Added a histogram to track number of http/https resources.

content/browser/renderer_host/duplicate_content_resource_handler.cc

 // 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_content_resource_handler.h"
 
 #include <set>
 
 #include "base/lazy_instance.h"
 #include "base/logging.h"
@@ skipping 74 matching lines @@
   if (status.is_success())
     RecordContentMetrics();
 
   return next_handler_->OnResponseCompleted(request_id, status, security_info);
 }
 
 void DuplicateContentResourceHandler::RecordContentMetrics() {
   MH_UINT32 contents_hash = PMurHash32_Result(pmurhash_ph1_,
                                               pmurhash_pcarry_, bytes_read_);
 
+  bool is_http_or_https = request_->url().SchemeIs("http") ||

[darin (slow to review), 2012/08/11 06:18:36]

weren't you going to modify this to stop recording metrics for non-http(s) URLs?
 or otherwise put a cap on the memory leak?

+                          request_->url().SchemeIs("https");
+  UMA_HISTOGRAM_BOOLEAN("Duplicate.IsHttpOrHttps", is_http_or_https);
+
   // Combine the contents_hash with the url, so we can test if future content
   // identical resources have the same original url or not.
   MH_UINT32 hashed_with_url;
   const std::string& url_spec = request_->url().spec();
   PMurHash32_Process(&pmurhash_ph1_, &pmurhash_pcarry_,
                      url_spec.data(), url_spec.length());
   hashed_with_url = PMurHash32_Result(pmurhash_ph1_, pmurhash_pcarry_,
                                       url_spec.length() + bytes_read_);
 
   DVLOG(4) << "url: " << url_spec;
   DVLOG(4) << "contents hash: " << contents_hash;
   DVLOG(4) << "hash with url: " << hashed_with_url;
 
   std::set<MH_UINT32>* content_matches =
       GlobalDuplicateRecords::GetInstance()->content_matches();
   std::set<MH_UINT32>* content_and_url_matches =
       GlobalDuplicateRecords::GetInstance()->content_and_url_matches();
 
   const bool did_match_contents = content_matches->count(contents_hash) > 0;
   const bool did_match_contents_and_url =
       content_and_url_matches->count(hashed_with_url) > 0;
 
   UMA_HISTOGRAM_BOOLEAN("Duplicate.Hits", did_match_contents);
   UMA_HISTOGRAM_BOOLEAN("Duplicate.HitsSameUrl",
                          did_match_contents && did_match_contents_and_url);
+  UMA_HISTOGRAM_ENUMERATION("Duplicate.ResourceType.All", resource_type_,

[darin (slow to review), 2012/08/11 06:18:36]

this was part of the other CL, right?

+                             ResourceType::LAST_TYPE);
   if (did_match_contents && !did_match_contents_and_url) {
     content_and_url_matches->insert(hashed_with_url);
     UMA_HISTOGRAM_CUSTOM_COUNTS("Duplicate.Size.HashHitUrlMiss", bytes_read_,
                                  1, 0x7FFFFFFF, 50);
     UMA_HISTOGRAM_ENUMERATION("Duplicate.ResourceType.HashHitUrlMiss",
                                resource_type_, ResourceType::LAST_TYPE);
   }
   content_matches->insert(contents_hash);
   content_and_url_matches->insert(hashed_with_url);
 
   bytes_read_ = 0;
   read_buffer_ = NULL;
 }
 
 }  // namespace content