Match the derived features to the training data set

components/dom_distiller/core/page_features.cc

 // Copyright 2015 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/dom_distiller/core/page_features.h"
 
 #include <stddef.h>
 
 #include <string>
 
 #include "base/json/json_reader.h"
 #include "third_party/re2/src/re2/re2.h"
 #include "url/gurl.h"
 
 namespace dom_distiller {
 /* This code needs to derive features in the same way and order in which they
  * are derived when training the model. Parts of that code are reproduced in the
  * comments below.
  */
 
 namespace {
 
 std::string GetLastSegment(const std::string& path) {
   // return re.search('[^/]*\/?$', path).group(0)
   if (path.size() == 0)
     return "";
-  size_t start = path.rfind("/", path.size() - 1);
+  if (path.size() == 1) {
+    DCHECK(path[0] == '/');
+    return path;
+  }
+  size_t start = path.rfind("/", path.size() - 2);
   return start == std::string::npos ? "" : path.substr(start + 1);
 }
 
 int CountMatches(const std::string& s, const std::string& p) {
   // return len(re.findall(p, s))
   re2::StringPiece sp(s);
   re2::RE2 regexp(p);
   int count = 0;
   while (re2::RE2::FindAndConsume(&sp, regexp))
     count++;
@@ skipping 198 matching lines @@
   features.push_back(mozScore);
   // 'mozScoreAllSqrt'
   features.push_back(mozScoreAllSqrt);
   // 'mozScoreAllLinear'
   features.push_back(mozScoreAllLinear);
 
   return features;
 }
 
 }  // namespace dom_distiller