recognize and parse Schema.org Markup

src/com/dom_distiller/client/SchemaOrgParserAccessor.java

+// Copyright 2014 The Chromium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+package com.dom_distiller.client;
+
+import java.util.ArrayList;
+import java.util.List;
+
+import com.google.gwt.dom.client.Element;
+
+/**
+ * This class instantiates SchemaOrgParser and implements MarkupParser.Parser interface to provide
+ * access to properties that SchemaOrgParser has parsed.
+ */
+public class SchemaOrgParserAccessor implements MarkupParser.Parser {
+    private final SchemaOrgParser parser;
+
+    /**
+     * The object that instantiates SchemaOrgParser and implements its MarkupParser.Parser
+     * interface.
+     */
+    public SchemaOrgParserAccessor(Element root) {
+        parser = new SchemaOrgParser(root);
+    }
+
+    @Override
+    public String getTitle() {
+        String title = parser.findStringProperty(SchemaOrgParser.HEADLINE_PROP);
+        // If there's no "headline" property, use "name" property of first article.
+        if (title.isEmpty()) {
+            SchemaOrgParser.ThingItem item = parser.findFirstArticle();
+            if (item != null) title = item.getStringProperty(SchemaOrgParser.NAME_PROP);
+        }
+        return title;
+    }
+
+    @Override
+    public String getType() {
+        // TODO(kuan): consolidate/standardize types returned from all 3 parsers in MarkupParser.
+        // Returns ARTICLe if there's an article.
+        return parser.findFirstArticle() != null ? SchemaOrgParser.Type.ARTICLE.toString() : "";
+    }
+
+    @Override
+    public String getUrl() {
+        SchemaOrgParser.ThingItem item = parser.findFirstArticle();
+        return item != null ? item.getStringProperty(SchemaOrgParser.URL_PROP) : "";
+    }
+
+    @Override
+    public MarkupParser.Image[] getImages() {
+        List<SchemaOrgParser.ThingItem> itemScopes = parser.getItemScopes();
+        if (itemScopes.isEmpty()) return null;
+
+        List<MarkupParser.Image> images = new ArrayList<MarkupParser.Image>();
+        boolean hasRepresentativeImage = false;
+        MarkupParser.Image imageOfArticle = null;
+
+        for (int i = 0; i < itemScopes.size(); i++) {
+            SchemaOrgParser.ThingItem item = itemScopes.get(i);
+            MarkupParser.Image image = item.getImage();
+            if (image == null) continue;
+            // If |image| is from an article with the "image" property, remember it for now;
+            // it'll be added to to the list later when its position in the list can be determined.
+            if (imageOfArticle == null && item.getType() == SchemaOrgParser.Type.ARTICLE) {
+                imageOfArticle = image;
+                continue;
+            }
+            // Otherwise, |image| is from an ImageObject, insert it at beginning of list if it's
+            // the first image that's representative of page.
+            if (!hasRepresentativeImage && item.isImageRepresentativeOfPage()) {
+                hasRepresentativeImage = true;
+                // Image should be the dominant, i.e. first, one.
+                images.add(0, image);
+            } else {
+                images.add(image);
+            }
+        }
+
+        // Prepend |imageOfArticle| to list if there's no image representative of page; append it
+        // otherwise.
+        if (imageOfArticle != null) {
+            if (!hasRepresentativeImage) images.add(0, imageOfArticle);
+            else images.add(imageOfArticle);
+        }
+
+        if (images.isEmpty()) return null;
+
+        return images.toArray(new MarkupParser.Image[images.size()]);
+    }
+
+    @Override
+    public String getDescription() {
+        SchemaOrgParser.ThingItem item = parser.findFirstArticle();
+        return item != null ? item.getStringProperty(SchemaOrgParser.DESCRIPTION_PROP) : "";
+    }
+
+    @Override
+    public String getPublisher() {
+        SchemaOrgParser.ThingItem item = parser.findFirstArticle();
+        if (item == null) return "";
+        String publisher = item.getStringProperty(SchemaOrgParser.PUBLISHER_PROP);
+        // If there's no "publisher" property, use "copyrightHolder" property of first article.
+        if (publisher.isEmpty()) {
+            publisher = item.getStringProperty(SchemaOrgParser.COPYRIGHT_HOLDER_PROP);
+        }
+        return publisher;
+    }
+
+    @Override
+    public String getCopyright() {
+        // Returns a concatenated string of copyright year and copyright holder of the first article
+        // that has these properties, delimited by a whitespace.
+        SchemaOrgParser.ThingItem item = parser.findFirstArticle();
+        if (item == null) return "";
+        String copyright = SchemaOrgParser.concat(
+                item.getStringProperty(SchemaOrgParser.COPYRIGHT_YEAR_PROP),
+                item.getStringProperty(SchemaOrgParser.COPYRIGHT_HOLDER_PROP));
+        return copyright.isEmpty() ? copyright : "Copyright " + copyright;
+    }
+
+    @Override
+    public String getAuthor() {
+        String author = "";
+        SchemaOrgParser.ThingItem item = parser.findFirstArticle();
+        if (item != null) {
+             author = item.getStringProperty(SchemaOrgParser.AUTHOR_PROP);
+             // If there's no "author" property, use "creator" property.
+             if (author.isEmpty()) author = item.getStringProperty(SchemaOrgParser.CREATOR_PROP);
+        }
+        // Otherwise, use "rel=author" tag.
+        return author.isEmpty() ? parser.getAuthorFromRel() : author;
+    }
+
+    @Override
+    public MarkupParser.Article getArticle() {
+        SchemaOrgParser.ThingItem item = parser.findFirstArticle();
+        return item != null ? item.getArticle() : null;
+    }
+
+    @Override
+    public boolean optOut() {
+        return false;
+    }
+}