|
1 // Copyright (c) 2006-2008 The Chromium Authors. All rights reserved. |
|
2 // Use of this source code is governed by a BSD-style license that can be |
|
3 // found in the LICENSE file. |
|
4 // |
|
5 // This file defines utility functions for working with strings. |
|
6 |
|
7 #ifndef BASE_STRING_UTIL_H_ |
|
8 #define BASE_STRING_UTIL_H_ |
|
9 |
|
10 #include <stdarg.h> // va_list |
|
11 #include <ctype.h> |
|
12 |
|
13 #include <string> |
|
14 #include <vector> |
|
15 |
|
16 #include "base/basictypes.h" |
|
17 #include "base/string16.h" |
|
18 #include "base/string_piece.h" // For implicit conversions. |
|
19 |
|
20 // Safe standard library wrappers for all platforms. |
|
21 |
|
22 namespace base { |
|
23 |
|
24 // C standard-library functions like "strncasecmp" and "snprintf" that aren't |
|
25 // cross-platform are provided as "base::strncasecmp", and their prototypes |
|
26 // are listed below. These functions are then implemented as inline calls |
|
27 // to the platform-specific equivalents in the platform-specific headers. |
|
28 |
|
29 // Compare the two strings s1 and s2 without regard to case using |
|
30 // the current locale; returns 0 if they are equal, 1 if s1 > s2, and -1 if |
|
31 // s2 > s1 according to a lexicographic comparison. |
|
32 int strcasecmp(const char* s1, const char* s2); |
|
33 |
|
34 // Compare up to count characters of s1 and s2 without regard to case using |
|
35 // the current locale; returns 0 if they are equal, 1 if s1 > s2, and -1 if |
|
36 // s2 > s1 according to a lexicographic comparison. |
|
37 int strncasecmp(const char* s1, const char* s2, size_t count); |
|
38 |
|
39 // Wrapper for vsnprintf that always null-terminates and always returns the |
|
40 // number of characters that would be in an untruncated formatted |
|
41 // string, even when truncation occurs. |
|
42 int vsnprintf(char* buffer, size_t size, const char* format, va_list arguments); |
|
43 |
|
44 // vswprintf always null-terminates, but when truncation occurs, it will either |
|
45 // return -1 or the number of characters that would be in an untruncated |
|
46 // formatted string. The actual return value depends on the underlying |
|
47 // C library's vswprintf implementation. |
|
48 int vswprintf(wchar_t* buffer, size_t size, |
|
49 const wchar_t* format, va_list arguments); |
|
50 |
|
51 // Some of these implementations need to be inlined. |
|
52 |
|
53 inline int snprintf(char* buffer, size_t size, const char* format, ...) { |
|
54 va_list arguments; |
|
55 va_start(arguments, format); |
|
56 int result = vsnprintf(buffer, size, format, arguments); |
|
57 va_end(arguments); |
|
58 return result; |
|
59 } |
|
60 |
|
61 inline int swprintf(wchar_t* buffer, size_t size, const wchar_t* format, ...) { |
|
62 va_list arguments; |
|
63 va_start(arguments, format); |
|
64 int result = vswprintf(buffer, size, format, arguments); |
|
65 va_end(arguments); |
|
66 return result; |
|
67 } |
|
68 |
|
69 // BSD-style safe and consistent string copy functions. |
|
70 // Copies |src| to |dst|, where |dst_size| is the total allocated size of |dst|. |
|
71 // Copies at most |dst_size|-1 characters, and always NULL terminates |dst|, as |
|
72 // long as |dst_size| is not 0. Returns the length of |src| in characters. |
|
73 // If the return value is >= dst_size, then the output was truncated. |
|
74 // NOTE: All sizes are in number of characters, NOT in bytes. |
|
75 size_t strlcpy(char* dst, const char* src, size_t dst_size); |
|
76 size_t wcslcpy(wchar_t* dst, const wchar_t* src, size_t dst_size); |
|
77 |
|
78 // Scan a wprintf format string to determine whether it's portable across a |
|
79 // variety of systems. This function only checks that the conversion |
|
80 // specifiers used by the format string are supported and have the same meaning |
|
81 // on a variety of systems. It doesn't check for other errors that might occur |
|
82 // within a format string. |
|
83 // |
|
84 // Nonportable conversion specifiers for wprintf are: |
|
85 // - 's' and 'c' without an 'l' length modifier. %s and %c operate on char |
|
86 // data on all systems except Windows, which treat them as wchar_t data. |
|
87 // Use %ls and %lc for wchar_t data instead. |
|
88 // - 'S' and 'C', which operate on wchar_t data on all systems except Windows, |
|
89 // which treat them as char data. Use %ls and %lc for wchar_t data |
|
90 // instead. |
|
91 // - 'F', which is not identified by Windows wprintf documentation. |
|
92 // - 'D', 'O', and 'U', which are deprecated and not available on all systems. |
|
93 // Use %ld, %lo, and %lu instead. |
|
94 // |
|
95 // Note that there is no portable conversion specifier for char data when |
|
96 // working with wprintf. |
|
97 // |
|
98 // This function is intended to be called from base::vswprintf. |
|
99 bool IsWprintfFormatPortable(const wchar_t* format); |
|
100 |
|
101 } // namespace base |
|
102 |
|
103 #if defined(OS_WIN) |
|
104 #include "base/string_util_win.h" |
|
105 #elif defined(OS_POSIX) |
|
106 #include "base/string_util_posix.h" |
|
107 #else |
|
108 #error Define string operations appropriately for your platform |
|
109 #endif |
|
110 |
|
111 // Trims any whitespace from either end of the input string. Returns where |
|
112 // whitespace was found. |
|
113 // The non-wide version has two functions: |
|
114 // * TrimWhitespaceASCII() |
|
115 // This function is for ASCII strings and only looks for ASCII whitespace; |
|
116 // * TrimWhitespaceUTF8() |
|
117 // This function is for UTF-8 strings and looks for Unicode whitespace. |
|
118 // Please choose the best one according to your usage. |
|
119 // NOTE: Safe to use the same variable for both input and output. |
|
120 enum TrimPositions { |
|
121 TRIM_NONE = 0, |
|
122 TRIM_LEADING = 1 << 0, |
|
123 TRIM_TRAILING = 1 << 1, |
|
124 TRIM_ALL = TRIM_LEADING | TRIM_TRAILING |
|
125 }; |
|
126 TrimPositions TrimWhitespace(const std::wstring& input, |
|
127 TrimPositions positions, |
|
128 std::wstring* output); |
|
129 TrimPositions TrimWhitespaceASCII(const std::string& input, |
|
130 TrimPositions positions, |
|
131 std::string* output); |
|
132 |
|
133 // Deprecated. This function is only for backward compatibility and calls |
|
134 // TrimWhitespaceASCII(). |
|
135 TrimPositions TrimWhitespace(const std::string& input, |
|
136 TrimPositions positions, |
|
137 std::string* output); |
|
138 |
|
139 // Searches for CR or LF characters. Removes all contiguous whitespace |
|
140 // strings that contain them. This is useful when trying to deal with text |
|
141 // copied from terminals. |
|
142 // Returns |text, with the following three transformations: |
|
143 // (1) Leading and trailing whitespace is trimmed. |
|
144 // (2) If |trim_sequences_with_line_breaks| is true, any other whitespace |
|
145 // sequences containing a CR or LF are trimmed. |
|
146 // (3) All other whitespace sequences are converted to single spaces. |
|
147 std::wstring CollapseWhitespace(const std::wstring& text, |
|
148 bool trim_sequences_with_line_breaks); |
|
149 |
|
150 // These convert between ASCII (7-bit) and Wide/UTF16 strings. |
|
151 std::string WideToASCII(const std::wstring& wide); |
|
152 std::wstring ASCIIToWide(const std::string& ascii); |
|
153 std::string UTF16ToASCII(const string16& utf16); |
|
154 string16 ASCIIToUTF16(const std::string& ascii); |
|
155 |
|
156 // These convert between UTF-8, -16, and -32 strings. They are potentially slow, |
|
157 // so avoid unnecessary conversions. The low-level versions return a boolean |
|
158 // indicating whether the conversion was 100% valid. In this case, it will still |
|
159 // do the best it can and put the result in the output buffer. The versions that |
|
160 // return strings ignore this error and just return the best conversion |
|
161 // possible. |
|
162 bool WideToUTF8(const wchar_t* src, size_t src_len, std::string* output); |
|
163 std::string WideToUTF8(const std::wstring& wide); |
|
164 bool UTF8ToWide(const char* src, size_t src_len, std::wstring* output); |
|
165 std::wstring UTF8ToWide(const ::StringPiece& utf8); |
|
166 |
|
167 bool IsStringASCII(const std::wstring& str); |
|
168 bool IsStringASCII(const std::string& str); |
|
169 bool IsStringASCII(const string16& str); |
|
170 |
|
171 // Specialized string-conversion functions. |
|
172 std::string IntToString(int value); |
|
173 std::wstring IntToWString(int value); |
|
174 std::string UintToString(unsigned int value); |
|
175 std::wstring UintToWString(unsigned int value); |
|
176 std::string Int64ToString(int64_t value); |
|
177 std::wstring Int64ToWString(int64_t value); |
|
178 std::string Uint64ToString(uint64_t value); |
|
179 std::wstring Uint64ToWString(uint64_t value); |
|
180 // The DoubleToString methods convert the double to a string format that |
|
181 // ignores the locale. If you want to use locale specific formatting, use ICU. |
|
182 std::string DoubleToString(double value); |
|
183 std::wstring DoubleToWString(double value); |
|
184 |
|
185 // Perform a best-effort conversion of the input string to a numeric type, |
|
186 // setting |*output| to the result of the conversion. Returns true for |
|
187 // "perfect" conversions; returns false in the following cases: |
|
188 // - Overflow/underflow. |*output| will be set to the maximum value supported |
|
189 // by the data type. |
|
190 // - Trailing characters in the string after parsing the number. |*output| |
|
191 // will be set to the value of the number that was parsed. |
|
192 // - No characters parseable as a number at the beginning of the string. |
|
193 // |*output| will be set to 0. |
|
194 // - Empty string. |*output| will be set to 0. |
|
195 bool StringToInt(const std::string& input, int* output); |
|
196 bool StringToInt(const string16& input, int* output); |
|
197 bool StringToInt64(const std::string& input, int64_t* output); |
|
198 bool StringToInt64(const string16& input, int64_t* output); |
|
199 |
|
200 // Convenience forms of the above, when the caller is uninterested in the |
|
201 // boolean return value. These return only the |*output| value from the |
|
202 // above conversions: a best-effort conversion when possible, otherwise, 0. |
|
203 int StringToInt(const std::string& value); |
|
204 int StringToInt(const string16& value); |
|
205 int64_t StringToInt64(const std::string& value); |
|
206 int64_t StringToInt64(const string16& value); |
|
207 |
|
208 // Return a C++ string given printf-like input. |
|
209 std::string StringPrintf(const char* format, ...); |
|
210 std::wstring StringPrintf(const wchar_t* format, ...); |
|
211 |
|
212 // Store result into a supplied string and return it |
|
213 const std::string& SStringPrintf(std::string* dst, const char* format, ...); |
|
214 const std::wstring& SStringPrintf(std::wstring* dst, |
|
215 const wchar_t* format, ...); |
|
216 |
|
217 // Append result to a supplied string |
|
218 void StringAppendF(std::string* dst, const char* format, ...); |
|
219 void StringAppendF(std::wstring* dst, const wchar_t* format, ...); |
|
220 |
|
221 //----------------------------------------------------------------------------- |
|
222 |
|
223 // Splits |str| into a vector of strings delimited by |s|. Append the results |
|
224 // into |r| as they appear. If several instances of |s| are contiguous, or if |
|
225 // |str| begins with or ends with |s|, then an empty string is inserted. |
|
226 // |
|
227 // Every substring is trimmed of any leading or trailing white space. |
|
228 void SplitString(const std::wstring& str, |
|
229 wchar_t s, |
|
230 std::vector<std::wstring>* r); |
|
231 void SplitString(const std::string& str, |
|
232 char s, |
|
233 std::vector<std::string>* r); |
|
234 |
|
235 #endif // BASE_STRING_UTIL_H_ |