DuplicateContentResourceHandler to monitor resources and track how many times th…

content/browser/renderer_host/duplicate_resource_handler.cc

Index: content/browser/renderer_host/duplicate_resource_handler.cc
===================================================================
--- content/browser/renderer_host/duplicate_resource_handler.cc	(revision 0)
+++ content/browser/renderer_host/duplicate_resource_handler.cc	(revision 0)
@@ -0,0 +1,83 @@
+// Copyright (c) 2012 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.
+
+#include "content/browser/renderer_host/duplicate_resource_handler.h"
+
+#include <string>
+
+#include "base/logging.h"
+#include "base/metrics/histogram.h"
+#include "content/browser/renderer_host/resource_request_info_impl.h"
+#include "net/base/io_buffer.h"
+#include "third_party/smhasher/src/MurmurHash3.h"
+
+
+namespace content {
+
+namespace{
+ 
+std::set<uint32>* GetSetOfHashes() {
+  static std::set<uint32> seen_resources;
+  return &seen_resources;
+}
+
+}  // namespace
+
+DuplicateResourceHandler::DuplicateResourceHandler(
+    scoped_ptr<ResourceHandler> next_handler,
+    net::URLRequest* request)
+    : LayeredResourceHandler(next_handler.Pass()),
+      bytes_hit_(0),
+      bytes_miss_(0),
+      read_buffer_size_(0),  // keep track of bytes in read buffer
+      request_(request) {
+}
+
+DuplicateResourceHandler::~DuplicateResourceHandler() {
+}
+
+bool DuplicateResourceHandler::OnWillRead(int request_id, net::IOBuffer** buf, 
+                                        int* buf_size, int min_size) {
+  DCHECK_EQ(-1, min_size);
+  
+  if (next_handler_->OnWillRead(request_id, buf, buf_size, min_size)) {
+    read_buffer_ = *buf; 
+    read_buffer_size_ = *buf_size;
+    return true;
+  }
+  
+  return false;
+}
+
+bool DuplicateResourceHandler::OnReadCompleted(int request_id, int bytes_read,
+                                              bool* defer) {
+  if (!next_handler_->OnReadCompleted(request_id, bytes_read, defer))
+    return false;
+  
+  DVLOG(4) << "read_buffer_ is " << read_buffer_.get();
+  DVLOG(4) << "read_buffer_->data() is " << read_buffer_->data();
+
+  // find hash of resource
+  uint32 buf_hash;
+  MurmurHash3_x86_32(read_buffer_->data(), bytes_read, 0x0, &buf_hash);

[gavinp, 2012/07/13 21:16:04]

0, but in hexadecimal? Just use 0.

If it helps, you should know that 0 is an octal constant according to the
grammar in the original K&R.

+
+  const bool did_we_find_it = GetSetOfHashes()->find(buf_hash) != GetSetOfHashes()->end();
+  UMA_HISTOGRAM_BOOLEAN("Duplicate.HitRate", did_we_find_it);  
+
+  // if element is in hash

[gavinp, 2012/07/13 21:16:04]

Your comment not full sentence with punctuation

+  if (did_we_find_it) {
+    bytes_hit_ += bytes_read;
+
+  } else {
+    bytes_miss_ += bytes_read;
+    GetSetOfHashes()->insert(buf_hash);
+  }
+  
+  read_buffer_.release();
+  read_buffer_size_ = 0;
+
+  return true;
+}
+
+}  //namespace content