Extract image URLs in srcset as well

java/org/chromium/distiller/ContentExtractor.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 org.chromium.distiller;
 
 import org.chromium.distiller.document.TextDocument;
 import org.chromium.distiller.document.TextDocumentStatistics;
 import org.chromium.distiller.extractors.ArticleExtractor;
 import org.chromium.distiller.proto.DomDistillerProtos.StatisticsInfo;
@@ skipping 16 matching lines @@
 import java.util.LinkedList;
 import java.util.List;
 import java.util.Set;
 
 public class ContentExtractor {
     private final Element documentElement;
     private final List<String> candidateTitles;
     private final TimingInfo mTimingInfo;
     private final StatisticsInfo mStatisticsInfo;
     private final MarkupParser parser;
-    private final List<String> imageUrls;
+    private List<String> imageUrls;
     private String textDirection;
 
     private class WebDocumentInfo {
         WebDocument document;
         Set<Node> hiddenElements;
     }
 
     public ContentExtractor(Element root) {
         documentElement = root;
         candidateTitles = new LinkedList<String>();
         mTimingInfo = TimingInfo.create();
         mStatisticsInfo = StatisticsInfo.create();
-        imageUrls = new ArrayList<String>();
 
         double startTime = DomUtil.getTime();
         parser = new MarkupParser(root, mTimingInfo);
         mTimingInfo.setMarkupParsingTime(DomUtil.getTime() - startTime);
         textDirection = "";
     }
 
     // Grabs a list of candidate titles in descending priority order:
     // 1) meta-information
     // 2) The document's title element, modified based on some readability heuristics
@@ skipping 28 matching lines @@
         double now = DomUtil.getTime();
         WebDocumentInfo documentInfo = createWebDocumentInfoFromPage();
         mTimingInfo.setDocumentConstructionTime(DomUtil.getTime() - now);
 
         now = DomUtil.getTime();
         processDocument(documentInfo.document);
         RelevantElements.process(documentInfo.document);
         LeadImageFinder.process(documentInfo.document);
         NestedElementRetainer.process(documentInfo.document);
 
-        List<WebImage> images = documentInfo.document.getContentImages();
-        for (WebImage wi : images) {
-            imageUrls.add(wi.getSrc());
-        }
         mTimingInfo.setArticleProcessingTime(DomUtil.getTime() - now);
 
         now = DomUtil.getTime();
         String html = documentInfo.document.generateOutput(textOnly);
         mTimingInfo.setFormattingTime(DomUtil.getTime() - now);
 
+        imageUrls = documentInfo.document.getImageUrls();
+
         if (LogUtil.isLoggable(LogUtil.DEBUG_LEVEL_TIMING_INFO)) {
             for (int i = 0; i < mTimingInfo.getOtherTimesCount(); i++) {
                 TimingEntry entry =  mTimingInfo.getOtherTimes(i);
                 LogUtil.logToConsole("Timing: " + entry.getName() + " = " + entry.getTime());
             }
 
             LogUtil.logToConsole(
                     "Timing: MarkupParsingTime = " +
                     mTimingInfo.getMarkupParsingTime() +
                     "\nTiming: DocumentConstructionTime = " +
@@ skipping 67 matching lines @@
      *
      * @param document the WebDocument representation of the page extracted from the DOM.
      */
     private void processDocument(WebDocument document) {
         TextDocument textDocument = document.createTextDocumentView();
         ArticleExtractor.INSTANCE.process(textDocument, candidateTitles);
         mStatisticsInfo.setWordCount(TextDocumentStatistics.countWordsInContent(textDocument));
         textDocument.applyToModel();
     }
 }