1.1 --- /dev/null Thu Jan 01 00:00:00 1970 +0000
1.2 +++ b/ipc/chromium/src/base/string_util.h Wed Dec 31 06:09:35 2014 +0100
1.3 @@ -0,0 +1,235 @@
1.4 +// Copyright (c) 2006-2008 The Chromium Authors. All rights reserved.
1.5 +// Use of this source code is governed by a BSD-style license that can be
1.6 +// found in the LICENSE file.
1.7 +//
1.8 +// This file defines utility functions for working with strings.
1.9 +
1.10 +#ifndef BASE_STRING_UTIL_H_
1.11 +#define BASE_STRING_UTIL_H_
1.12 +
1.13 +#include <stdarg.h> // va_list
1.14 +#include <ctype.h>
1.15 +
1.16 +#include <string>
1.17 +#include <vector>
1.18 +
1.19 +#include "base/basictypes.h"
1.20 +#include "base/string16.h"
1.21 +#include "base/string_piece.h" // For implicit conversions.
1.22 +
1.23 +// Safe standard library wrappers for all platforms.
1.24 +
1.25 +namespace base {
1.26 +
1.27 +// C standard-library functions like "strncasecmp" and "snprintf" that aren't
1.28 +// cross-platform are provided as "base::strncasecmp", and their prototypes
1.29 +// are listed below. These functions are then implemented as inline calls
1.30 +// to the platform-specific equivalents in the platform-specific headers.
1.31 +
1.32 +// Compare the two strings s1 and s2 without regard to case using
1.33 +// the current locale; returns 0 if they are equal, 1 if s1 > s2, and -1 if
1.34 +// s2 > s1 according to a lexicographic comparison.
1.35 +int strcasecmp(const char* s1, const char* s2);
1.36 +
1.37 +// Compare up to count characters of s1 and s2 without regard to case using
1.38 +// the current locale; returns 0 if they are equal, 1 if s1 > s2, and -1 if
1.39 +// s2 > s1 according to a lexicographic comparison.
1.40 +int strncasecmp(const char* s1, const char* s2, size_t count);
1.41 +
1.42 +// Wrapper for vsnprintf that always null-terminates and always returns the
1.43 +// number of characters that would be in an untruncated formatted
1.44 +// string, even when truncation occurs.
1.45 +int vsnprintf(char* buffer, size_t size, const char* format, va_list arguments);
1.46 +
1.47 +// vswprintf always null-terminates, but when truncation occurs, it will either
1.48 +// return -1 or the number of characters that would be in an untruncated
1.49 +// formatted string. The actual return value depends on the underlying
1.50 +// C library's vswprintf implementation.
1.51 +int vswprintf(wchar_t* buffer, size_t size,
1.52 + const wchar_t* format, va_list arguments);
1.53 +
1.54 +// Some of these implementations need to be inlined.
1.55 +
1.56 +inline int snprintf(char* buffer, size_t size, const char* format, ...) {
1.57 + va_list arguments;
1.58 + va_start(arguments, format);
1.59 + int result = vsnprintf(buffer, size, format, arguments);
1.60 + va_end(arguments);
1.61 + return result;
1.62 +}
1.63 +
1.64 +inline int swprintf(wchar_t* buffer, size_t size, const wchar_t* format, ...) {
1.65 + va_list arguments;
1.66 + va_start(arguments, format);
1.67 + int result = vswprintf(buffer, size, format, arguments);
1.68 + va_end(arguments);
1.69 + return result;
1.70 +}
1.71 +
1.72 +// BSD-style safe and consistent string copy functions.
1.73 +// Copies |src| to |dst|, where |dst_size| is the total allocated size of |dst|.
1.74 +// Copies at most |dst_size|-1 characters, and always NULL terminates |dst|, as
1.75 +// long as |dst_size| is not 0. Returns the length of |src| in characters.
1.76 +// If the return value is >= dst_size, then the output was truncated.
1.77 +// NOTE: All sizes are in number of characters, NOT in bytes.
1.78 +size_t strlcpy(char* dst, const char* src, size_t dst_size);
1.79 +size_t wcslcpy(wchar_t* dst, const wchar_t* src, size_t dst_size);
1.80 +
1.81 +// Scan a wprintf format string to determine whether it's portable across a
1.82 +// variety of systems. This function only checks that the conversion
1.83 +// specifiers used by the format string are supported and have the same meaning
1.84 +// on a variety of systems. It doesn't check for other errors that might occur
1.85 +// within a format string.
1.86 +//
1.87 +// Nonportable conversion specifiers for wprintf are:
1.88 +// - 's' and 'c' without an 'l' length modifier. %s and %c operate on char
1.89 +// data on all systems except Windows, which treat them as wchar_t data.
1.90 +// Use %ls and %lc for wchar_t data instead.
1.91 +// - 'S' and 'C', which operate on wchar_t data on all systems except Windows,
1.92 +// which treat them as char data. Use %ls and %lc for wchar_t data
1.93 +// instead.
1.94 +// - 'F', which is not identified by Windows wprintf documentation.
1.95 +// - 'D', 'O', and 'U', which are deprecated and not available on all systems.
1.96 +// Use %ld, %lo, and %lu instead.
1.97 +//
1.98 +// Note that there is no portable conversion specifier for char data when
1.99 +// working with wprintf.
1.100 +//
1.101 +// This function is intended to be called from base::vswprintf.
1.102 +bool IsWprintfFormatPortable(const wchar_t* format);
1.103 +
1.104 +} // namespace base
1.105 +
1.106 +#if defined(OS_WIN)
1.107 +#include "base/string_util_win.h"
1.108 +#elif defined(OS_POSIX)
1.109 +#include "base/string_util_posix.h"
1.110 +#else
1.111 +#error Define string operations appropriately for your platform
1.112 +#endif
1.113 +
1.114 +// Trims any whitespace from either end of the input string. Returns where
1.115 +// whitespace was found.
1.116 +// The non-wide version has two functions:
1.117 +// * TrimWhitespaceASCII()
1.118 +// This function is for ASCII strings and only looks for ASCII whitespace;
1.119 +// * TrimWhitespaceUTF8()
1.120 +// This function is for UTF-8 strings and looks for Unicode whitespace.
1.121 +// Please choose the best one according to your usage.
1.122 +// NOTE: Safe to use the same variable for both input and output.
1.123 +enum TrimPositions {
1.124 + TRIM_NONE = 0,
1.125 + TRIM_LEADING = 1 << 0,
1.126 + TRIM_TRAILING = 1 << 1,
1.127 + TRIM_ALL = TRIM_LEADING | TRIM_TRAILING
1.128 +};
1.129 +TrimPositions TrimWhitespace(const std::wstring& input,
1.130 + TrimPositions positions,
1.131 + std::wstring* output);
1.132 +TrimPositions TrimWhitespaceASCII(const std::string& input,
1.133 + TrimPositions positions,
1.134 + std::string* output);
1.135 +
1.136 +// Deprecated. This function is only for backward compatibility and calls
1.137 +// TrimWhitespaceASCII().
1.138 +TrimPositions TrimWhitespace(const std::string& input,
1.139 + TrimPositions positions,
1.140 + std::string* output);
1.141 +
1.142 +// Searches for CR or LF characters. Removes all contiguous whitespace
1.143 +// strings that contain them. This is useful when trying to deal with text
1.144 +// copied from terminals.
1.145 +// Returns |text, with the following three transformations:
1.146 +// (1) Leading and trailing whitespace is trimmed.
1.147 +// (2) If |trim_sequences_with_line_breaks| is true, any other whitespace
1.148 +// sequences containing a CR or LF are trimmed.
1.149 +// (3) All other whitespace sequences are converted to single spaces.
1.150 +std::wstring CollapseWhitespace(const std::wstring& text,
1.151 + bool trim_sequences_with_line_breaks);
1.152 +
1.153 +// These convert between ASCII (7-bit) and Wide/UTF16 strings.
1.154 +std::string WideToASCII(const std::wstring& wide);
1.155 +std::wstring ASCIIToWide(const std::string& ascii);
1.156 +std::string UTF16ToASCII(const string16& utf16);
1.157 +string16 ASCIIToUTF16(const std::string& ascii);
1.158 +
1.159 +// These convert between UTF-8, -16, and -32 strings. They are potentially slow,
1.160 +// so avoid unnecessary conversions. The low-level versions return a boolean
1.161 +// indicating whether the conversion was 100% valid. In this case, it will still
1.162 +// do the best it can and put the result in the output buffer. The versions that
1.163 +// return strings ignore this error and just return the best conversion
1.164 +// possible.
1.165 +bool WideToUTF8(const wchar_t* src, size_t src_len, std::string* output);
1.166 +std::string WideToUTF8(const std::wstring& wide);
1.167 +bool UTF8ToWide(const char* src, size_t src_len, std::wstring* output);
1.168 +std::wstring UTF8ToWide(const ::StringPiece& utf8);
1.169 +
1.170 +bool IsStringASCII(const std::wstring& str);
1.171 +bool IsStringASCII(const std::string& str);
1.172 +bool IsStringASCII(const string16& str);
1.173 +
1.174 +// Specialized string-conversion functions.
1.175 +std::string IntToString(int value);
1.176 +std::wstring IntToWString(int value);
1.177 +std::string UintToString(unsigned int value);
1.178 +std::wstring UintToWString(unsigned int value);
1.179 +std::string Int64ToString(int64_t value);
1.180 +std::wstring Int64ToWString(int64_t value);
1.181 +std::string Uint64ToString(uint64_t value);
1.182 +std::wstring Uint64ToWString(uint64_t value);
1.183 +// The DoubleToString methods convert the double to a string format that
1.184 +// ignores the locale. If you want to use locale specific formatting, use ICU.
1.185 +std::string DoubleToString(double value);
1.186 +std::wstring DoubleToWString(double value);
1.187 +
1.188 +// Perform a best-effort conversion of the input string to a numeric type,
1.189 +// setting |*output| to the result of the conversion. Returns true for
1.190 +// "perfect" conversions; returns false in the following cases:
1.191 +// - Overflow/underflow. |*output| will be set to the maximum value supported
1.192 +// by the data type.
1.193 +// - Trailing characters in the string after parsing the number. |*output|
1.194 +// will be set to the value of the number that was parsed.
1.195 +// - No characters parseable as a number at the beginning of the string.
1.196 +// |*output| will be set to 0.
1.197 +// - Empty string. |*output| will be set to 0.
1.198 +bool StringToInt(const std::string& input, int* output);
1.199 +bool StringToInt(const string16& input, int* output);
1.200 +bool StringToInt64(const std::string& input, int64_t* output);
1.201 +bool StringToInt64(const string16& input, int64_t* output);
1.202 +
1.203 +// Convenience forms of the above, when the caller is uninterested in the
1.204 +// boolean return value. These return only the |*output| value from the
1.205 +// above conversions: a best-effort conversion when possible, otherwise, 0.
1.206 +int StringToInt(const std::string& value);
1.207 +int StringToInt(const string16& value);
1.208 +int64_t StringToInt64(const std::string& value);
1.209 +int64_t StringToInt64(const string16& value);
1.210 +
1.211 +// Return a C++ string given printf-like input.
1.212 +std::string StringPrintf(const char* format, ...);
1.213 +std::wstring StringPrintf(const wchar_t* format, ...);
1.214 +
1.215 +// Store result into a supplied string and return it
1.216 +const std::string& SStringPrintf(std::string* dst, const char* format, ...);
1.217 +const std::wstring& SStringPrintf(std::wstring* dst,
1.218 + const wchar_t* format, ...);
1.219 +
1.220 +// Append result to a supplied string
1.221 +void StringAppendF(std::string* dst, const char* format, ...);
1.222 +void StringAppendF(std::wstring* dst, const wchar_t* format, ...);
1.223 +
1.224 +//-----------------------------------------------------------------------------
1.225 +
1.226 +// Splits |str| into a vector of strings delimited by |s|. Append the results
1.227 +// into |r| as they appear. If several instances of |s| are contiguous, or if
1.228 +// |str| begins with or ends with |s|, then an empty string is inserted.
1.229 +//
1.230 +// Every substring is trimmed of any leading or trailing white space.
1.231 +void SplitString(const std::wstring& str,
1.232 + wchar_t s,
1.233 + std::vector<std::wstring>* r);
1.234 +void SplitString(const std::string& str,
1.235 + char s,
1.236 + std::vector<std::string>* r);
1.237 +
1.238 +#endif // BASE_STRING_UTIL_H_
