summaryrefslogtreecommitdiffstats
path: root/base/string_util.h
blob: a13518026a3001c78e4c2af53b41f89b8677db3a (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
// Copyright (c) 2011 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.
//
// This file defines utility functions for working with strings.

#ifndef BASE_STRING_UTIL_H_
#define BASE_STRING_UTIL_H_
#pragma once

#include <ctype.h>
#include <stdarg.h>   // va_list

#include <string>
#include <vector>

#include "base/base_export.h"
#include "base/basictypes.h"
#include "base/compiler_specific.h"
#include "base/string16.h"
#include "base/string_piece.h"  // For implicit conversions.

// Safe standard library wrappers for all platforms.

namespace base {

// C standard-library functions like "strncasecmp" and "snprintf" that aren't
// cross-platform are provided as "base::strncasecmp", and their prototypes
// are listed below.  These functions are then implemented as inline calls
// to the platform-specific equivalents in the platform-specific headers.

// Compares the two strings s1 and s2 without regard to case using
// the current locale; returns 0 if they are equal, 1 if s1 > s2, and -1 if
// s2 > s1 according to a lexicographic comparison.
int strcasecmp(const char* s1, const char* s2);

// Compares up to count characters of s1 and s2 without regard to case using
// the current locale; returns 0 if they are equal, 1 if s1 > s2, and -1 if
// s2 > s1 according to a lexicographic comparison.
int strncasecmp(const char* s1, const char* s2, size_t count);

// Same as strncmp but for char16 strings.
int strncmp16(const char16* s1, const char16* s2, size_t count);

// Wrapper for vsnprintf that always null-terminates and always returns the
// number of characters that would be in an untruncated formatted
// string, even when truncation occurs.
int vsnprintf(char* buffer, size_t size, const char* format, va_list arguments)
    PRINTF_FORMAT(3, 0);

// vswprintf always null-terminates, but when truncation occurs, it will either
// return -1 or the number of characters that would be in an untruncated
// formatted string.  The actual return value depends on the underlying
// C library's vswprintf implementation.
int vswprintf(wchar_t* buffer, size_t size,
              const wchar_t* format, va_list arguments)
    WPRINTF_FORMAT(3, 0);

// Some of these implementations need to be inlined.

// We separate the declaration from the implementation of this inline
// function just so the PRINTF_FORMAT works.
inline int snprintf(char* buffer, size_t size, const char* format, ...)
    PRINTF_FORMAT(3, 4);
inline int snprintf(char* buffer, size_t size, const char* format, ...) {
  va_list arguments;
  va_start(arguments, format);
  int result = vsnprintf(buffer, size, format, arguments);
  va_end(arguments);
  return result;
}

// We separate the declaration from the implementation of this inline
// function just so the WPRINTF_FORMAT works.
inline int swprintf(wchar_t* buffer, size_t size, const wchar_t* format, ...)
    WPRINTF_FORMAT(3, 4);
inline int swprintf(wchar_t* buffer, size_t size, const wchar_t* format, ...) {
  va_list arguments;
  va_start(arguments, format);
  int result = vswprintf(buffer, size, format, arguments);
  va_end(arguments);
  return result;
}

// BSD-style safe and consistent string copy functions.
// Copies |src| to |dst|, where |dst_size| is the total allocated size of |dst|.
// Copies at most |dst_size|-1 characters, and always NULL terminates |dst|, as
// long as |dst_size| is not 0.  Returns the length of |src| in characters.
// If the return value is >= dst_size, then the output was truncated.
// NOTE: All sizes are in number of characters, NOT in bytes.
BASE_EXPORT size_t strlcpy(char* dst, const char* src, size_t dst_size);
BASE_EXPORT size_t wcslcpy(wchar_t* dst, const wchar_t* src, size_t dst_size);

// Scan a wprintf format string to determine whether it's portable across a
// variety of systems.  This function only checks that the conversion
// specifiers used by the format string are supported and have the same meaning
// on a variety of systems.  It doesn't check for other errors that might occur
// within a format string.
//
// Nonportable conversion specifiers for wprintf are:
//  - 's' and 'c' without an 'l' length modifier.  %s and %c operate on char
//     data on all systems except Windows, which treat them as wchar_t data.
//     Use %ls and %lc for wchar_t data instead.
//  - 'S' and 'C', which operate on wchar_t data on all systems except Windows,
//     which treat them as char data.  Use %ls and %lc for wchar_t data
//     instead.
//  - 'F', which is not identified by Windows wprintf documentation.
//  - 'D', 'O', and 'U', which are deprecated and not available on all systems.
//     Use %ld, %lo, and %lu instead.
//
// Note that there is no portable conversion specifier for char data when
// working with wprintf.
//
// This function is intended to be called from base::vswprintf.
BASE_EXPORT bool IsWprintfFormatPortable(const wchar_t* format);

// ASCII-specific tolower.  The standard library's tolower is locale sensitive,
// so we don't want to use it here.
template <class Char> inline Char ToLowerASCII(Char c) {
  return (c >= 'A' && c <= 'Z') ? (c + ('a' - 'A')) : c;
}

// ASCII-specific toupper.  The standard library's toupper is locale sensitive,
// so we don't want to use it here.
template <class Char> inline Char ToUpperASCII(Char c) {
  return (c >= 'a' && c <= 'z') ? (c + ('A' - 'a')) : c;
}

// Function objects to aid in comparing/searching strings.

template<typename Char> struct CaseInsensitiveCompare {
 public:
  bool operator()(Char x, Char y) const {
    // TODO(darin): Do we really want to do locale sensitive comparisons here?
    // See http://crbug.com/24917
    return tolower(x) == tolower(y);
  }
};

template<typename Char> struct CaseInsensitiveCompareASCII {
 public:
  bool operator()(Char x, Char y) const {
    return ToLowerASCII(x) == ToLowerASCII(y);
  }
};

}  // namespace base

#if defined(OS_WIN)
#include "base/string_util_win.h"
#elif defined(OS_POSIX)
#include "base/string_util_posix.h"
#else
#error Define string operations appropriately for your platform
#endif

// These threadsafe functions return references to globally unique empty
// strings.
//
// DO NOT USE THESE AS A GENERAL-PURPOSE SUBSTITUTE FOR DEFAULT CONSTRUCTORS.
// There is only one case where you should use these: functions which need to
// return a string by reference (e.g. as a class member accessor), and don't
// have an empty string to use (e.g. in an error case).  These should not be
// used as initializers, function arguments, or return values for functions
// which return by value or outparam.
BASE_EXPORT const std::string& EmptyString();
BASE_EXPORT const std::wstring& EmptyWString();
BASE_EXPORT const string16& EmptyString16();

BASE_EXPORT extern const wchar_t kWhitespaceWide[];
BASE_EXPORT extern const char16 kWhitespaceUTF16[];
BASE_EXPORT extern const char kWhitespaceASCII[];

BASE_EXPORT extern const char kUtf8ByteOrderMark[];

// Removes characters in |remove_chars| from anywhere in |input|.  Returns true
// if any characters were removed.  |remove_chars| must be null-terminated.
// NOTE: Safe to use the same variable for both |input| and |output|.
BASE_EXPORT bool RemoveChars(const string16& input,
                             const char16 remove_chars[],
                             string16* output);
BASE_EXPORT bool RemoveChars(const std::string& input,
                             const char remove_chars[],
                             std::string* output);

// Replaces characters in |replace_chars| from anywhere in |input| with
// |replace_with|.  Each character in |replace_chars| will be replaced with
// the |replace_with| string.  Returns true if any characters were replaced.
// |replace_chars| must be null-terminated.
// NOTE: Safe to use the same variable for both |input| and |output|.
BASE_EXPORT bool ReplaceChars(const string16& input,
                              const char16 replace_chars[],
                              const string16& replace_with,
                              string16* output);
BASE_EXPORT bool ReplaceChars(const std::string& input,
                              const char replace_chars[],
                              const std::string& replace_with,
                              std::string* output);

// Removes characters in |trim_chars| from the beginning and end of |input|.
// |trim_chars| must be null-terminated.
// NOTE: Safe to use the same variable for both |input| and |output|.
BASE_EXPORT bool TrimString(const std::wstring& input,
                            const wchar_t trim_chars[],
                            std::wstring* output);
BASE_EXPORT bool TrimString(const string16& input,
                            const char16 trim_chars[],
                            string16* output);
BASE_EXPORT bool TrimString(const std::string& input,
                            const char trim_chars[],
                            std::string* output);

// Truncates a string to the nearest UTF-8 character that will leave
// the string less than or equal to the specified byte size.
BASE_EXPORT void TruncateUTF8ToByteSize(const std::string& input,
                                        const size_t byte_size,
                                        std::string* output);

// Trims any whitespace from either end of the input string.  Returns where
// whitespace was found.
// The non-wide version has two functions:
// * TrimWhitespaceASCII()
//   This function is for ASCII strings and only looks for ASCII whitespace;
// Please choose the best one according to your usage.
// NOTE: Safe to use the same variable for both input and output.
enum TrimPositions {
  TRIM_NONE     = 0,
  TRIM_LEADING  = 1 << 0,
  TRIM_TRAILING = 1 << 1,
  TRIM_ALL      = TRIM_LEADING | TRIM_TRAILING,
};
BASE_EXPORT TrimPositions TrimWhitespace(const string16& input,
                                         TrimPositions positions,
                                         string16* output);
BASE_EXPORT TrimPositions TrimWhitespaceASCII(const std::string& input,
                                              TrimPositions positions,
                                              std::string* output);

// Deprecated. This function is only for backward compatibility and calls
// TrimWhitespaceASCII().
BASE_EXPORT TrimPositions TrimWhitespace(const std::string& input,
                                         TrimPositions positions,
                                         std::string* output);

// Searches  for CR or LF characters.  Removes all contiguous whitespace
// strings that contain them.  This is useful when trying to deal with text
// copied from terminals.
// Returns |text|, with the following three transformations:
// (1) Leading and trailing whitespace is trimmed.
// (2) If |trim_sequences_with_line_breaks| is true, any other whitespace
//     sequences containing a CR or LF are trimmed.
// (3) All other whitespace sequences are converted to single spaces.
BASE_EXPORT std::wstring CollapseWhitespace(
    const std::wstring& text,
    bool trim_sequences_with_line_breaks);
BASE_EXPORT string16 CollapseWhitespace(
    const string16& text,
    bool trim_sequences_with_line_breaks);
BASE_EXPORT std::string CollapseWhitespaceASCII(
    const std::string& text,
    bool trim_sequences_with_line_breaks);

// Returns true if the passed string is empty or contains only white-space
// characters.
BASE_EXPORT bool ContainsOnlyWhitespaceASCII(const std::string& str);
BASE_EXPORT bool ContainsOnlyWhitespace(const string16& str);

// Returns true if |input| is empty or contains only characters found in
// |characters|.
BASE_EXPORT bool ContainsOnlyChars(const std::wstring& input,
                                   const std::wstring& characters);
BASE_EXPORT bool ContainsOnlyChars(const string16& input,
                                   const string16& characters);
BASE_EXPORT bool ContainsOnlyChars(const std::string& input,
                                   const std::string& characters);

// Converts to 7-bit ASCII by truncating. The result must be known to be ASCII
// beforehand.
BASE_EXPORT std::string WideToASCII(const std::wstring& wide);
BASE_EXPORT std::string UTF16ToASCII(const string16& utf16);

// Converts the given wide string to the corresponding Latin1. This will fail
// (return false) if any characters are more than 255.
BASE_EXPORT bool WideToLatin1(const std::wstring& wide, std::string* latin1);

// Returns true if the specified string matches the criteria. How can a wide
// string be 8-bit or UTF8? It contains only characters that are < 256 (in the
// first case) or characters that use only 8-bits and whose 8-bit
// representation looks like a UTF-8 string (the second case).
//
// Note that IsStringUTF8 checks not only if the input is structurally
// valid but also if it doesn't contain any non-character codepoint
// (e.g. U+FFFE). It's done on purpose because all the existing callers want
// to have the maximum 'discriminating' power from other encodings. If
// there's a use case for just checking the structural validity, we have to
// add a new function for that.
BASE_EXPORT bool IsStringUTF8(const std::string& str);
BASE_EXPORT bool IsStringASCII(const std::wstring& str);
BASE_EXPORT bool IsStringASCII(const base::StringPiece& str);
BASE_EXPORT bool IsStringASCII(const string16& str);

// Converts the elements of the given string.  This version uses a pointer to
// clearly differentiate it from the non-pointer variant.
template <class str> inline void StringToLowerASCII(str* s) {
  for (typename str::iterator i = s->begin(); i != s->end(); ++i)
    *i = base::ToLowerASCII(*i);
}

template <class str> inline str StringToLowerASCII(const str& s) {
  // for std::string and std::wstring
  str output(s);
  StringToLowerASCII(&output);
  return output;
}

// Converts the elements of the given string.  This version uses a pointer to
// clearly differentiate it from the non-pointer variant.
template <class str> inline void StringToUpperASCII(str* s) {
  for (typename str::iterator i = s->begin(); i != s->end(); ++i)
    *i = base::ToUpperASCII(*i);
}

template <class str> inline str StringToUpperASCII(const str& s) {
  // for std::string and std::wstring
  str output(s);
  StringToUpperASCII(&output);
  return output;
}

// Compare the lower-case form of the given string against the given ASCII
// string.  This is useful for doing checking if an input string matches some
// token, and it is optimized to avoid intermediate string copies.  This API is
// borrowed from the equivalent APIs in Mozilla.
BASE_EXPORT bool LowerCaseEqualsASCII(const std::string& a, const char* b);
BASE_EXPORT bool LowerCaseEqualsASCII(const std::wstring& a, const char* b);
BASE_EXPORT bool LowerCaseEqualsASCII(const string16& a, const char* b);

// Same thing, but with string iterators instead.
BASE_EXPORT bool LowerCaseEqualsASCII(std::string::const_iterator a_begin,
                                      std::string::const_iterator a_end,
                                      const char* b);
BASE_EXPORT bool LowerCaseEqualsASCII(std::wstring::const_iterator a_begin,
                                      std::wstring::const_iterator a_end,
                                      const char* b);
BASE_EXPORT bool LowerCaseEqualsASCII(string16::const_iterator a_begin,
                                      string16::const_iterator a_end,
                                      const char* b);
BASE_EXPORT bool LowerCaseEqualsASCII(const char* a_begin,
                                      const char* a_end,
                                      const char* b);
BASE_EXPORT bool LowerCaseEqualsASCII(const wchar_t* a_begin,
                                      const wchar_t* a_end,
                                      const char* b);
BASE_EXPORT bool LowerCaseEqualsASCII(const char16* a_begin,
                                      const char16* a_end,
                                      const char* b);

// Performs a case-sensitive string compare. The behavior is undefined if both
// strings are not ASCII.
BASE_EXPORT bool EqualsASCII(const string16& a, const base::StringPiece& b);

// Returns true if str starts with search, or false otherwise.
BASE_EXPORT bool StartsWithASCII(const std::string& str,
                                 const std::string& search,
                                 bool case_sensitive);
BASE_EXPORT bool StartsWith(const std::wstring& str,
                            const std::wstring& search,
                            bool case_sensitive);
BASE_EXPORT bool StartsWith(const string16& str,
                            const string16& search,
                            bool case_sensitive);

// Returns true if str ends with search, or false otherwise.
BASE_EXPORT bool EndsWith(const std::string& str,
                          const std::string& search,
                          bool case_sensitive);
BASE_EXPORT bool EndsWith(const std::wstring& str,
                          const std::wstring& search,
                          bool case_sensitive);
BASE_EXPORT bool EndsWith(const string16& str,
                          const string16& search,
                          bool case_sensitive);


// Determines the type of ASCII character, independent of locale (the C
// library versions will change based on locale).
template <typename Char>
inline bool IsAsciiWhitespace(Char c) {
  return c == ' ' || c == '\r' || c == '\n' || c == '\t';
}
template <typename Char>
inline bool IsAsciiAlpha(Char c) {
  return ((c >= 'A') && (c <= 'Z')) || ((c >= 'a') && (c <= 'z'));
}
template <typename Char>
inline bool IsAsciiDigit(Char c) {
  return c >= '0' && c <= '9';
}

template <typename Char>
inline bool IsHexDigit(Char c) {
  return (c >= '0' && c <= '9') ||
         (c >= 'A' && c <= 'F') ||
         (c >= 'a' && c <= 'f');
}

template <typename Char>
inline Char HexDigitToInt(Char c) {
  DCHECK(IsHexDigit(c));
  if (c >= '0' && c <= '9')
    return c - '0';
  if (c >= 'A' && c <= 'F')
    return c - 'A' + 10;
  if (c >= 'a' && c <= 'f')
    return c - 'a' + 10;
  return 0;
}

// Returns true if it's a whitespace character.
inline bool IsWhitespace(wchar_t c) {
  return wcschr(kWhitespaceWide, c) != NULL;
}

// Return a byte string in human-readable format with a unit suffix. Not
// appropriate for use in any UI; use of FormatBytes and friends in ui/base is
// highly recommended instead. TODO(avi): Figure out how to get callers to use
// FormatBytes instead; remove this.
BASE_EXPORT string16 FormatBytesUnlocalized(int64 bytes);

// Starting at |start_offset| (usually 0), replace the first instance of
// |find_this| with |replace_with|.
BASE_EXPORT void ReplaceFirstSubstringAfterOffset(
    string16* str,
    string16::size_type start_offset,
    const string16& find_this,
    const string16& replace_with);
BASE_EXPORT void ReplaceFirstSubstringAfterOffset(
    std::string* str,
    std::string::size_type start_offset,
    const std::string& find_this,
    const std::string& replace_with);

// Starting at |start_offset| (usually 0), look through |str| and replace all
// instances of |find_this| with |replace_with|.
//
// This does entire substrings; use std::replace in <algorithm> for single
// characters, for example:
//   std::replace(str.begin(), str.end(), 'a', 'b');
BASE_EXPORT void ReplaceSubstringsAfterOffset(
    string16* str,
    string16::size_type start_offset,
    const string16& find_this,
    const string16& replace_with);
BASE_EXPORT void ReplaceSubstringsAfterOffset(
    std::string* str,
    std::string::size_type start_offset,
    const std::string& find_this,
    const std::string& replace_with);

// Reserves enough memory in |str| to accommodate |length_with_null| characters,
// sets the size of |str| to |length_with_null - 1| characters, and returns a
// pointer to the underlying contiguous array of characters.  This is typically
// used when calling a function that writes results into a character array, but
// the caller wants the data to be managed by a string-like object.  It is
// convenient in that is can be used inline in the call, and fast in that it
// avoids copying the results of the call from a char* into a string.
//
// |length_with_null| must be at least 2, since otherwise the underlying string
// would have size 0, and trying to access &((*str)[0]) in that case can result
// in a number of problems.
//
// Internally, this takes linear time because the resize() call 0-fills the
// underlying array for potentially all
// (|length_with_null - 1| * sizeof(string_type::value_type)) bytes.  Ideally we
// could avoid this aspect of the resize() call, as we expect the caller to
// immediately write over this memory, but there is no other way to set the size
// of the string, and not doing that will mean people who access |str| rather
// than str.c_str() will get back a string of whatever size |str| had on entry
// to this function (probably 0).
template <class string_type>
inline typename string_type::value_type* WriteInto(string_type* str,
                                                   size_t length_with_null) {
  DCHECK_GT(length_with_null, 1u);
  str->reserve(length_with_null);
  str->resize(length_with_null - 1);
  return &((*str)[0]);
}

//-----------------------------------------------------------------------------

// Splits a string into its fields delimited by any of the characters in
// |delimiters|.  Each field is added to the |tokens| vector.  Returns the
// number of tokens found.
BASE_EXPORT size_t Tokenize(const std::wstring& str,
                            const std::wstring& delimiters,
                            std::vector<std::wstring>* tokens);
BASE_EXPORT size_t Tokenize(const string16& str,
                            const string16& delimiters,
                            std::vector<string16>* tokens);
BASE_EXPORT size_t Tokenize(const std::string& str,
                            const std::string& delimiters,
                            std::vector<std::string>* tokens);
BASE_EXPORT size_t Tokenize(const base::StringPiece& str,
                            const base::StringPiece& delimiters,
                            std::vector<base::StringPiece>* tokens);

// Does the opposite of SplitString().
BASE_EXPORT string16 JoinString(const std::vector<string16>& parts, char16 s);
BASE_EXPORT std::string JoinString(
    const std::vector<std::string>& parts, char s);

// Replace $1-$2-$3..$9 in the format string with |a|-|b|-|c|..|i| respectively.
// Additionally, any number of consecutive '$' characters is replaced by that
// number less one. Eg $$->$, $$$->$$, etc. The offsets parameter here can be
// NULL. This only allows you to use up to nine replacements.
BASE_EXPORT string16 ReplaceStringPlaceholders(
    const string16& format_string,
    const std::vector<string16>& subst,
    std::vector<size_t>* offsets);

BASE_EXPORT std::string ReplaceStringPlaceholders(
    const base::StringPiece& format_string,
    const std::vector<std::string>& subst,
    std::vector<size_t>* offsets);

// Single-string shortcut for ReplaceStringHolders. |offset| may be NULL.
BASE_EXPORT string16 ReplaceStringPlaceholders(const string16& format_string,
                                               const string16& a,
                                               size_t* offset);

// Returns true if the string passed in matches the pattern. The pattern
// string can contain wildcards like * and ?
// The backslash character (\) is an escape character for * and ?
// We limit the patterns to having a max of 16 * or ? characters.
// ? matches 0 or 1 character, while * matches 0 or more characters.
BASE_EXPORT bool MatchPattern(const base::StringPiece& string,
                              const base::StringPiece& pattern);
BASE_EXPORT bool MatchPattern(const string16& string, const string16& pattern);

// Hack to convert any char-like type to its unsigned counterpart.
// For example, it will convert char, signed char and unsigned char to unsigned
// char.
template<typename T>
struct ToUnsigned {
  typedef T Unsigned;
};

template<>
struct ToUnsigned<char> {
  typedef unsigned char Unsigned;
};
template<>
struct ToUnsigned<signed char> {
  typedef unsigned char Unsigned;
};
template<>
struct ToUnsigned<wchar_t> {
#if defined(WCHAR_T_IS_UTF16)
  typedef unsigned short Unsigned;
#elif defined(WCHAR_T_IS_UTF32)
  typedef uint32 Unsigned;
#endif
};
template<>
struct ToUnsigned<short> {
  typedef unsigned short Unsigned;
};

#endif  // BASE_STRING_UTIL_H_