]> git.saurik.com Git - wxWidgets.git/blob - src/common/encconv.cpp
5ad61864642eca6e5238292ef6dbc447c3335b12
[wxWidgets.git] / src / common / encconv.cpp
1 /////////////////////////////////////////////////////////////////////////////
2 // Name: encconv.cpp
3 // Purpose: wxEncodingConverter class for converting between different
4 // font encodings
5 // Author: Vaclav Slavik
6 // Copyright: (c) 1999 Vaclav Slavik
7 // Licence: wxWindows Licence
8 /////////////////////////////////////////////////////////////////////////////
9
10 #ifdef __GNUG__
11 #pragma implementation "encconv.h"
12 #endif
13
14 // For compilers that support precompilation, includes "wx.h".
15 #include "wx/wxprec.h"
16
17 #ifdef __BORLANDC__
18 #pragma hdrstop
19 #endif
20
21 #if wxUSE_FONTMAP
22
23 #include "wx/encconv.h"
24
25 #include <stdlib.h>
26
27 // conversion tables, generated by scripts in $(WXWIN)/misc/unictabl:
28 #ifdef __BORLANDC__
29 #include "../common/unictabl.inc"
30 #else
31 #include "unictabl.inc"
32 #endif
33
34 #if wxUSE_WCHAR_T
35 typedef wchar_t tchar;
36 #else
37 typedef char tchar;
38 #endif
39
40 #ifdef __WXWINCE__
41 #undef LINKAGEMODE
42 #define LINKAGEMODE __cdecl
43 #endif
44
45 static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc)
46 {
47 for (int i = 0; encodings_list[i].table != NULL; i++)
48 {
49 if (encodings_list[i].encoding == enc)
50 return encodings_list[i].table;
51 }
52 return NULL;
53 }
54
55 typedef struct {
56 wxUint16 u;
57 wxUint8 c;
58 } CharsetItem;
59
60 extern "C" int LINKAGEMODE CompareCharsetItems(const void *i1, const void *i2)
61 {
62 return ( ((CharsetItem*)i1) -> u - ((CharsetItem*)i2) -> u );
63 }
64
65
66 static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl)
67 {
68 CharsetItem *rev = new CharsetItem[128];
69
70 for (int i = 0; i < 128; i++)
71 rev[i].c = 128 + i, rev[i].u = tbl[i];
72
73 qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems);
74
75 return rev;
76 }
77
78
79
80 wxEncodingConverter::wxEncodingConverter()
81 {
82 m_Table = NULL;
83 m_UnicodeInput = m_UnicodeOutput = FALSE;
84 m_JustCopy = FALSE;
85 }
86
87
88
89 bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method)
90 {
91 unsigned i;
92 wxUint16 *in_tbl = NULL, *out_tbl = NULL;
93
94 if (m_Table) {delete[] m_Table; m_Table = NULL;}
95
96 #if !wxUSE_WCHAR_T
97 if (input_enc == wxFONTENCODING_UNICODE || output_enc == wxFONTENCODING_UNICODE) return FALSE;
98 #endif
99
100 if (input_enc == output_enc) {m_JustCopy = TRUE; return TRUE;}
101
102 m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE);
103 m_JustCopy = FALSE;
104
105 if (input_enc == wxFONTENCODING_UNICODE)
106 {
107 if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;
108
109 m_Table = new tchar[65536];
110 for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII
111 for (i = 128; i < 65536; i++) m_Table[i] = (tchar)'?';
112 // FIXME - this should be character that means `unicode to charset' impossible, not '?'
113
114 if (method == wxCONVERT_SUBSTITUTE)
115 {
116 for (i = 0; i < encoding_unicode_fallback_count; i++)
117 m_Table[encoding_unicode_fallback[i].c] = (tchar) encoding_unicode_fallback[i].s;
118 }
119
120 for (i = 0; i < 128; i++)
121 m_Table[out_tbl[i]] = (tchar)(128 + i);
122
123 m_UnicodeInput = TRUE;
124 return TRUE;
125 }
126
127 else
128 {
129 if ((in_tbl = GetEncTable(input_enc)) == NULL) return FALSE;
130 if (output_enc != wxFONTENCODING_UNICODE)
131 if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;
132
133 m_UnicodeInput = FALSE;
134
135 m_Table = new tchar[256];
136 for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII
137
138 if (output_enc == wxFONTENCODING_UNICODE)
139 {
140 for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i];
141 return TRUE;
142 }
143 // FIXME: write a substitute for bsearch
144 #ifndef __WXWINCE__
145 else
146 {
147 CharsetItem *rev = BuildReverseTable(out_tbl);
148 CharsetItem *item;
149 CharsetItem key;
150
151 for (i = 0; i < 128; i++)
152 {
153 key.u = in_tbl[i];
154 item = (CharsetItem*) bsearch(&key, rev, 128, sizeof(CharsetItem), CompareCharsetItems);
155 if (item == NULL && method == wxCONVERT_SUBSTITUTE)
156 item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback,
157 encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems);
158 if (item)
159 m_Table[128 + i] = (tchar)item -> c;
160 else
161 #if wxUSE_WCHAR_T
162 m_Table[128 + i] = (wchar_t)(128 + i);
163 #else
164 m_Table[128 + i] = (char)(128 + i);
165 #endif
166 }
167
168 delete[] rev;
169 return TRUE;
170 }
171 #endif
172 return TRUE;
173 }
174 }
175
176
177
178 void wxEncodingConverter::Convert(const char* input, char* output)
179 {
180 wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
181 wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
182
183 const char *i;
184 char *o;
185
186 if (m_JustCopy)
187 {
188 strcpy(output, input);
189 return;
190 }
191
192 wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
193
194 for (i = input, o = output; *i != 0;)
195 *(o++) = (char)(m_Table[(wxUint8)*(i++)]);
196 *o = 0;
197 }
198
199
200 #if wxUSE_WCHAR_T
201
202 void wxEncodingConverter::Convert(const char* input, wchar_t* output)
203 {
204 wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
205 wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
206
207 const char *i;
208 wchar_t *o;
209
210 if (m_JustCopy)
211 {
212 for (i = input, o = output; *i != 0;)
213 *(o++) = (wchar_t)(*(i++));
214 *o = 0;
215 return;
216 }
217
218 wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
219
220 for (i = input, o = output; *i != 0;)
221 *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]);
222 *o = 0;
223 }
224
225
226
227 void wxEncodingConverter::Convert(const wchar_t* input, char* output)
228 {
229 wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
230 wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
231
232 const wchar_t *i;
233 char *o;
234
235 if (m_JustCopy)
236 {
237 for (i = input, o = output; *i != 0;)
238 *(o++) = (char)(*(i++));
239 *o = 0;
240 return;
241 }
242
243 wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
244
245 for (i = input, o = output; *i != 0;)
246 *(o++) = (char)(m_Table[(wxUint16)*(i++)]);
247 *o = 0;
248 }
249
250
251
252 void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output)
253 {
254 wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
255 wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
256
257 const wchar_t *i;
258 wchar_t *o;
259
260 if (m_JustCopy)
261 {
262 // wcscpy() is not guaranteed to exist
263 for (i = input, o = output; *i != 0;)
264 *(o++) = (*(i++));
265 *o = 0;
266 return;
267 }
268
269 wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
270
271 for (i = input, o = output; *i != 0;)
272 *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]);
273 *o = 0;
274 }
275
276 #endif // wxUSE_WCHAR_T
277
278
279 wxString wxEncodingConverter::Convert(const wxString& input)
280 {
281 if (m_JustCopy) return input;
282
283 wxString s;
284 const wxChar *i;
285
286 wxCHECK_MSG(m_Table != NULL, s,
287 wxT("You must call wxEncodingConverter::Init() before actually converting!"));
288
289 if (m_UnicodeInput)
290 {
291 for (i = input.c_str(); *i != 0; i++)
292 s << (wxChar)(m_Table[(wxUint16)*i]);
293 }
294 else
295 {
296 for (i = input.c_str(); *i != 0; i++)
297 s << (wxChar)(m_Table[(wxUint8)*i]);
298 }
299
300 return s;
301 }
302
303
304
305
306
307
308
309 // Following tables describe classes of encoding equivalence.
310 //
311
312 #define STOP wxFONTENCODING_SYSTEM
313
314 #define NUM_OF_PLATFORMS 4 /*must conform to enum wxPLATFORM_XXXX !!!*/
315 #define ENC_PER_PLATFORM 5
316 // max no. of encodings for one language used on one platform
317 // Anybody thinks 5 is not enough? ;-)
318
319 static wxFontEncoding
320 EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = {
321
322 // *** Please put more common encodings as first! ***
323
324 // Western European
325 {
326 /* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP},
327 /* windows */ {wxFONTENCODING_CP1252, STOP},
328 /* os2 */ {STOP},
329 /* mac */ {STOP}
330 },
331
332 // Central European
333 {
334 /* unix */ {wxFONTENCODING_ISO8859_2, STOP},
335 /* windows */ {wxFONTENCODING_CP1250, STOP},
336 /* os2 */ {STOP},
337 /* mac */ {STOP}
338 },
339
340 // Baltic
341 {
342 /* unix */ {wxFONTENCODING_ISO8859_13, wxFONTENCODING_ISO8859_4, STOP},
343 /* windows */ {wxFONTENCODING_CP1257, STOP},
344 /* os2 */ {STOP},
345 /* mac */ {STOP}
346 },
347
348 // Hebrew
349 {
350 /* unix */ {wxFONTENCODING_ISO8859_8, STOP},
351 /* windows */ {wxFONTENCODING_CP1255, STOP},
352 /* os2 */ {STOP},
353 /* mac */ {STOP}
354 },
355
356 // Greek
357 {
358 /* unix */ {wxFONTENCODING_ISO8859_7, STOP},
359 /* windows */ {wxFONTENCODING_CP1253, STOP},
360 /* os2 */ {STOP},
361 /* mac */ {STOP}
362 },
363
364 // Arabic
365 {
366 /* unix */ {wxFONTENCODING_ISO8859_6, STOP},
367 /* windows */ {wxFONTENCODING_CP1256, STOP},
368 /* os2 */ {STOP},
369 /* mac */ {STOP}
370 },
371
372 // Turkish
373 {
374 /* unix */ {wxFONTENCODING_ISO8859_9, STOP},
375 /* windows */ {wxFONTENCODING_CP1254, STOP},
376 /* os2 */ {STOP},
377 /* mac */ {STOP}
378 },
379
380 // Cyrillic
381 {
382 /* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_ISO8859_5, STOP},
383 /* windows */ {wxFONTENCODING_CP1251, STOP},
384 /* os2 */ {STOP},
385 /* mac */ {STOP}
386 },
387
388 {{STOP},{STOP},{STOP},{STOP}} /* Terminator */
389 /* no, _not_ Arnold! */
390 };
391
392
393 static bool FindEncoding(const wxFontEncodingArray& arr, wxFontEncoding f)
394 {
395 for (wxFontEncodingArray::const_iterator it = arr.begin(), en = arr.end();
396 it != en; ++it)
397 if (*it == f)
398 return true;
399 return false;
400 }
401
402 wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding enc, int platform)
403 {
404 if (platform == wxPLATFORM_CURRENT)
405 {
406 #if defined(__WXMSW__)
407 platform = wxPLATFORM_WINDOWS;
408 #elif defined(__WXGTK__) || defined(__WXMOTIF__)
409 platform = wxPLATFORM_UNIX;
410 #elif defined(__WXOS2__)
411 platform = wxPLATFORM_OS2;
412 #elif defined(__WXMAC__)
413 platform = wxPLATFORM_MAC;
414 #endif
415 }
416
417 int i, clas, e ;
418 wxFontEncoding *f;
419 wxFontEncodingArray arr;
420
421 clas = 0;
422 while (EquivalentEncodings[clas][0][0] != STOP)
423 {
424 for (i = 0; i < NUM_OF_PLATFORMS; i++)
425 for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
426 if (EquivalentEncodings[clas][i][e] == enc)
427 {
428 for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
429 if (*f == enc) arr.push_back(enc);
430 for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
431 if (!FindEncoding(arr, *f)) arr.push_back(*f);
432 i = NUM_OF_PLATFORMS/*hack*/; break;
433 }
434 clas++;
435 }
436
437 return arr;
438 }
439
440
441
442 wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc)
443 {
444 int i, clas, e, j ;
445 wxFontEncoding *f;
446 wxFontEncodingArray arr;
447
448 arr = GetPlatformEquivalents(enc); // we want them to be first items in array
449
450 clas = 0;
451 while (EquivalentEncodings[clas][0][0] != STOP)
452 {
453 for (i = 0; i < NUM_OF_PLATFORMS; i++)
454 for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
455 if (EquivalentEncodings[clas][i][e] == enc)
456 {
457 for (j = 0; j < NUM_OF_PLATFORMS; j++)
458 for (f = EquivalentEncodings[clas][j]; *f != STOP; f++)
459 if (!FindEncoding(arr, *f)) arr.push_back(*f);
460 i = NUM_OF_PLATFORMS/*hack*/; break;
461 }
462 clas++;
463 }
464
465 return arr;
466 }
467
468 #endif // wxUSE_FONTMAP