(element);
 }
 
-PageSerializer::PageSerializer(Vector<SerializedResource>* resources, PassOwnPtr<Delegate> delegate)
+bool SerializerMarkupAccumulator::shouldIgnoreElement(const Element& element) const
+{
+    return isHTMLScriptElement(element) || isHTMLNoScriptElement(element) || isCharsetSpecifyingNode(element);
+}
+
+PageSerializer* SerializerMarkupAccumulator::pageSerializer()
+{
+    return m_serializer;
+}
+
+const Document& SerializerMarkupAccumulator::document()
+{
+    return *m_document;
+}
+
+class LinkChangeSerializerMarkupAccumulator final : public SerializerMarkupAccumulator {
+public:
+    LinkChangeSerializerMarkupAccumulator(PageSerializer*, const Document&, WillBeHeapVector<RawPtrWillBeMember<Node>>&, LinkLocalPathMap*, String);
+
+protected:

[yosin_UTC9, 2015/06/16 01:21:50]

Since this class is marked |final|, please move protected members to private.

[Tiger (Sony Mobile), 2015/06/16 09:28:47]

Done.

+    void appendElement(StringBuilder&, Element&, Namespaces*) override;
+    void appendAttribute(StringBuilder&, const Element&, const Attribute&, Namespaces*) override;
+
+    bool shouldIgnoreElement(const Element&) const override;
+
+private:
+    // m_replaceLinks include all pair of local resource path and corresponding original link.
+    LinkLocalPathMap* m_replaceLinks;
+    String m_directoryName;
+};
+
+LinkChangeSerializerMarkupAccumulator::LinkChangeSerializerMarkupAccumulator(PageSerializer* serializer,  const Document& document, WillBeHeapVector<RawPtrWillBeMember<Node>>& nodes, LinkLocalPathMap* links, String directoryName)
+    : SerializerMarkupAccumulator(serializer, document, nodes)
+    , m_replaceLinks(links)
+    , m_directoryName(directoryName)
+{
+}
+
+void LinkChangeSerializerMarkupAccumulator::appendElement(StringBuilder& result, Element& element, Namespaces* namespaces)
+{
+    if (element.hasTagName(HTMLNames::htmlTag)) {
+        // Add MOTW (Mark of the Web) declaration before html tag.
+        // See http://msdn2.microsoft.com/en-us/library/ms537628(VS.85).aspx.
+        result.append('\n');
+        MarkupFormatter::appendComment(result, String::format(" saved from url=(%04d)%s ",
+            static_cast<int>(document().url().string().utf8().length()),
+            document().url().string().utf8().data()));
+        result.append('\n');
+    }
+
+    SerializerMarkupAccumulator::appendElement(result, element, namespaces);
+
+    if (element.hasTagName(HTMLNames::baseTag)) {
+        // TODO(tiger): Refactor MarkupAccumulator so it is easier to append an element like this, without special cases for XHTML
+        // Append a new base tag declaration.
+        result.appendLiteral("<base href=\".\"");
+        if (!document().baseTarget().isEmpty()) {
+            result.appendLiteral(" target=\"");
+            result.append(document().baseTarget());

[hajimehoshi, 2015/06/16 04:46:54]

MarkupFormatter::appendAttributeValue

[Tiger (Sony Mobile), 2015/06/16 09:28:47]

Done.

+            result.append('"');
+        }
+        if (document().isXHTMLDocument())
+            result.appendLiteral(" />");
+        else
+            result.appendLiteral(">");
+    }
+}
+
+void LinkChangeSerializerMarkupAccumulator::appendAttribute(StringBuilder& result, const Element& element, const Attribute& attribute, Namespaces* namespaces)
+{
+    if (m_replaceLinks && element.isURLAttribute(attribute) && !element.isJavaScriptURLAttribute(attribute)) {
+
+        String completeURL = document().completeURL(attribute.value());
+
+        if (m_replaceLinks->contains(completeURL)) {
+            // TODO(tiger): Refactor MarkupAccumulator so it is easier to append an attribute like this.
+            result.append(' ');
+            result.append(attribute.name().toString());
+            result.appendLiteral("=\"");
+            if (!m_directoryName.isEmpty()) {
+                result.appendLiteral("./");
+                MarkupFormatter::appendAttributeValue(result, m_directoryName, document().isHTMLDocument());
+                result.append('/');

[hajimehoshi, 2015/06/16 04:46:55]

MarkupFormatter::appendAttributeValue(result, "./" + m_directoryName + "/",
document().isHTMLDocument());

[Tiger (Sony Mobile), 2015/06/16 09:28:47]

Done.

+            }
+            MarkupFormatter::appendAttributeValue(result, m_replaceLinks->get(completeURL), document().isHTMLDocument());
+            result.appendLiteral("\"");
+            return;
+        }
+    }
+    MarkupAccumulator::appendAttribute(result, element, attribute, namespaces);
+}
+
+bool LinkChangeSerializerMarkupAccumulator::shouldIgnoreElement(const Element& element) const
+{
+    return SerializerMarkupAccumulator::shouldIgnoreElement(element) || isHTMLBaseElement(element);
+}
+
+
+PageSerializer::PageSerializer(Vector<SerializedResource>* resources, PassOwnPtr<Delegate> delegate, LinkLocalPathMap* urls, String directory)
     : m_resources(resources)
+    , m_URLs(urls)
+    , m_directory(directory)
     , m_blankFrameCounter(0)
     , m_delegate(delegate)
 {
@@ -241,8 +341,15 @@ void PageSerializer::serializeFrame(LocalFrame* frame)
     }
 
     WillBeHeapVector<RawPtrWillBeMember<Node>> serializedNodes;
-    SerializerMarkupAccumulator accumulator(this, document, serializedNodes);
-    String text = serializeNodes<EditingStrategy>(accumulator, document, IncludeNode);
+    String text;
+    if (m_URLs) {
+        LinkChangeSerializerMarkupAccumulator accumulator(this, document, serializedNodes, m_URLs, m_directory);
+        text = serializeNodes<EditingStrategy>(accumulator, document, IncludeNode);
+    } else {
+        SerializerMarkupAccumulator accumulator(this, document, serializedNodes);
+        text = serializeNodes<EditingStrategy>(accumulator, document, IncludeNode);
+    }
+
     WTF::TextEncoding textEncoding(document.charset());
     CString frameHTML = textEncoding.normalizeAndEncode(text, WTF::EntitiesForUnencodables);
     m_resources->append(SerializedResource(url, document.suggestedMIMEType(), SharedBuffer::create(frameHTML.data(), frameHTML.length())));