MetricsService: Send a hash of the UMA log in a header.

chrome/common/metrics/metrics_log_manager.h

 // 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.
 
 #ifndef CHROME_COMMON_METRICS_METRICS_LOG_MANAGER_H_
 #define CHROME_COMMON_METRICS_METRICS_LOG_MANAGER_H_
 
 #include "base/basictypes.h"
 #include "base/memory/scoped_ptr.h"
 
 #include <string>
 #include <vector>
 
 class MetricsLogBase;
 
 // Manages all the log objects used by a MetricsService implementation. Keeps
 // track of both an in progress log and a log that is staged for uploading as
 // text, as well as saving logs to, and loading logs from, persistent storage.
 class MetricsLogManager {
  public:
   MetricsLogManager();
   ~MetricsLogManager();
 
   enum LogType {
     INITIAL_LOG,  // The first log of a session.
     ONGOING_LOG,  // Subsequent logs in a session.
     NO_LOG,       // Placeholder value for when there is no log.
   };
 
+  class SerializedLog {
+   public:
+    SerializedLog();
+    ~SerializedLog();
+
+    const std::string& log_text() const { return log_text_; }
+    const std::string& log_hash() const { return log_hash_; }
+
+    // Returns true if the log is empty.
+    bool IsEmpty() const;
+
+    // Sets the log text and updates the hash.
+    void SetLogText(const std::string& log_text);

[Ilya Sherman, 2013/10/17 01:29:03]

Can we pass |log_text| as a pointer, and swap() the contents?  The code
previously did this for "efficiency"; it might be that the performance
difference actually matters, since the strings can indeed be pretty large.

[Alexei Svitkine (slow), 2013/10/17 19:47:27]

I added a SwapLogText() function that does that now and changed the production
code to use it. I've kept SetLogText() since a lot of the test code sets the
text via a string literal (which doesn't work well when the function takes a
pointer).

[Ilya Sherman, 2013/10/17 22:06:25]

Thanks.  If SetLogText() is only used in tests, can you rename it to something
like SetLogTextForTesting()?  Even better might be to define a wrapper function
in the test file that looks something like this:

void SetLogText(const std::string& log_text, SerializedLog* log) {
  std::string log_text_copy = log_text;
  log->SwapLogText(&log_text_copy);
}

[Alexei Svitkine (slow), 2013/10/18 18:51:06]

Done.

+
+    // Clears the log.
+    void Clear();
+
+    // Swaps log contents with |other|.
+    void Swap(SerializedLog* other);
+
+   private:
+    // Non-human readable log text (serialized proto).
+    std::string log_text_;
+
+    // Non-human readable SHA1 of |log_text| or empty if |log_text| is empty.
+    std::string log_hash_;
+  };

[Ilya Sherman, 2013/10/17 01:29:03]

nit: Please either add a DISALLOW_COPY_AND_ASSIGN, or add an explicit comment
indicating that it's intentionally omitted.

[Alexei Svitkine (slow), 2013/10/17 19:47:27]

Done.

+
   enum StoreType {
     NORMAL_STORE,       // A standard store operation.
     PROVISIONAL_STORE,  // A store operation that can be easily reverted later.
   };
 
   // Takes ownership of |log|, which has type |log_type|, and makes it the
   // current_log. This should only be called if there is not a current log.
   void BeginLoggingWithLog(MetricsLogBase* log, LogType log_type);
 
   // Returns the in-progress log.
@@ skipping 10 matching lines @@
 
   // Populates staged_log_text() with the next stored log to send.
   // Should only be called if has_unsent_logs() is true.
   void StageNextLogForUpload();
 
   // Returns true if there is a log that needs to be, or is being, uploaded.
   bool has_staged_log() const;
 
   // The text of the staged log, as a serialized protobuf. Empty if there is no
   // staged log, or if compression of the staged log failed.
-  const std::string& staged_log_text() const { return staged_log_text_; }
+  const std::string& staged_log_text() const { return staged_log_.log_text(); }
+
+  // The SHA1 hash (non-human readable) of the staged log or empty if there is
+  // no staged log.
+  const std::string& staged_log_hash() const { return staged_log_.log_hash(); }
 
   // Discards the staged log.
   void DiscardStagedLog();
 
   // Closes and discards |current_log|.
   void DiscardCurrentLog();
 
   // Sets current_log to NULL, but saves the current log for future use with
   // ResumePausedLog(). Only one log may be paused at a time.
   // TODO(stuartmorgan): Pause/resume support is really a workaround for a
@@ skipping 27 matching lines @@
   }
 
   // Interface for a utility class to serialize and deserialize logs for
   // persistent storage.
   class LogSerializer {
    public:
     virtual ~LogSerializer() {}
 
     // Serializes |logs| to persistent storage, replacing any previously
     // serialized logs of the same type.
-    virtual void SerializeLogs(const std::vector<std::string>& logs,
+    virtual void SerializeLogs(const std::vector<SerializedLog>& logs,
                                LogType log_type) = 0;
 
     // Populates |logs| with logs of type |log_type| deserialized from
     // persistent storage.
     virtual void DeserializeLogs(LogType log_type,
-                                 std::vector<std::string>* logs) = 0;
+                                 std::vector<SerializedLog>* logs) = 0;
   };
 
   // Sets the serializer to use for persisting and loading logs; takes ownership
   // of |serializer|.
   void set_log_serializer(LogSerializer* serializer) {
     log_serializer_.reset(serializer);
   }
 
   // Saves any unsent logs to persistent storage using the current log
   // serializer. Can only be called after set_log_serializer.
   void PersistUnsentLogs();
 
   // Loads any unsent logs from persistent storage using the current log
   // serializer. Can only be called after set_log_serializer.
   void LoadPersistedUnsentLogs();
 
  private:
-  // Saves |log_text| as the given type (or discards it in accordance with
+  // Saves |log| as the given type (or discards it in accordance with
   // |max_ongoing_log_store_size_|).
-  // NOTE: This clears the contents of |log_text| (to avoid an expensive
-  // string copy), so the log should be discarded after this call.
-  void StoreLog(std::string* log_text,
+  // NOTE: This clears the contents of |log| (to avoid an expensive copy),
+  // so the log should be discarded after this call.
+  void StoreLog(SerializedLog* log,
                 LogType log_type,
                 StoreType store_type);
 
-  // Compresses current_log_ into compressed_log.
-  void CompressCurrentLog(std::string* compressed_log);
+  // Compresses |current_log_| into |compressed_log|.
+  void CompressCurrentLog(SerializedLog* compressed_log);
 
   // The log that we are still appending to.
   scoped_ptr<MetricsLogBase> current_log_;
   LogType current_log_type_;
 
   // A paused, previously-current log.
   scoped_ptr<MetricsLogBase> paused_log_;
   LogType paused_log_type_;
 
   // Helper class to handle serialization/deserialization of logs for persistent
   // storage. May be NULL.
   scoped_ptr<LogSerializer> log_serializer_;
 
-  // The text representations of the staged log, ready for upload to the server.
-  std::string staged_log_text_;
+  // The current staged log, ready for upload to the server.
+  SerializedLog staged_log_;
   LogType staged_log_type_;
 
   // Logs from a previous session that have not yet been sent.
   // Note that the vector has the oldest logs listed first (early in the
   // vector), and we'll discard old logs if we have gathered too many logs.
-  std::vector<std::string> unsent_initial_logs_;
-  std::vector<std::string> unsent_ongoing_logs_;
+  std::vector<SerializedLog> unsent_initial_logs_;
+  std::vector<SerializedLog> unsent_ongoing_logs_;
 
   size_t max_ongoing_log_store_size_;
 
   // The index and type of the last provisional store. If nothing has been
   // provisionally stored, or the last provisional store has already been
   // re-staged, the index will be -1;
   // This is necessary because during an upload there are two logs (staged
   // and current) and a client might store them in either order, so it's
   // not necessarily the case that the provisional store is the last store.
   int last_provisional_store_index_;
   LogType last_provisional_store_type_;
 
   DISALLOW_COPY_AND_ASSIGN(MetricsLogManager);
 };
 
 #endif  // CHROME_COMMON_METRICS_METRICS_LOG_MANAGER_H_