Add support for Schema.org/Recipe

java/org/chromium/distiller/SchemaOrgParserAccessor.java

Index: java/org/chromium/distiller/SchemaOrgParserAccessor.java
diff --git a/java/org/chromium/distiller/SchemaOrgParserAccessor.java b/java/org/chromium/distiller/SchemaOrgParserAccessor.java
index 83779ecea65af6b358c1e9dc6d8d57c231a15df5..351b57cf8ed2cfc8bfaea1b7b3d2efd1a57bb3e4 100644
--- a/java/org/chromium/distiller/SchemaOrgParserAccessor.java
+++ b/java/org/chromium/distiller/SchemaOrgParserAccessor.java
@@ -19,9 +19,11 @@ import java.util.List;
  */
 public class SchemaOrgParserAccessor implements MarkupParser.Accessor {
 
+    private static List<SchemaOrgParser.Type> supportedTypes;

[wychen, 2016/07/24 23:06:34]

I suspect this still causes some slowness.
Would lazy initialization make it faster?

     private SchemaOrgParser mParser;
     private final Element mRoot;
     private final TimingInfo mTimingInfo;
+    private static final String ENGLISH_LANGUAGE = "en";
 
     /**
      * The object that instantiates SchemaOrgParser and implements its MarkupParser.Accessor
@@ -36,6 +38,11 @@ public class SchemaOrgParserAccessor implements MarkupParser.Accessor {
         mTimingInfo = timingInfo;
     }
 
+    static {
+        supportedTypes = new ArrayList<>();
+        supportedTypes.add(SchemaOrgParser.Type.RECIPE);
+    }
+
     private void init() {
         if (mParser == null) {
             mParser = new SchemaOrgParser(mRoot, mTimingInfo);
@@ -184,6 +191,32 @@ public class SchemaOrgParserAccessor implements MarkupParser.Accessor {
     }
 
     /**
+     * Get schema.org parsed content.
+     *
+     * Content is generated when a main entity is identified or
+     * has only one entity item parsed and it is supported
+     *
+     * @return HTML output from the supported item.
+     */
+    @Override
+    public String getStructuredData() {
+        String output = "";
+        if (DomUtil.getLanguage(mRoot).startsWith(ENGLISH_LANGUAGE)) {
+            init();
+            SchemaOrgParser.ThingItem entity = mParser.getMainEntity();
+            List<SchemaOrgParser.ThingItem> topLevelEntities =
+                    mParser.getTopLevelEntities();
+            if (entity == null && topLevelEntities.size() == 1) {
+                entity = topLevelEntities.get(0);
+            }
+            if (entity != null && supportedTypes.contains(entity.getType())) {
+                output = entity.generateOutput();
+            }
+        }
+        return output;
+    }
+
+    /**
      * Sort a list of {@link SchemaOrgParser.ArticleItem}s by their area
      * in descending order.
      * @param articles List of {@link SchemaOrgParser.ArticleItem}s to