Fixes Issue 7377: Regression: Omnibox trims URL ending with 0x85...

base/string_util.cc

 // Copyright (c) 2006-2008 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 "base/string_util.h"
 
 #include "build/build_config.h"
 
 #include <ctype.h>
 #include <errno.h>
@@ skipping 319 matching lines @@
   0x3000,  // Ideographic Space
   0
 };
 const char kWhitespaceASCII[] = {
   0x09,    // <control-0009> to <control-000D>
   0x0A,
   0x0B,
   0x0C,
   0x0D,
   0x20,    // Space
-  '\x85',  // <control-0085>
-  '\xa0',  // No-Break Space
   0
 };
 const char* const kCodepageUTF8 = "UTF-8";
 
 template<typename STR>
 TrimPositions TrimStringT(const STR& input,
                           const typename STR::value_type trim_chars[],
                           TrimPositions positions,
                           STR* output) {
   // Find the edges of leading/trailing whitespace as desired.
@@ skipping 34 matching lines @@
                 std::string* output) {
   return TrimStringT(input, trim_chars, TRIM_ALL, output) != TRIM_NONE;
 }
 
 TrimPositions TrimWhitespace(const std::wstring& input,
                              TrimPositions positions,
                              std::wstring* output) {
   return TrimStringT(input, kWhitespaceWide, positions, output);
 }
 
+TrimPositions TrimWhitespaceASCII(const std::string& input,
+                                  TrimPositions positions,
+                                  std::string* output) {
+  return TrimStringT(input, kWhitespaceASCII, positions, output);
+}
+
+TrimPositions TrimWhitespaceUTF8(const std::string& input,
+                                 TrimPositions positions,
+                                 std::string* output) {
+  // This implementation is not so fast since it converts the text encoding
+  // twice. Please feel free to file a bug if this function hurts the
+  // performance of Chrome.
+  DCHECK(IsStringUTF8(input));
+  std::wstring input_wide = UTF8ToWide(input);
+  std::wstring output_wide;
+  TrimPositions result = TrimWhitespace(input_wide, positions, &output_wide);
+  *output = WideToUTF8(output_wide);
+  return result;
+}
+
+// This function is only for backward-compatibility.
+// To be removed when all callers are updated.
 TrimPositions TrimWhitespace(const std::string& input,
                              TrimPositions positions,
                              std::string* output) {
-  return TrimStringT(input, kWhitespaceASCII, positions, output);
+  return TrimWhitespaceASCII(input, positions, output);
 }
 
 std::wstring CollapseWhitespace(const std::wstring& text,
                                 bool trim_sequences_with_line_breaks) {
   std::wstring result;
   result.resize(text.size());
 
   // Set flags to pretend we're already in a trimmed whitespace sequence, so we
   // will trim any leading whitespace.
   bool in_whitespace = true;
@@ skipping 1196 matching lines @@
   // Each input byte creates two output hex characters.
   std::string ret(size * 2, '\0');
 
   for (size_t i = 0; i < size; ++i) {
     char b = reinterpret_cast<const char*>(bytes)[i];
     ret[(i * 2)] = kHexChars[(b >> 4) & 0xf];
     ret[(i * 2) + 1] = kHexChars[b & 0xf];
   }
   return ret;
 }