Clean up attributes of image elements

java/org/chromium/distiller/DomUtil.java

Index: java/org/chromium/distiller/DomUtil.java
diff --git a/java/org/chromium/distiller/DomUtil.java b/java/org/chromium/distiller/DomUtil.java
index d0bcb1bfd7330c1cc8c56be59fdbc6320918284f..f757cb76c02f73df586e8c393456a1120f303113 100644
--- a/java/org/chromium/distiller/DomUtil.java
+++ b/java/org/chromium/distiller/DomUtil.java
@@ -216,6 +216,7 @@ public class DomUtil {
         stripFontColorAttributes(clonedSubtree);
         stripTableBackgroundColorAttributes(clonedSubtree);
         stripStyleAttributes(clonedSubtree);
+        stripImageElements(clonedSubtree);

[mdjones, 2015/12/10 01:12:45]

It could be argued that images as part of a table should retain their
formatting, but since it looks like we are removing other table attributes, this
is fine.

[wychen, 2015/12/10 05:39:16]

Acknowledged.

 
         if (textOnly) {
             return DomUtil.getTextFromTree(clonedSubtree);
@@ -260,8 +261,14 @@ public class DomUtil {
     }
 
     public static void makeSrcSetAbsolute(ImageElement ie) {
+        String srcset = ie.getAttribute("srcset");
+        if (srcset == "") {
+            ie.removeAttribute("srcset");
+            return;
+        }
+
         String oldsrc = ie.getSrc();
-        String[] sizes = StringUtil.jsSplit(ie.getAttribute("srcset"), ",");
+        String[] sizes = StringUtil.jsSplit(srcset, ",");
         for(int i = 0; i < sizes.length; i++) {
             String size = StringUtil.jsTrim(sizes[i]);
             if (size.isEmpty()) continue;
@@ -274,6 +281,27 @@ public class DomUtil {
         ie.setSrc(oldsrc);
     }
 
+    public static void stripImageElements(Node root) {
+        NodeList<Element> imgs = DomUtil.querySelectorAll(root, "IMG");
+        for (int i = 0; i < imgs.getLength(); i++) {
+            stripImageElement(ImageElement.as(imgs.getItem(i)));
+        }
+    }
+
+    /**
+     * Only keep src and optionally srcset attributes for image elements.
+     * @param ie The image element to strip in-place.
+     */
+    public static void stripImageElement(ImageElement imgElement) {
+        JsArray<Node> attrs = DomUtil.getAttributes(imgElement);
+        for (int i = 0; i < attrs.length(); i++) {
+            String name = attrs.get(i).getNodeName();
+            if (name != "src" && name != "srcset") {
+                imgElement.removeAttribute(name);
+            }
+        }
+    }
+
     /**
      * Makes all "img", "source", "track", and "video" tags have an absolute "src" attribute.
      * @param root The root element to look through.