OLD | NEW |
| (Empty) |
1 // Copyright (c) 2009 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 #ifndef BASE_I18N_STRING_CONVERSIONS_H_ | |
6 #define BASE_I18N_STRING_CONVERSIONS_H_ | |
7 | |
8 #include <string> | |
9 | |
10 #include "base/string16.h" | |
11 #include "base/string_piece.h" | |
12 | |
13 // These convert between UTF-8, -16, and -32 strings. They are potentially slow, | |
14 // so avoid unnecessary conversions. The low-level versions return a boolean | |
15 // indicating whether the conversion was 100% valid. In this case, it will still | |
16 // do the best it can and put the result in the output buffer. The versions that | |
17 // return strings ignore this error and just return the best conversion | |
18 // possible. | |
19 // | |
20 // Note that only the structural validity is checked and non-character | |
21 // codepoints and unassigned are regarded as valid. | |
22 // TODO(jungshik): Consider replacing an invalid input sequence with | |
23 // the Unicode replacement character or adding |replacement_char| parameter. | |
24 // Currently, it's skipped in the ouput, which could be problematic in | |
25 // some situations. | |
26 bool WideToUTF8(const wchar_t* src, size_t src_len, std::string* output); | |
27 std::string WideToUTF8(const std::wstring& wide); | |
28 bool UTF8ToWide(const char* src, size_t src_len, std::wstring* output); | |
29 std::wstring UTF8ToWide(const base::StringPiece& utf8); | |
30 | |
31 bool WideToUTF16(const wchar_t* src, size_t src_len, string16* output); | |
32 string16 WideToUTF16(const std::wstring& wide); | |
33 bool UTF16ToWide(const char16* src, size_t src_len, std::wstring* output); | |
34 std::wstring UTF16ToWide(const string16& utf16); | |
35 | |
36 bool UTF8ToUTF16(const char* src, size_t src_len, string16* output); | |
37 string16 UTF8ToUTF16(const std::string& utf8); | |
38 bool UTF16ToUTF8(const char16* src, size_t src_len, std::string* output); | |
39 std::string UTF16ToUTF8(const string16& utf16); | |
40 | |
41 // We are trying to get rid of wstring as much as possible, but it's too big | |
42 // a mess to do it all at once. These conversions should be used when we | |
43 // really should just be passing a string16 around, but we haven't finished | |
44 // porting whatever module uses wstring and the conversion is being used as a | |
45 // stopcock. This makes it easy to grep for the ones that should be removed. | |
46 #if defined(OS_WIN) | |
47 # define WideToUTF16Hack | |
48 # define UTF16ToWideHack | |
49 #else | |
50 # define WideToUTF16Hack WideToUTF16 | |
51 # define UTF16ToWideHack UTF16ToWide | |
52 #endif | |
53 | |
54 // Defines the error handling modes of UTF16ToCodepage, CodepageToUTF16, | |
55 // WideToCodepage and CodepageToWide. | |
56 class OnStringUtilConversionError { | |
57 public: | |
58 enum Type { | |
59 // The function will return failure. The output buffer will be empty. | |
60 FAIL, | |
61 | |
62 // The offending characters are skipped and the conversion will proceed as | |
63 // if they did not exist. | |
64 SKIP, | |
65 | |
66 // When converting to Unicode, the offending byte sequences are substituted | |
67 // by Unicode replacement character (U+FFFD). When converting from Unicode, | |
68 // this is the same as SKIP. | |
69 SUBSTITUTE, | |
70 }; | |
71 | |
72 private: | |
73 OnStringUtilConversionError(); | |
74 }; | |
75 | |
76 // Converts between UTF-16 strings and the encoding specified. If the | |
77 // encoding doesn't exist or the encoding fails (when on_error is FAIL), | |
78 // returns false. | |
79 bool UTF16ToCodepage(const string16& utf16, | |
80 const char* codepage_name, | |
81 OnStringUtilConversionError::Type on_error, | |
82 std::string* encoded); | |
83 | |
84 bool CodepageToUTF16(const std::string& encoded, | |
85 const char* codepage_name, | |
86 OnStringUtilConversionError::Type on_error, | |
87 string16* utf16); | |
88 | |
89 // Converts between wide strings and the encoding specified. If the | |
90 // encoding doesn't exist or the encoding fails (when on_error is FAIL), | |
91 // returns false. | |
92 bool WideToCodepage(const std::wstring& wide, | |
93 const char* codepage_name, | |
94 OnStringUtilConversionError::Type on_error, | |
95 std::string* encoded); | |
96 bool CodepageToWide(const std::string& encoded, | |
97 const char* codepage_name, | |
98 OnStringUtilConversionError::Type on_error, | |
99 std::wstring* wide); | |
100 | |
101 #endif // BASE_I18N_STRING_CONVERSIONS_H_ | |
OLD | NEW |