Added anonymizer to SystemLogUploader.

chrome/browser/chromeos/policy/system_log_uploader.cc

Index: chrome/browser/chromeos/policy/system_log_uploader.cc
diff --git a/chrome/browser/chromeos/policy/system_log_uploader.cc b/chrome/browser/chromeos/policy/system_log_uploader.cc
index bb68674f1ae547bbfaaed5a604ecd28242df19b6..b3a33a1162aa1122235c8b2b95eb98621511c4d1 100644
--- a/chrome/browser/chromeos/policy/system_log_uploader.cc
+++ b/chrome/browser/chromeos/policy/system_log_uploader.cc
@@ -26,7 +26,6 @@
 #include "components/policy/core/common/cloud/enterprise_metrics.h"
 #include "content/public/browser/browser_thread.h"
 #include "net/http/http_request_headers.h"
-#include "third_party/re2/src/re2/re2.h"
 
 namespace {
 // The maximum number of successive retries.
@@ -44,34 +43,6 @@ const char* const kSystemLogFileNames[] = {
     "/var/log/net.log",       "/var/log/net.1.log",
     "/var/log/ui/ui.LATEST",  "/var/log/update_engine.log"};
 
-const char kEmailAddress[] =
-    "[a-zA-Z0-9\\+\\.\\_\\%\\-\\+]{1,256}\\@"
-    "[a-zA-Z0-9][a-zA-Z0-9\\-]{0,64}(\\.[a-zA-Z0-9][a-zA-Z0-9\\-]{0,25})+";
-const char kIPAddress[] =
-    "((25[0-5]|2[0-4][0-9]|[0-1][0-9]{2}|[1-9][0-9]|[1-9])"
-    "\\.(25[0-5]|2[0-4][0-9]|[0-1][0-9]{2}|[1-9][0-9]|[1-9]|0)\\.(25[0-5]|2"
-    "[0-4][0-9]|[0-1][0-9]{2}|[1-9][0-9]|[1-9]|0)\\.(25[0-5]|2[0-4][0-9]|[0-1]"
-    "[0-9]{2}|[1-9][0-9]|[0-9]))";
-const char kIPv6Address[] =
-    "(([0-9a-fA-F]{1,4}:){7,7}[0-9a-fA-F]{1,4}|"
-    "([0-9a-fA-F]{1,4}:){1,7}:|"
-    "([0-9a-fA-F]{1,4}:){1,6}:[0-9a-fA-F]{1,4}|"
-    "([0-9a-fA-F]{1,4}:){1,5}(:[0-9a-fA-F]{1,4}){1,2}|"
-    "([0-9a-fA-F]{1,4}:){1,4}(:[0-9a-fA-F]{1,4}){1,3}|"
-    "([0-9a-fA-F]{1,4}:){1,3}(:[0-9a-fA-F]{1,4}){1,4}|"
-    "([0-9a-fA-F]{1,4}:){1,2}(:[0-9a-fA-F]{1,4}){1,5}|"
-    "[0-9a-fA-F]{1,4}:((:[0-9a-fA-F]{1,4}){1,6})|"
-    ":((:[0-9a-fA-F]{1,4}){1,7}|:)|"
-    "fe80:(:[0-9a-fA-F]{0,4}){0,4}%[0-9a-zA-Z]{1,}|"
-    "::(ffff(:0{1,4}){0,1}:){0,1}"
-    "((25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])\\.){3,3}"
-    "(25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])|"
-    "([0-9a-fA-F]{1,4}:){1,4}:"
-    "((25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9])\\.){3,3}"
-    "(25[0-5]|(2[0-4]|1{0,1}[0-9]){0,1}[0-9]))";
-
-const char kWebUrl[] = "(http|https|Http|Https|rtsp|Rtsp):\\/\\/";
-
 // Reads the system log files as binary files, stores the files as pairs
 // (file name, data) and returns. Called on blocking thread.
 scoped_ptr<policy::SystemLogUploader::SystemLogs> ReadFiles() {
@@ -85,8 +56,7 @@ scoped_ptr<policy::SystemLogUploader::SystemLogs> ReadFiles() {
       LOG(ERROR) << "Failed to read the system log file from the disk "
                  << file_path << std::endl;
     }
-    system_logs->push_back(std::make_pair(
-        file_path, policy::SystemLogUploader::RemoveSensitiveData(data)));
+    system_logs->push_back(std::make_pair(file_path, data));

[Thiemo Nagel, 2016/01/21 14:22:29]

Why have you moved the anonymizing step to UploadSystemLogs()?  In my opinion,
it is safer to anonymize as soon as possible.  (This reduces the chance of
introducing errors in future changes to the code.)

[Polina Bondarenko, 2016/01/21 16:16:08]

Yes, you're right, I like mostly the idea of moving AnonymizerTool and all
anonymization to ReadFiles.

   }
   return system_logs;
 }
@@ -155,11 +125,6 @@ base::TimeDelta GetUploadFrequency() {
   return upload_frequency;
 }
 
-void RecordSystemLogPIILeak(policy::SystemLogPIIType type) {
-  UMA_HISTOGRAM_ENUMERATION(policy::kMetricSystemLogPII, type,
-                            policy::SYSTEM_LOG_PII_TYPE_SIZE);
-}
-
 std::string GetUploadUrl() {
   return policy::BrowserPolicyConnector::GetDeviceManagementUrl() +
          kSystemLogUploadUrlTail;
@@ -247,47 +212,11 @@ void SystemLogUploader::OnFailure(UploadJob::ErrorCode error_code) {
   }
 }
 
-// static
 std::string SystemLogUploader::RemoveSensitiveData(const std::string& data) {
-  std::string result = "";
-  RE2 email_pattern(kEmailAddress), ipv4_pattern(kIPAddress),
-      ipv6_pattern(kIPv6Address), url_pattern(kWebUrl);
-
-  for (const std::string& line : base::SplitString(
-           data, "\n", base::KEEP_WHITESPACE, base::SPLIT_WANT_ALL)) {
-    // Email.
-    if (RE2::PartialMatch(line, email_pattern)) {
-      RecordSystemLogPIILeak(SYSTEM_LOG_PII_TYPE_EMAIL_ADDRESS);
-      continue;
-    }
-
-    // IPv4 address.
-    if (RE2::PartialMatch(line, ipv4_pattern)) {
-      RecordSystemLogPIILeak(SYSTEM_LOG_PII_TYPE_IP_ADDRESS);
-      continue;
-    }
-
-    // IPv6 address.
-    if (RE2::PartialMatch(line, ipv6_pattern)) {
-      RecordSystemLogPIILeak(SYSTEM_LOG_PII_TYPE_IP_ADDRESS);
-      continue;
-    }
-
-    // URL.
-    if (RE2::PartialMatch(line, url_pattern)) {
-      RecordSystemLogPIILeak(SYSTEM_LOG_PII_TYPE_WEB_URL);
-      continue;
-    }
-
-    // SSID.
-    if (line.find("SSID=") != std::string::npos) {
-      RecordSystemLogPIILeak(SYSTEM_LOG_PII_TYPE_SSID);
-      continue;
-    }
-
-    result += line + "\n";
-  }
-  return result;
+  std::string anonymized = anonymizer_.Anonymize(data);
+  UMA_HISTOGRAM_BOOLEAN(policy::kMetricSystemLogPII,
+                        anonymized.compare(data) != 0);
+  return anonymized;
 }
 
 void SystemLogUploader::RefreshUploadSettings() {
@@ -322,8 +251,8 @@ void SystemLogUploader::UploadSystemLogs(scoped_ptr<SystemLogs> system_logs) {
   int file_number = 1;
   for (const auto& syslog_entry : *system_logs) {
     std::map<std::string, std::string> header_fields;
-    scoped_ptr<std::string> data =
-        make_scoped_ptr(new std::string(syslog_entry.second));
+    scoped_ptr<std::string> data = make_scoped_ptr(
+        new std::string(RemoveSensitiveData(syslog_entry.second)));
     header_fields.insert(std::make_pair(kFileTypeHeaderName, kFileTypeLogFile));
     header_fields.insert(std::make_pair(net::HttpRequestHeaders::kContentType,
                                         kContentTypePlainText));