]> git.saurik.com Git - wxWidgets.git/blob - include/wx/stringops.h
made UTF-8 tables const
[wxWidgets.git] / include / wx / stringops.h
1 ///////////////////////////////////////////////////////////////////////////////
2 // Name: wx/stringops.h
3 // Purpose: implementation of wxString primitive operations
4 // Author: Vaclav Slavik
5 // Modified by:
6 // Created: 2007-04-16
7 // RCS-ID: $Id$
8 // Copyright: (c) 2007 REA Elektronik GmbH
9 // Licence: wxWindows licence
10 ///////////////////////////////////////////////////////////////////////////////
11
12 #ifndef _WX_WXSTRINGOPS_H__
13 #define _WX_WXSTRINGOPS_H__
14
15 #include "wx/chartype.h"
16 #include "wx/stringimpl.h"
17 #include "wx/unichar.h"
18 #include "wx/buffer.h"
19
20 // This header contains wxStringOperations "namespace" class that implements
21 // elementary operations on string data as static methods; wxString methods and
22 // iterators are implemented in terms of it. Two implementations are available,
23 // one for UTF-8 encoded char* string and one for "raw" wchar_t* strings (or
24 // char* in ANSI build).
25
26 // FIXME-UTF8: only wchar after we remove ANSI build
27 #if wxUSE_UNICODE_WCHAR || !wxUSE_UNICODE
28 struct WXDLLIMPEXP_BASE wxStringOperationsWchar
29 {
30 // moves the iterator to the next Unicode character
31 static void IncIter(wxStringImpl::iterator& i) { ++i; }
32 static void IncIter(wxStringImpl::const_iterator& i) { ++i; }
33
34 // moves the iterator to the previous Unicode character
35 static void DecIter(wxStringImpl::iterator& i) { --i; }
36 static void DecIter(wxStringImpl::const_iterator& i) { --i; }
37
38 // moves the iterator by n Unicode characters
39 static wxStringImpl::iterator AddToIter(const wxStringImpl::iterator& i, int n)
40 { return i + n; }
41 static wxStringImpl::const_iterator AddToIter(const wxStringImpl::const_iterator& i, int n)
42 { return i + n; }
43
44 // returns distance of the two iterators in Unicode characters
45 static int DiffIters(const wxStringImpl::iterator& i1,
46 const wxStringImpl::iterator& i2)
47 { return i1 - i2; }
48 static int DiffIters(const wxStringImpl::const_iterator& i1,
49 const wxStringImpl::const_iterator& i2)
50 { return i1 - i2; }
51
52 // encodes the character to a form used to represent it in internal
53 // representation (returns a string in UTF8 version)
54 static wxChar EncodeChar(const wxUniChar& ch) { return (wxChar)ch; }
55
56 static wxUniChar DecodeChar(const wxStringImpl::const_iterator& i)
57 { return *i; }
58 };
59 #endif // wxUSE_UNICODE_WCHAR || !wxUSE_UNICODE
60
61
62 #if wxUSE_UNICODE_UTF8
63 struct WXDLLIMPEXP_BASE wxStringOperationsUtf8
64 {
65 // checks correctness of UTF-8 sequence
66 static bool IsValidUtf8String(const char *c,
67 size_t len = wxStringImpl::npos);
68 #ifdef __WXDEBUG__
69 static bool IsValidUtf8LeadByte(unsigned char c);
70 #endif
71
72 // table of offsets to skip forward when iterating over UTF-8 sequence
73 static const unsigned char ms_utf8IterTable[256];
74
75
76 template<typename Iterator>
77 static void IncIter(Iterator& i)
78 {
79 wxASSERT( IsValidUtf8LeadByte(*i) );
80 i += ms_utf8IterTable[(unsigned char)*i];
81 }
82
83 template<typename Iterator>
84 static void DecIter(Iterator& i)
85 {
86 wxASSERT( IsValidUtf8LeadByte(*i) );
87
88 // Non-lead bytes are all in the 0x80..0xBF range (i.e. 10xxxxxx in
89 // binary), so we just have to go back until we hit a byte that is
90 // either < 0x80 (i.e. 0xxxxxxx in binary) or 0xC0..0xFF (11xxxxxx in
91 // binary; this includes some invalid values, but we can ignore it
92 // here, because we assume valid UTF-8 input for the purpose of
93 // efficient implementation).
94 --i;
95 while ( ((*i) & 0xC0) == 0x80 /* 2 highest bits are '10' */ )
96 --i;
97 }
98
99 template<typename Iterator>
100 static Iterator AddToIter(const Iterator& i, int n)
101 {
102 Iterator out(i);
103
104 if ( n > 0 )
105 {
106 for ( int j = 0; j < n; ++j )
107 IncIter(out);
108 }
109 else if ( n < 0 )
110 {
111 for ( int j = 0; j > n; --j )
112 DecIter(out);
113 }
114
115 return out;
116 }
117
118 template<typename Iterator>
119 static int DiffIters(Iterator i1, Iterator i2)
120 {
121 int dist = 0;
122
123 if ( i1 < i2 )
124 {
125 while ( i1 != i2 )
126 {
127 IncIter(i1);
128 dist--;
129 }
130 }
131 else if ( i2 < i1 )
132 {
133 while ( i2 != i1 )
134 {
135 IncIter(i2);
136 dist++;
137 }
138 }
139
140 return dist;
141 }
142
143 // encodes the character as UTF-8:
144 typedef wxUniChar::Utf8CharBuffer Utf8CharBuffer;
145 static Utf8CharBuffer EncodeChar(const wxUniChar& ch)
146 { return ch.AsUTF8(); }
147
148 // returns n copies of ch encoded in UTF-8 string
149 static wxCharBuffer EncodeNChars(size_t n, const wxUniChar& ch);
150
151 // returns the length of UTF-8 encoding of the character with lead byte 'c'
152 static size_t GetUtf8CharLength(char c)
153 {
154 wxASSERT( IsValidUtf8LeadByte(c) );
155 return ms_utf8IterTable[(unsigned char)c];
156 }
157
158 // decodes single UTF-8 character from UTF-8 string
159 static wxUniChar DecodeChar(wxStringImpl::const_iterator i)
160 {
161 if ( (unsigned char)*i < 0x80 )
162 return (int)*i;
163 return DecodeNonAsciiChar(i);
164 }
165
166 private:
167 static wxUniChar DecodeNonAsciiChar(wxStringImpl::const_iterator i);
168 };
169 #endif // wxUSE_UNICODE_UTF8
170
171
172 #if wxUSE_UNICODE_UTF8
173 typedef wxStringOperationsUtf8 wxStringOperations;
174 #else
175 typedef wxStringOperationsWchar wxStringOperations;
176 #endif
177
178 #endif // _WX_WXSTRINGOPS_H_