Support gzip-compressed seed data from the variations server.

components/variations/service/variations_service.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 "components/variations/service/variations_service.h"
 
 #include "base/build_time.h"
 #include "base/command_line.h"
 #include "base/metrics/histogram.h"
 #include "base/metrics/sparse_histogram.h"
 #include "base/prefs/pref_registry_simple.h"
 #include "base/prefs/pref_service.h"
+#include "base/strings/string_util.h"
 #include "base/sys_info.h"
 #include "base/task_runner_util.h"
 #include "base/timer/elapsed_timer.h"
 #include "base/values.h"
 #include "base/version.h"
 #include "components/data_use_measurement/core/data_use_user_data.h"
 #include "components/metrics/metrics_state_manager.h"
 #include "components/network_time/network_time_tracker.h"
 #include "components/pref_registry/pref_registry_syncable.h"
 #include "components/variations/pref_names.h"
@@ skipping 184 matching lines @@
 
 // Returns the header value for |name| from |headers| or an empty string if not
 // set.
 std::string GetHeaderValue(const net::HttpResponseHeaders* headers,
                            const base::StringPiece& name) {
   std::string value;
   headers->EnumerateHeader(NULL, name, &value);
   return value;
 }
 
+// Returns the list of values for |name| from |headers|. If the header in not
+// set, return an empty list.
+std::vector<std::string> GetHeaderValuesList(
+    const net::HttpResponseHeaders* headers, const base::StringPiece& name) {

[Alexei Svitkine (slow), 2015/10/16 15:16:44]

Nit: Prefer 1 param per line if the first one wraps.

[veranika, 2015/10/16 21:01:00]

Done.

+  std::vector<std::string> values;
+  void* iter = NULL;
+  std::string value;
+  while (headers->EnumerateHeader(&iter, name, &value)) {
+    values.push_back(value);
+  }
+  return values;
+}
+
+// Looks for delta and gzip compression instance manipulation flags set by the
+// server in |headers|. Checks the order of flags and presence of unknown
+// instance manipulations. If successful, |is_delta_compressed| and
+// |is_gzip_compressed| contain compression flags and true is returned.
+bool GetInstanceManipulations(const net::HttpResponseHeaders* headers,
+                              bool* is_delta_compressed,
+                              bool* is_gzip_compressed) {
+  std::vector<std::string> ims = GetHeaderValuesList(headers, "IM");
+  const auto delta_im = std::find(ims.begin(), ims.end(), "x-bm");
+  const auto gzip_im = std::find(ims.begin(), ims.end(), "gzip");
+  *is_delta_compressed = delta_im != ims.end();
+  *is_gzip_compressed = gzip_im != ims.end();
+
+  // The IM field should not have anything but x-bm and gzip.
+  size_t im_count = (*is_delta_compressed ? 1 : 0) +
+      (*is_gzip_compressed ? 1 : 0);
+  if (im_count != ims.size()) {
+    DVLOG(1) << "Unrecognized instance manipulations in "
+             << base::JoinString(ims, ",")
+             << "; only x-bm and gzip are supported";
+    return false;
+  }
+
+  // The IM field defines order in which instance manipulations were applied.
+  // The client requests and supports gzip-compressed delta-compressed seeds,
+  // but not vice versa.
+  if (*is_delta_compressed && *is_gzip_compressed && delta_im > gzip_im) {
+    DVLOG(1) << "Unsupported instance manipulations order: "
+             << "requested x-bm,gzip but received gzip,x-bm";
+    return false;
+  }
+
+  return true;
+}
+
 }  // namespace
 
 VariationsService::VariationsService(
     scoped_ptr<VariationsServiceClient> client,
     scoped_ptr<web_resource::ResourceRequestAllowedNotifier> notifier,
     PrefService* local_state,
     metrics::MetricsStateManager* state_manager,
     const UIStringOverrider& ui_string_overrider)
     : client_(client.Pass()),
       ui_string_overrider_(ui_string_overrider),
@@ skipping 240 matching lines @@
 
   pending_seed_request_ = net::URLFetcher::Create(0, variations_server_url_,
                                                   net::URLFetcher::GET, this);
   data_use_measurement::DataUseUserData::AttachToFetcher(
       pending_seed_request_.get(),
       data_use_measurement::DataUseUserData::VARIATIONS);
   pending_seed_request_->SetLoadFlags(net::LOAD_DO_NOT_SEND_COOKIES |
                                       net::LOAD_DO_NOT_SAVE_COOKIES);
   pending_seed_request_->SetRequestContext(client_->GetURLRequestContext());
   pending_seed_request_->SetMaxRetriesOn5xx(kMaxRetrySeedFetch);
+  bool enable_deltas = false;
   if (!seed_store_.variations_serial_number().empty() &&
       !disable_deltas_for_next_request_) {
     // If the current seed includes a country code, deltas are not supported (as
     // the serial number doesn't take into account the country code). The server
     // will update us with a seed that doesn't include a country code which will
     // enable deltas to work.
     // TODO(asvitkine): Remove the check in M50+ when the percentage of clients
     // that have an old seed with a country code becomes miniscule.
     if (!seed_store_.seed_has_country_code()) {
       // Tell the server that delta-compressed seeds are supported.
-      pending_seed_request_->AddExtraRequestHeader("A-IM:x-bm");
+      enable_deltas = true;
     }
     // Get the seed only if its serial number doesn't match what we have.
     pending_seed_request_->AddExtraRequestHeader(
         "If-None-Match:" + seed_store_.variations_serial_number());
   }
+  // Tell the server that delta-compressed and gzipped seeds are supported.
+  const char * supported_im = enable_deltas ? "A-IM:x-bm,gzip" : "A-IM:gzip";

[Alexei Svitkine (slow), 2015/10/16 15:16:44]

Nit: No space between char and *.

[veranika, 2015/10/16 21:01:00]

Done.

+  pending_seed_request_->AddExtraRequestHeader(supported_im);
+
   pending_seed_request_->Start();
 
   const base::TimeTicks now = base::TimeTicks::Now();
   base::TimeDelta time_since_last_fetch;
   // Record a time delta of 0 (default value) if there was no previous fetch.
   if (!last_request_started_time_.is_null())
     time_since_last_fetch = now - last_request_started_time_;
   UMA_HISTOGRAM_CUSTOM_COUNTS("Variations.TimeSinceLastFetchAttempt",
                               time_since_last_fetch.InMinutes(), 0,
                               base::TimeDelta::FromDays(7).InMinutes(), 50);
   UMA_HISTOGRAM_COUNTS_100("Variations.RequestCount", request_count_);
   ++request_count_;
   last_request_started_time_ = now;
   disable_deltas_for_next_request_ = false;
 }
 
 bool VariationsService::StoreSeed(const std::string& seed_data,
                                   const std::string& seed_signature,
                                   const std::string& country_code,
                                   const base::Time& date_fetched,
-                                  bool is_delta_compressed) {
+                                  bool is_delta_compressed,
+                                  bool is_gzip_compressed) {
   DCHECK(thread_checker_.CalledOnValidThread());
 
   scoped_ptr<variations::VariationsSeed> seed(new variations::VariationsSeed);
   if (!seed_store_.StoreSeedData(seed_data, seed_signature, country_code,
                                  date_fetched, is_delta_compressed,
-                                 seed.get())) {
+                                 is_gzip_compressed, seed.get())) {
     return false;
   }
   RecordLastFetchTime();
 
   // Perform seed simulation only if |state_manager_| is not-NULL. The state
   // manager may be NULL for some unit tests.
   if (!state_manager_)
     return true;
 
   base::PostTaskAndReplyWithResult(
@@ skipping 92 matching lines @@
     }
     return;
   }
 
   std::string seed_data;
   bool success = request->GetResponseAsString(&seed_data);
   DCHECK(success);
 
   net::HttpResponseHeaders* headers = request->GetResponseHeaders();
   const std::string signature = GetHeaderValue(headers, "X-Seed-Signature");
   const std::string country_code = GetHeaderValue(headers, "X-Country");

[Alexei Svitkine (slow), 2015/10/16 15:16:44]

Nit: Can you move these two lines to right above the StoreSeed() call that uses
them?

[veranika, 2015/10/16 21:01:00]

Done.

-  const bool is_delta_compressed = (GetHeaderValue(headers, "IM") == "x-bm");
+
+  bool is_delta_compressed;
+  bool is_gzip_compressed;
+  if (!GetInstanceManipulations(headers, &is_delta_compressed,
+                                &is_gzip_compressed)) {
+    // Details of errors are logged by GetInstanceManipulations, just return.

[Alexei Svitkine (slow), 2015/10/16 15:16:44]

What I'd like in this comment is why it's okay to return - i.e. that the
instance manipulations specified in the response do not match any of our
supported formats.

[veranika, 2015/10/16 21:01:00]

Done.

+    return;

[Alexei Svitkine (slow), 2015/10/16 15:16:44]

Also, it would be good to log this case in a histogram.

Instead of adding a new histogram, can you add an extra case for this in
Variations.SeedStoreResult histogram? Then, you can expose a method on the
SeedStore to log this specific enum. (i.e. the enum def can still stay in the
seed store .cc file, but you can have a method for logging this specific case)

[veranika, 2015/10/16 21:01:00]

Done.

+  }
+
   const bool store_success = StoreSeed(seed_data, signature, country_code,
-                                       response_date, is_delta_compressed);
+                                       response_date, is_delta_compressed,
+                                       is_gzip_compressed);
   if (!store_success && is_delta_compressed) {
     disable_deltas_for_next_request_ = true;
     request_scheduler_->ScheduleFetchShortly();
   }
 }
 
 void VariationsService::OnResourceRequestsAllowed() {
   DCHECK(thread_checker_.CalledOnValidThread());
 
   // Note that this only attempts to fetch the seed at most once per period
@@ skipping 141 matching lines @@
   std::string stored_country;
 
   if (list_value->GetSize() == 2) {
     list_value->GetString(1, &stored_country);
   }
 
   return stored_country;
 }
 
 }  // namespace variations