OOPIFs: Deduplicating MHTML parts across frames.

third_party/WebKit/Source/web/WebPageSerializer.cpp

Index: third_party/WebKit/Source/web/WebPageSerializer.cpp
diff --git a/third_party/WebKit/Source/web/WebPageSerializer.cpp b/third_party/WebKit/Source/web/WebPageSerializer.cpp
index 48441a08048fc1b8d2bdb290acef211b75a00590..7e9c76be809169ae0af4406036e28583d522581b 100644
--- a/third_party/WebKit/Source/web/WebPageSerializer.cpp
+++ b/third_party/WebKit/Source/web/WebPageSerializer.cpp
@@ -166,16 +166,34 @@ WebData WebPageSerializer::generateMHTMLHeader(
     return buffer.release();
 }
 
+WebVector<WebURL> WebPageSerializer::enumerateMHTMLResources(WebLocalFrame* webFrame)
+{
+    Vector<SerializedResource> resources;
+    PageSerializer serializer(resources, nullptr);
+    serializer.serializeFrame(*toWebLocalFrameImpl(webFrame)->frame());
+
+    Vector<WebURL> result;
+    for (size_t i = 0; i < resources.size(); i++) {
+        const SerializedResource& resource = resources[i];
+        if (i != 0) // Report only resources, not the frame.
+            result.append(resource.url);
+    }
+
+    return result;
+}
+
 WebData WebPageSerializer::generateMHTMLParts(
     const WebString& boundary, WebLocalFrame* webFrame, bool useBinaryEncoding,
-    const WebVector<std::pair<WebFrame*, WebString>>& webFrameToContentID)
+    const WebVector<std::pair<WebFrame*, WebString>>& webFrameToContentID,
+    const WebVector<WebURL>& resourcesToSkip)
 {
     // Translate arguments from public to internal blink APIs.
     LocalFrame* frame = toWebLocalFrameImpl(webFrame)->frame();
     MHTMLArchive::EncodingPolicy encodingPolicy = useBinaryEncoding
         ? MHTMLArchive::EncodingPolicy::UseBinaryEncoding
         : MHTMLArchive::EncodingPolicy::UseDefaultEncoding;
-    ContentIDMap frameToContentID = createFrameToContentIDMap(webFrameToContentID);
+    ContentIDMap frameToContentID =
+        createFrameToContentIDMap(webFrameToContentID);
 
     // Serialize.
     Vector<SerializedResource> resources;
@@ -191,6 +209,10 @@ WebData WebPageSerializer::generateMHTMLParts(
         // comment). Frames need a Content-ID header.
         String contentID = isFirstResource ? frameToContentID.get(frame) : String();
 
+        // Resources don't yet get a Content ID and are deduped based on URI.
+        if (contentID.isEmpty() && resourcesToSkip.contains(resource.url))
+            continue;
+
         MHTMLArchive::generateMHTMLPart(
             boundary, contentID, encodingPolicy, resource, *output);