Revert "Remove old serializer" It caused a regression, crbug.com/510422

Source/web/WebPageSerializerImpl.cpp

Index: Source/web/WebPageSerializerImpl.cpp
diff --git a/Source/web/WebPageSerializerImpl.cpp b/Source/web/WebPageSerializerImpl.cpp
new file mode 100644
index 0000000000000000000000000000000000000000..53a421db9cd5b5cfaeac2cdf96615fc3d9ae73b8
--- /dev/null
+++ b/Source/web/WebPageSerializerImpl.cpp
@@ -0,0 +1,520 @@
+/*
+ * Copyright (C) 2009 Google Inc. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are
+ * met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following disclaimer
+ * in the documentation and/or other materials provided with the
+ * distribution.
+ *     * Neither the name of Google Inc. nor the names of its
+ * contributors may be used to endorse or promote products derived from
+ * this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+// How we handle the base tag better.
+// Current status:
+// At now the normal way we use to handling base tag is
+// a) For those links which have corresponding local saved files, such as
+// savable CSS, JavaScript files, they will be written to relative URLs which
+// point to local saved file. Why those links can not be resolved as absolute
+// file URLs, because if they are resolved as absolute URLs, after moving the
+// file location from one directory to another directory, the file URLs will
+// be dead links.
+// b) For those links which have not corresponding local saved files, such as
+// links in A, AREA tags, they will be resolved as absolute URLs.
+// c) We comment all base tags when serialzing DOM for the page.
+// FireFox also uses above way to handle base tag.
+//
+// Problem:
+// This way can not handle the following situation:
+// the base tag is written by JavaScript.
+// For example. The page "www.yahoo.com" use
+// "document.write('<base href="http://www.yahoo.com/"...');" to setup base URL
+// of page when loading page. So when saving page as completed-HTML, we assume
+// that we save "www.yahoo.com" to "c:\yahoo.htm". After then we load the saved
+// completed-HTML page, then the JavaScript will insert a base tag
+// <base href="http://www.yahoo.com/"...> to DOM, so all URLs which point to
+// local saved resource files will be resolved as
+// "http://www.yahoo.com/yahoo_files/...", which will cause all saved  resource
+// files can not be loaded correctly. Also the page will be rendered ugly since
+// all saved sub-resource files (such as CSS, JavaScript files) and sub-frame
+// files can not be fetched.
+// Now FireFox, IE and WebKit based Browser all have this problem.
+//
+// Solution:
+// My solution is that we comment old base tag and write new base tag:
+// <base href="." ...> after the previous commented base tag. In WebKit, it
+// always uses the latest "href" attribute of base tag to set document's base
+// URL. Based on this behavior, when we encounter a base tag, we comment it and
+// write a new base tag <base href="."> after the previous commented base tag.
+// The new added base tag can help engine to locate correct base URL for
+// correctly loading local saved resource files. Also I think we need to inherit
+// the base target value from document object when appending new base tag.
+// If there are multiple base tags in original document, we will comment all old
+// base tags and append new base tag after each old base tag because we do not
+// know those old base tags are original content or added by JavaScript. If
+// they are added by JavaScript, it means when loading saved page, the script(s)
+// will still insert base tag(s) to DOM, so the new added base tag(s) can
+// override the incorrect base URL and make sure we alway load correct local
+// saved resource files.
+
+#include "config.h"
+#include "web/WebPageSerializerImpl.h"
+
+#include "core/HTMLNames.h"
+#include "core/dom/Document.h"
+#include "core/dom/DocumentType.h"
+#include "core/dom/Element.h"
+#include "core/editing/Serialization.h"
+#include "core/html/HTMLAllCollection.h"
+#include "core/html/HTMLElement.h"
+#include "core/html/HTMLFormElement.h"
+#include "core/html/HTMLHtmlElement.h"
+#include "core/html/HTMLMetaElement.h"
+#include "core/loader/DocumentLoader.h"
+#include "core/loader/FrameLoader.h"
+#include "public/platform/WebVector.h"
+#include "web/WebLocalFrameImpl.h"
+#include "wtf/text/TextEncoding.h"
+
+namespace blink {
+
+// Maximum length of data buffer which is used to temporary save generated
+// html content data. This is a soft limit which might be passed if a very large
+// contegious string is found in the page.
+static const unsigned dataBufferCapacity = 65536;
+
+WebPageSerializerImpl::SerializeDomParam::SerializeDomParam(const KURL& url,
+                                                            const WTF::TextEncoding& textEncoding,
+                                                            Document* document,
+                                                            const String& directoryName)
+    : url(url)
+    , textEncoding(textEncoding)
+    , document(document)
+    , directoryName(directoryName)
+    , isHTMLDocument(document->isHTMLDocument())
+    , haveSeenDocType(false)
+    , haveAddedCharsetDeclaration(false)
+    , skipMetaElement(nullptr)
+    , isInScriptOrStyleTag(false)
+    , haveAddedXMLProcessingDirective(false)
+    , haveAddedContentsBeforeEnd(false)
+{
+}
+
+String WebPageSerializerImpl::preActionBeforeSerializeOpenTag(
+    const Element* element, SerializeDomParam* param, bool* needSkip)
+{
+    StringBuilder result;
+
+    *needSkip = false;
+    if (param->isHTMLDocument) {
+        // Skip the open tag of original META tag which declare charset since we
+        // have overrided the META which have correct charset declaration after
+        // serializing open tag of HEAD element.
+        ASSERT(element);
+        if (isHTMLMetaElement(*element)) {
+            const HTMLMetaElement& meta = toHTMLMetaElement(*element);
+            // Check whether the META tag has declared charset or not.
+            String equiv = meta.httpEquiv();
+            if (equalIgnoringCase(equiv, "content-type")) {
+                String content = meta.content();
+                if (content.length() && content.contains("charset", TextCaseInsensitive)) {
+                    // Find META tag declared charset, we need to skip it when
+                    // serializing DOM.
+                    param->skipMetaElement = element;
+                    *needSkip = true;
+                }
+            }
+        } else if (isHTMLHtmlElement(*element)) {
+            // Check something before processing the open tag of HEAD element.
+            // First we add doc type declaration if original document has it.
+            if (!param->haveSeenDocType) {
+                param->haveSeenDocType = true;
+                result.append(createMarkup(param->document->doctype()));
+            }
+
+            // Add MOTW declaration before html tag.
+            // See http://msdn2.microsoft.com/en-us/library/ms537628(VS.85).aspx.
+            result.append(WebPageSerializer::generateMarkOfTheWebDeclaration(param->url));
+        } else if (isHTMLBaseElement(*element)) {
+            // Comment the BASE tag when serializing dom.
+            result.appendLiteral("<!--");
+        }
+    } else {
+        // Write XML declaration.
+        if (!param->haveAddedXMLProcessingDirective) {
+            param->haveAddedXMLProcessingDirective = true;
+            // Get encoding info.
+            String xmlEncoding = param->document->xmlEncoding();
+            if (xmlEncoding.isEmpty())
+                xmlEncoding = param->document->encodingName();
+            if (xmlEncoding.isEmpty())
+                xmlEncoding = UTF8Encoding().name();
+            result.appendLiteral("<?xml version=\"");
+            result.append(param->document->xmlVersion());
+            result.appendLiteral("\" encoding=\"");
+            result.append(xmlEncoding);
+            if (param->document->xmlStandalone())
+                result.appendLiteral("\" standalone=\"yes");
+            result.appendLiteral("\"?>\n");
+        }
+        // Add doc type declaration if original document has it.
+        if (!param->haveSeenDocType) {
+            param->haveSeenDocType = true;
+            result.append(createMarkup(param->document->doctype()));
+        }
+    }
+    return result.toString();
+}
+
+String WebPageSerializerImpl::postActionAfterSerializeOpenTag(
+    const Element* element, SerializeDomParam* param)
+{
+    StringBuilder result;
+
+    param->haveAddedContentsBeforeEnd = false;
+    if (!param->isHTMLDocument)
+        return result.toString();
+    // Check after processing the open tag of HEAD element
+    if (!param->haveAddedCharsetDeclaration
+        && isHTMLHeadElement(*element)) {
+        param->haveAddedCharsetDeclaration = true;
+        // Check meta element. WebKit only pre-parse the first 512 bytes
+        // of the document. If the whole <HEAD> is larger and meta is the
+        // end of head part, then this kind of pages aren't decoded correctly
+        // because of this issue. So when we serialize the DOM, we need to
+        // make sure the meta will in first child of head tag.
+        // See http://bugs.webkit.org/show_bug.cgi?id=16621.
+        // First we generate new content for writing correct META element.
+        result.append(WebPageSerializer::generateMetaCharsetDeclaration(
+            String(param->textEncoding.name())));
+
+        param->haveAddedContentsBeforeEnd = true;
+        // Will search each META which has charset declaration, and skip them all
+        // in PreActionBeforeSerializeOpenTag.
+    } else if (isHTMLScriptElement(*element) || isHTMLScriptElement(*element)) {
+        param->isInScriptOrStyleTag = true;
+    }
+
+    return result.toString();
+}
+
+String WebPageSerializerImpl::preActionBeforeSerializeEndTag(
+    const Element* element, SerializeDomParam* param, bool* needSkip)
+{
+    String result;
+
+    *needSkip = false;
+    if (!param->isHTMLDocument)
+        return result;
+    // Skip the end tag of original META tag which declare charset.
+    // Need not to check whether it's META tag since we guarantee
+    // skipMetaElement is definitely META tag if it's not 0.
+    if (param->skipMetaElement == element) {
+        *needSkip = true;
+    } else if (isHTMLScriptElement(*element) || isHTMLScriptElement(*element)) {
+        ASSERT(param->isInScriptOrStyleTag);
+        param->isInScriptOrStyleTag = false;
+    }
+
+    return result;
+}
+
+// After we finish serializing end tag of a element, we give the target
+// element a chance to do some post work to add some additional data.
+String WebPageSerializerImpl::postActionAfterSerializeEndTag(
+    const Element* element, SerializeDomParam* param)
+{
+    StringBuilder result;
+
+    if (!param->isHTMLDocument)
+        return result.toString();
+    // Comment the BASE tag when serializing DOM.
+    if (isHTMLBaseElement(*element)) {
+        result.appendLiteral("-->");
+        // Append a new base tag declaration.
+        result.append(WebPageSerializer::generateBaseTagDeclaration(
+            param->document->baseTarget()));
+    }
+
+    return result.toString();
+}
+
+void WebPageSerializerImpl::saveHTMLContentToBuffer(
+    const String& result, SerializeDomParam* param)
+{
+    m_dataBuffer.append(result);
+    encodeAndFlushBuffer(WebPageSerializerClient::CurrentFrameIsNotFinished,
+                         param,
+                         DoNotForceFlush);
+}
+
+void WebPageSerializerImpl::encodeAndFlushBuffer(
+    WebPageSerializerClient::PageSerializationStatus status,
+    SerializeDomParam* param,
+    FlushOption flushOption)
+{
+    // Data buffer is not full nor do we want to force flush.
+    if (flushOption != ForceFlush && m_dataBuffer.length() <= dataBufferCapacity)
+        return;
+
+    String content = m_dataBuffer.toString();
+    m_dataBuffer.clear();
+
+    CString encodedContent = param->textEncoding.normalizeAndEncode(content, WTF::EntitiesForUnencodables);
+
+    // Send result to the client.
+    m_client->didSerializeDataForFrame(param->url,
+                                       WebCString(encodedContent.data(), encodedContent.length()),
+                                       status);
+}
+
+void WebPageSerializerImpl::openTagToString(Element* element,
+                                            SerializeDomParam* param)
+{
+    bool needSkip;
+    StringBuilder result;
+    // Do pre action for open tag.
+    result.append(preActionBeforeSerializeOpenTag(element, param, &needSkip));
+    if (needSkip)
+        return;
+    // Add open tag
+    result.append('<');
+    result.append(element->nodeName().lower());
+    // Go through all attributes and serialize them.
+    AttributeCollection attributes = element->attributes();
+    AttributeCollection::iterator end = attributes.end();
+    for (AttributeCollection::iterator it = attributes.begin(); it != end; ++it) {
+        result.append(' ');
+        // Add attribute pair
+        result.append(it->name().toString());
+        result.appendLiteral("=\"");
+        if (!it->value().isEmpty()) {
+            const String& attrValue = it->value();
+
+            // Check whether we need to replace some resource links
+            // with local resource paths.
+            const QualifiedName& attrName = it->name();
+            if (element->hasLegalLinkAttribute(attrName)) {
+                // For links start with "javascript:", we do not change it.
+                if (attrValue.startsWith("javascript:", TextCaseInsensitive)) {
+                    result.append(attrValue);
+                } else {
+                    // Get the absolute link
+                    WebLocalFrameImpl* subFrame = WebLocalFrameImpl::fromFrameOwnerElement(element);
+                    String completeURL = subFrame ? subFrame->frame()->document()->url() :
+                                                    param->document->completeURL(attrValue);
+                    // Check whether we have local files for those link.
+                    if (m_localLinks.contains(completeURL)) {
+                        if (!param->directoryName.isEmpty()) {
+                            result.appendLiteral("./");
+                            result.append(param->directoryName);
+                            result.append('/');
+                        }
+                        result.append(m_localLinks.get(completeURL));
+                    } else {
+                        result.append(completeURL);
+                    }
+                }
+            } else {
+                if (param->isHTMLDocument)
+                    result.append(m_htmlEntities.convertEntitiesInString(attrValue));
+                else
+                    result.append(m_xmlEntities.convertEntitiesInString(attrValue));
+            }
+        }
+        result.append('\"');
+    }
+
+    // Do post action for open tag.
+    String addedContents = postActionAfterSerializeOpenTag(element, param);
+    // Complete the open tag for element when it has child/children.
+    if (element->hasChildren() || param->haveAddedContentsBeforeEnd)
+        result.append('>');
+    // Append the added contents generate in  post action of open tag.
+    result.append(addedContents);
+    // Save the result to data buffer.
+    saveHTMLContentToBuffer(result.toString(), param);
+}
+
+// Serialize end tag of an specified element.
+void WebPageSerializerImpl::endTagToString(Element* element,
+                                           SerializeDomParam* param)
+{
+    bool needSkip;
+    StringBuilder result;
+    // Do pre action for end tag.
+    result.append(preActionBeforeSerializeEndTag(element, param, &needSkip));
+    if (needSkip)
+        return;
+    // Write end tag when element has child/children.
+    if (element->hasChildren() || param->haveAddedContentsBeforeEnd) {
+        result.appendLiteral("</");
+        result.append(element->nodeName().lower());
+        result.append('>');
+    } else {
+        // Check whether we have to write end tag for empty element.
+        if (param->isHTMLDocument) {
+            result.append('>');
+            // FIXME: This code is horribly wrong.  WebPageSerializerImpl must die.
+            if (!element->isHTMLElement() || !toHTMLElement(element)->ieForbidsInsertHTML()) {
+                // We need to write end tag when it is required.
+                result.appendLiteral("</");
+                result.append(element->nodeName().lower());
+                result.append('>');
+            }
+        } else {
+            // For xml base document.
+            result.appendLiteral(" />");
+        }
+    }
+    // Do post action for end tag.
+    result.append(postActionAfterSerializeEndTag(element, param));
+    // Save the result to data buffer.
+    saveHTMLContentToBuffer(result.toString(), param);
+}
+
+void WebPageSerializerImpl::buildContentForNode(Node* node,
+                                                SerializeDomParam* param)
+{
+    switch (node->nodeType()) {
+    case Node::ELEMENT_NODE:
+        // Process open tag of element.
+        openTagToString(toElement(node), param);
+        // Walk through the children nodes and process it.
+        for (Node *child = node->firstChild(); child; child = child->nextSibling())
+            buildContentForNode(child, param);
+        // Process end tag of element.
+        endTagToString(toElement(node), param);
+        break;
+    case Node::TEXT_NODE:
+        saveHTMLContentToBuffer(createMarkup(node), param);
+        break;
+    case Node::ATTRIBUTE_NODE:
+    case Node::DOCUMENT_NODE:
+    case Node::DOCUMENT_FRAGMENT_NODE:
+        // Should not exist.
+        ASSERT_NOT_REACHED();
+        break;
+    // Document type node can be in DOM?
+    case Node::DOCUMENT_TYPE_NODE:
+        param->haveSeenDocType = true;
+    default:
+        // For other type node, call default action.
+        saveHTMLContentToBuffer(createMarkup(node), param);
+        break;
+    }
+}
+
+WebPageSerializerImpl::WebPageSerializerImpl(WebFrame* frame,
+                                             bool recursiveSerialization,
+                                             WebPageSerializerClient* client,
+                                             const WebVector<WebURL>& links,
+                                             const WebVector<WebString>& localPaths,
+                                             const WebString& localDirectoryName)
+    : m_client(client)
+    , m_recursiveSerialization(recursiveSerialization)
+    , m_framesCollected(false)
+    , m_localDirectoryName(localDirectoryName)
+    , m_htmlEntities(false)
+    , m_xmlEntities(true)
+{
+    // Must specify available webframe.
+    ASSERT(frame);
+    m_specifiedWebLocalFrameImpl = toWebLocalFrameImpl(frame);
+    // Make sure we have non 0 client.
+    ASSERT(client);
+    // Build local resources map.
+    ASSERT(links.size() == localPaths.size());
+    for (size_t i = 0; i < links.size(); i++) {
+        KURL url = links[i];
+        ASSERT(!m_localLinks.contains(url.string()));
+        m_localLinks.set(url.string(), localPaths[i]);
+    }
+
+    ASSERT(m_dataBuffer.isEmpty());
+}
+
+void WebPageSerializerImpl::collectTargetFrames()
+{
+    ASSERT(!m_framesCollected);
+    m_framesCollected = true;
+
+    // First, process main frame.
+    m_frames.append(m_specifiedWebLocalFrameImpl);
+    // Return now if user only needs to serialize specified frame, not including
+    // all sub-frames.
+    if (!m_recursiveSerialization)
+        return;
+    // Collect all frames inside the specified frame.
+    for (WebLocalFrameImpl* frame : m_frames) {
+        // Get current using document.
+        Document* currentDoc = frame->frame()->document();
+        // Go through sub-frames.
+        RefPtrWillBeRawPtr<HTMLAllCollection> all = currentDoc->all();
+
+        for (unsigned i = 0; Element* element = all->item(i); ++i) {
+            if (!element->isHTMLElement())
+                continue;
+            WebLocalFrameImpl* webFrame =
+                WebLocalFrameImpl::fromFrameOwnerElement(element);
+            if (webFrame)
+                m_frames.append(webFrame);
+        }
+    }
+}
+
+bool WebPageSerializerImpl::serialize()
+{
+    if (!m_framesCollected)
+        collectTargetFrames();
+
+    bool didSerialization = false;
+    KURL mainURL = m_specifiedWebLocalFrameImpl->frame()->document()->url();
+
+    for (unsigned i = 0; i < m_frames.size(); ++i) {
+        WebLocalFrameImpl* webFrame = m_frames[i];
+        Document* document = webFrame->frame()->document();
+        const KURL& url = document->url();
+
+        if (!url.isValid() || !m_localLinks.contains(url.string()))
+            continue;
+
+        didSerialization = true;
+
+        const WTF::TextEncoding& textEncoding = document->encoding().isValid() ? document->encoding() : UTF8Encoding();
+        String directoryName = url == mainURL ? m_localDirectoryName : "";
+
+        SerializeDomParam param(url, textEncoding, document, directoryName);
+
+        Element* documentElement = document->documentElement();
+        if (documentElement)
+            buildContentForNode(documentElement, &param);
+
+        encodeAndFlushBuffer(WebPageSerializerClient::CurrentFrameIsFinished, &param, ForceFlush);
+    }
+
+    ASSERT(m_dataBuffer.isEmpty());
+    m_client->didSerializeDataForFrame(KURL(), WebCString("", 0), WebPageSerializerClient::AllFramesAreFinished);
+    return didSerialization;
+}
+
+}  // namespace blink