]>
Commit | Line | Data |
---|---|---|
dd0ef332 VS |
1 | ///////////////////////////////////////////////////////////////////////////// |
2 | // Name: src/common/unichar.cpp | |
3 | // Purpose: wxUniChar and wxUniCharRef classes | |
4 | // Author: Vaclav Slavik | |
5 | // Created: 2007-03-19 | |
6 | // RCS-ID: $Id$ | |
7 | // Copyright: (c) 2007 REA Elektronik GmbH | |
8 | // Licence: wxWindows licence | |
9 | /////////////////////////////////////////////////////////////////////////////// | |
10 | ||
11 | // =========================================================================== | |
12 | // headers | |
13 | // =========================================================================== | |
14 | ||
15 | // For compilers that support precompilation, includes "wx.h". | |
16 | #include "wx/wxprec.h" | |
17 | ||
18 | #ifdef __BORLANDC__ | |
19 | #pragma hdrstop | |
20 | #endif | |
21 | ||
0cb6a6e4 VZ |
22 | #ifndef WX_PRECOMP |
23 | #include "wx/strconv.h" // wxConvLibc | |
98d2df74 | 24 | #include "wx/log.h" |
0cb6a6e4 VZ |
25 | #endif |
26 | ||
dd0ef332 | 27 | #include "wx/unichar.h" |
155c2f6c | 28 | #include "wx/string.h" |
81727065 | 29 | |
dd0ef332 VS |
30 | // =========================================================================== |
31 | // implementation | |
32 | // =========================================================================== | |
33 | ||
81727065 VS |
34 | // --------------------------------------------------------------------------- |
35 | // wxUniChar | |
36 | // --------------------------------------------------------------------------- | |
37 | ||
dd0ef332 | 38 | /* static */ |
ce65118e | 39 | wxUniChar::value_type wxUniChar::FromHi8bit(char c) |
dd0ef332 | 40 | { |
111d9948 | 41 | #if wxUSE_UTF8_LOCALE_ONLY |
68fb51cd | 42 | wxFAIL_MSG( "invalid UTF-8 character" ); |
eea9eca5 VZ |
43 | wxUnusedVar(c); |
44 | ||
111d9948 VS |
45 | return wxT('?'); // FIXME-UTF8: what to use as failure character? |
46 | #else | |
7a8e90dd VZ |
47 | char cbuf[2]; |
48 | cbuf[0] = c; | |
49 | cbuf[1] = '\0'; | |
50 | wchar_t wbuf[2]; | |
51 | if ( wxConvLibc.ToWChar(wbuf, 2, cbuf, 2) != 2 ) | |
68fb51cd VS |
52 | { |
53 | wxFAIL_MSG( "invalid multibyte character" ); | |
dd0ef332 | 54 | return wxT('?'); // FIXME-UTF8: what to use as failure character? |
68fb51cd | 55 | } |
7a8e90dd | 56 | return wbuf[0]; |
111d9948 | 57 | #endif |
dd0ef332 VS |
58 | } |
59 | ||
60 | /* static */ | |
ce65118e | 61 | char wxUniChar::ToHi8bit(wxUniChar::value_type c) |
dd0ef332 | 62 | { |
111d9948 | 63 | #if wxUSE_UTF8_LOCALE_ONLY |
68fb51cd | 64 | wxFAIL_MSG( "character cannot be converted to single UTF-8 byte" ); |
eea9eca5 VZ |
65 | wxUnusedVar(c); |
66 | ||
111d9948 VS |
67 | return '?'; // FIXME-UTF8: what to use as failure character? |
68 | #else | |
7a8e90dd VZ |
69 | wchar_t wbuf[2]; |
70 | wbuf[0] = c; | |
71 | wbuf[1] = L'\0'; | |
72 | char cbuf[2]; | |
73 | if ( wxConvLibc.FromWChar(cbuf, 2, wbuf, 2) != 2 ) | |
68fb51cd VS |
74 | { |
75 | wxFAIL_MSG( "character cannot be converted to single byte" ); | |
dd0ef332 | 76 | return '?'; // FIXME-UTF8: what to use as failure character? |
68fb51cd | 77 | } |
7a8e90dd | 78 | return cbuf[0]; |
111d9948 | 79 | #endif |
dd0ef332 | 80 | } |
81727065 VS |
81 | |
82 | ||
83 | // --------------------------------------------------------------------------- | |
84 | // wxUniCharRef | |
85 | // --------------------------------------------------------------------------- | |
86 | ||
87 | #if wxUSE_UNICODE_UTF8 | |
467175ab VS |
88 | wxUniChar wxUniCharRef::UniChar() const |
89 | { | |
90 | return wxStringOperations::DecodeChar(m_pos); | |
91 | } | |
92 | ||
81727065 VS |
93 | wxUniCharRef& wxUniCharRef::operator=(const wxUniChar& c) |
94 | { | |
467175ab VS |
95 | wxStringOperations::Utf8CharBuffer utf(wxStringOperations::EncodeChar(c)); |
96 | size_t lenOld = wxStringOperations::GetUtf8CharLength(*m_pos); | |
97 | size_t lenNew = wxStringOperations::GetUtf8CharLength(utf[0]); | |
81727065 VS |
98 | |
99 | if ( lenNew == lenOld ) | |
100 | { | |
b0c4d5d7 VS |
101 | // this is the simpler case: if the new value's UTF-8 code has the |
102 | // same length, we can just replace it: | |
103 | ||
81727065 VS |
104 | iterator pos(m_pos); |
105 | for ( size_t i = 0; i < lenNew; ++i, ++pos ) | |
106 | *pos = utf[i]; | |
107 | } | |
68482dc5 | 108 | else // length of character encoding in UTF-8 changed |
81727065 | 109 | { |
b0c4d5d7 VS |
110 | // the worse case is when the new value has either longer or shorter |
111 | // code -- in that case, we have to use wxStringImpl::replace() and | |
112 | // this invalidates all iterators, so we have to update them too: | |
113 | ||
6bd4f281 | 114 | wxStringImpl& strimpl = m_str.m_impl; |
b0c4d5d7 VS |
115 | |
116 | int iterDiff = lenNew - lenOld; | |
117 | size_t posIdx = m_pos - strimpl.begin(); | |
118 | ||
119 | // compute positions of outstanding iterators for this string after the | |
120 | // replacement is done (there is only a small number of iterators at | |
121 | // any time, so we use an array on the stack to avoid unneeded | |
122 | // allocation): | |
123 | static const size_t STATIC_SIZE = 32; | |
124 | size_t indexes_a[STATIC_SIZE]; | |
125 | size_t *indexes = indexes_a; | |
126 | size_t iterNum = 0; | |
127 | wxStringIteratorNode *it; | |
6bd4f281 | 128 | for ( it = m_str.m_iterators.ptr; it; it = it->m_next, ++iterNum ) |
b0c4d5d7 VS |
129 | { |
130 | wxASSERT( it->m_iter || it->m_citer ); | |
131 | ||
132 | if ( iterNum == STATIC_SIZE ) | |
133 | { | |
9a83f860 | 134 | wxLogTrace( wxT("utf8"), wxT("unexpectedly many iterators") ); |
b0c4d5d7 VS |
135 | |
136 | size_t total = iterNum + 1; | |
137 | for ( wxStringIteratorNode *it2 = it; it2; it2 = it2->m_next ) | |
138 | total++; | |
139 | indexes = new size_t[total]; | |
140 | memcpy(indexes, indexes_a, sizeof(size_t) * STATIC_SIZE); | |
141 | } | |
142 | ||
143 | size_t idx = it->m_iter | |
144 | ? (*it->m_iter - strimpl.begin()) | |
145 | : (*it->m_citer - strimpl.begin()); | |
146 | ||
147 | if ( idx > posIdx ) | |
148 | idx += iterDiff; | |
149 | ||
150 | indexes[iterNum] = idx; | |
151 | } | |
152 | ||
153 | // update the string: | |
154 | strimpl.replace(m_pos, m_pos + lenOld, utf, lenNew); | |
155 | ||
68482dc5 VZ |
156 | #if wxUSE_STRING_POS_CACHE |
157 | m_str.InvalidateCache(); | |
158 | #endif // wxUSE_STRING_POS_CACHE | |
159 | ||
b0c4d5d7 VS |
160 | // finally, set the iterators to valid values again (note that this |
161 | // updates m_pos as well): | |
162 | size_t i; | |
6bd4f281 | 163 | for ( i = 0, it = m_str.m_iterators.ptr; it; it = it->m_next, ++i ) |
b0c4d5d7 VS |
164 | { |
165 | wxASSERT( i < iterNum ); | |
166 | wxASSERT( it->m_iter || it->m_citer ); | |
167 | ||
168 | if ( it->m_iter ) | |
169 | *it->m_iter = strimpl.begin() + indexes[i]; | |
170 | else // it->m_citer | |
171 | *it->m_citer = strimpl.begin() + indexes[i]; | |
172 | } | |
173 | ||
174 | if ( indexes != indexes_a ) | |
175 | delete[] indexes; | |
81727065 VS |
176 | } |
177 | ||
178 | return *this; | |
179 | } | |
180 | #endif // wxUSE_UNICODE_UTF8 |