]> git.saurik.com Git - wxWidgets.git/blob - src/common/encconv.cpp
String corrections
[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 #if defined(__GNUG__) && !defined(NO_GCC_PRAGMA)
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 #if defined( __BORLANDC__ ) || defined(__DARWIN__)
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 #ifdef __WXMAC__
46
47 #include "ATSUnicode.h"
48 #include "TextCommon.h"
49 #include "TextEncodingConverter.h"
50
51 #include "wx/fontutil.h"
52 #include "wx/mac/private.h" // includes mac headers
53
54 wxUint16 gMacEncodings[wxFONTENCODING_MACMAX-wxFONTENCODING_MACMIN+1][128] ;
55 bool gMacEncodingsInited[wxFONTENCODING_MACMAX-wxFONTENCODING_MACMIN+1] ;
56
57 #endif
58
59 static wxUint16* LINKAGEMODE GetEncTable(wxFontEncoding enc)
60 {
61 #ifdef __WXMAC__
62 if( enc >= wxFONTENCODING_MACMIN && enc <= wxFONTENCODING_MACMAX )
63 {
64 int i = enc-wxFONTENCODING_MACMIN ;
65 if ( gMacEncodingsInited[i] == false )
66 {
67 TECObjectRef converter ;
68 TextEncodingBase code = wxMacGetSystemEncFromFontEnc( enc ) ;
69 TextEncodingBase unicode = CreateTextEncoding(kTextEncodingUnicodeDefault,0,kUnicode16BitFormat) ;
70 OSStatus status = TECCreateConverter(&converter,code,unicode);
71 char s[2] ;
72 s[1] = 0 ;
73 ByteCount byteInLen, byteOutLen ;
74 for( unsigned char c = 255 ; c >= 128 ; --c )
75 {
76 s[0] = c ;
77 status = TECConvertText(converter, (ConstTextPtr) &s , 1, &byteInLen,
78 (TextPtr) &gMacEncodings[i][c-128] , 2, &byteOutLen);
79 }
80 status = TECDisposeConverter(converter);
81 gMacEncodingsInited[i]=true;
82 }
83 return gMacEncodings[i] ;
84 }
85 #endif
86
87 for (int i = 0; encodings_list[i].table != NULL; i++)
88 {
89 if (encodings_list[i].encoding == enc)
90 return encodings_list[i].table;
91 }
92 return NULL;
93 }
94
95 typedef struct {
96 wxUint16 u;
97 wxUint8 c;
98 } CharsetItem;
99
100 extern "C" int LINKAGEMODE CompareCharsetItems(const void *i1, const void *i2)
101 {
102 return ( ((CharsetItem*)i1) -> u - ((CharsetItem*)i2) -> u );
103 }
104
105
106 static CharsetItem* LINKAGEMODE BuildReverseTable(wxUint16 *tbl)
107 {
108 CharsetItem *rev = new CharsetItem[128];
109
110 for (int i = 0; i < 128; i++)
111 rev[i].c = 128 + i, rev[i].u = tbl[i];
112
113 qsort(rev, 128, sizeof(CharsetItem), CompareCharsetItems);
114
115 return rev;
116 }
117
118
119
120 wxEncodingConverter::wxEncodingConverter()
121 {
122 m_Table = NULL;
123 m_UnicodeInput = m_UnicodeOutput = FALSE;
124 m_JustCopy = FALSE;
125 }
126
127
128
129 bool wxEncodingConverter::Init(wxFontEncoding input_enc, wxFontEncoding output_enc, int method)
130 {
131 unsigned i;
132 wxUint16 *in_tbl, *out_tbl = NULL;
133
134 if (m_Table) {delete[] m_Table; m_Table = NULL;}
135
136 #if !wxUSE_WCHAR_T
137 if (input_enc == wxFONTENCODING_UNICODE || output_enc == wxFONTENCODING_UNICODE) return FALSE;
138 #endif
139
140 if (input_enc == output_enc) {m_JustCopy = TRUE; return TRUE;}
141
142 m_UnicodeOutput = (output_enc == wxFONTENCODING_UNICODE);
143 m_JustCopy = FALSE;
144
145 if (input_enc == wxFONTENCODING_UNICODE)
146 {
147 if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;
148
149 m_Table = new tchar[65536];
150 for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII
151 for (i = 128; i < 65536; i++) m_Table[i] = (tchar)'?';
152 // FIXME - this should be character that means `unicode to charset' impossible, not '?'
153
154 if (method == wxCONVERT_SUBSTITUTE)
155 {
156 for (i = 0; i < encoding_unicode_fallback_count; i++)
157 m_Table[encoding_unicode_fallback[i].c] = (tchar) encoding_unicode_fallback[i].s;
158 }
159
160 for (i = 0; i < 128; i++)
161 m_Table[out_tbl[i]] = (tchar)(128 + i);
162
163 m_UnicodeInput = TRUE;
164 }
165 else // input !Unicode
166 {
167 if ((in_tbl = GetEncTable(input_enc)) == NULL) return FALSE;
168 if (output_enc != wxFONTENCODING_UNICODE)
169 if ((out_tbl = GetEncTable(output_enc)) == NULL) return FALSE;
170
171 m_UnicodeInput = FALSE;
172
173 m_Table = new tchar[256];
174 for (i = 0; i < 128; i++) m_Table[i] = (tchar)i; // 7bit ASCII
175
176 if (output_enc == wxFONTENCODING_UNICODE)
177 {
178 for (i = 0; i < 128; i++) m_Table[128 + i] = (tchar)in_tbl[i];
179 return TRUE;
180 }
181 // FIXME: write a substitute for bsearch
182 #ifndef __WXWINCE__
183 else // output !Unicode
184 {
185 CharsetItem *rev = BuildReverseTable(out_tbl);
186 CharsetItem *item;
187 CharsetItem key;
188
189 for (i = 0; i < 128; i++)
190 {
191 key.u = in_tbl[i];
192 item = (CharsetItem*) bsearch(&key, rev, 128, sizeof(CharsetItem), CompareCharsetItems);
193 if (item == NULL && method == wxCONVERT_SUBSTITUTE)
194 item = (CharsetItem*) bsearch(&key, encoding_unicode_fallback,
195 encoding_unicode_fallback_count, sizeof(CharsetItem), CompareCharsetItems);
196 if (item)
197 m_Table[128 + i] = (tchar)item -> c;
198 else
199 #if wxUSE_WCHAR_T
200 m_Table[128 + i] = (wchar_t)(128 + i);
201 #else
202 m_Table[128 + i] = (char)(128 + i);
203 #endif
204 }
205
206 delete[] rev;
207 }
208 #endif // !__WXWINCE__
209 }
210
211 return TRUE;
212 }
213
214
215
216 void wxEncodingConverter::Convert(const char* input, char* output) const
217 {
218 wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
219 wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
220
221 const char *i;
222 char *o;
223
224 if (m_JustCopy)
225 {
226 strcpy(output, input);
227 return;
228 }
229
230 wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
231
232 for (i = input, o = output; *i != 0;)
233 *(o++) = (char)(m_Table[(wxUint8)*(i++)]);
234 *o = 0;
235 }
236
237
238 #if wxUSE_WCHAR_T
239
240 void wxEncodingConverter::Convert(const char* input, wchar_t* output) const
241 {
242 wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
243 wxASSERT_MSG(!m_UnicodeInput, wxT("You cannot convert from unicode if input is const char*!"));
244
245 const char *i;
246 wchar_t *o;
247
248 if (m_JustCopy)
249 {
250 for (i = input, o = output; *i != 0;)
251 *(o++) = (wchar_t)(*(i++));
252 *o = 0;
253 return;
254 }
255
256 wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
257
258 for (i = input, o = output; *i != 0;)
259 *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]);
260 *o = 0;
261 }
262
263
264
265 void wxEncodingConverter::Convert(const wchar_t* input, char* output) const
266 {
267 wxASSERT_MSG(!m_UnicodeOutput, wxT("You cannot convert to unicode if output is const char*!"));
268 wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
269
270 const wchar_t *i;
271 char *o;
272
273 if (m_JustCopy)
274 {
275 for (i = input, o = output; *i != 0;)
276 *(o++) = (char)(*(i++));
277 *o = 0;
278 return;
279 }
280
281 wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
282
283 for (i = input, o = output; *i != 0;)
284 *(o++) = (char)(m_Table[(wxUint16)*(i++)]);
285 *o = 0;
286 }
287
288
289
290 void wxEncodingConverter::Convert(const wchar_t* input, wchar_t* output) const
291 {
292 wxASSERT_MSG(m_UnicodeOutput, wxT("You cannot convert to 8-bit if output is const wchar_t*!"));
293 wxASSERT_MSG(m_UnicodeInput, wxT("You cannot convert from 8-bit if input is const wchar_t*!"));
294
295 const wchar_t *i;
296 wchar_t *o;
297
298 if (m_JustCopy)
299 {
300 // wcscpy() is not guaranteed to exist
301 for (i = input, o = output; *i != 0;)
302 *(o++) = (*(i++));
303 *o = 0;
304 return;
305 }
306
307 wxCHECK_RET(m_Table != NULL, wxT("You must call wxEncodingConverter::Init() before actually converting!"));
308
309 for (i = input, o = output; *i != 0;)
310 *(o++) = (wchar_t)(m_Table[(wxUint8)*(i++)]);
311 *o = 0;
312 }
313
314 #endif // wxUSE_WCHAR_T
315
316
317 wxString wxEncodingConverter::Convert(const wxString& input) const
318 {
319 if (m_JustCopy) return input;
320
321 wxString s;
322 const wxChar *i;
323
324 wxCHECK_MSG(m_Table != NULL, s,
325 wxT("You must call wxEncodingConverter::Init() before actually converting!"));
326
327 if (m_UnicodeInput)
328 {
329 for (i = input.c_str(); *i != 0; i++)
330 s << (wxChar)(m_Table[(wxUint16)*i]);
331 }
332 else
333 {
334 for (i = input.c_str(); *i != 0; i++)
335 s << (wxChar)(m_Table[(wxUint8)*i]);
336 }
337
338 return s;
339 }
340
341
342
343
344
345
346
347 // Following tables describe classes of encoding equivalence.
348 //
349
350 #define STOP wxFONTENCODING_SYSTEM
351
352 #define NUM_OF_PLATFORMS 4 /*must conform to enum wxPLATFORM_XXXX !!!*/
353 #define ENC_PER_PLATFORM 5
354 // max no. of encodings for one language used on one platform
355 // Anybody thinks 5 is not enough? ;-)
356
357 static wxFontEncoding
358 EquivalentEncodings[][NUM_OF_PLATFORMS][ENC_PER_PLATFORM+1] = {
359
360 // *** Please put more common encodings as first! ***
361
362 // Western European
363 {
364 /* unix */ {wxFONTENCODING_ISO8859_1, wxFONTENCODING_ISO8859_15, STOP},
365 /* windows */ {wxFONTENCODING_CP1252, STOP},
366 /* os2 */ {STOP},
367 /* mac */ {wxFONTENCODING_MACROMAN, STOP}
368 },
369
370 // Central European
371 {
372 /* unix */ {wxFONTENCODING_ISO8859_2, STOP},
373 /* windows */ {wxFONTENCODING_CP1250, STOP},
374 /* os2 */ {STOP},
375 /* mac */ {wxFONTENCODING_MACCENTRALEUR, STOP}
376 },
377
378 // Baltic
379 {
380 /* unix */ {wxFONTENCODING_ISO8859_13, wxFONTENCODING_ISO8859_4, STOP},
381 /* windows */ {wxFONTENCODING_CP1257, STOP},
382 /* os2 */ {STOP},
383 /* mac */ {STOP}
384 },
385
386 // Hebrew
387 {
388 /* unix */ {wxFONTENCODING_ISO8859_8, STOP},
389 /* windows */ {wxFONTENCODING_CP1255, STOP},
390 /* os2 */ {STOP},
391 /* mac */ {wxFONTENCODING_MACHEBREW, STOP}
392 },
393
394 // Greek
395 {
396 /* unix */ {wxFONTENCODING_ISO8859_7, STOP},
397 /* windows */ {wxFONTENCODING_CP1253, STOP},
398 /* os2 */ {STOP},
399 /* mac */ {wxFONTENCODING_MACGREEK, STOP}
400 },
401
402 // Arabic
403 {
404 /* unix */ {wxFONTENCODING_ISO8859_6, STOP},
405 /* windows */ {wxFONTENCODING_CP1256, STOP},
406 /* os2 */ {STOP},
407 /* mac */ {wxFONTENCODING_MACARABIC, STOP}
408 },
409
410 // Turkish
411 {
412 /* unix */ {wxFONTENCODING_ISO8859_9, STOP},
413 /* windows */ {wxFONTENCODING_CP1254, STOP},
414 /* os2 */ {STOP},
415 /* mac */ {wxFONTENCODING_MACTURKISH, STOP}
416 },
417
418 // Cyrillic
419 {
420 /* unix */ {wxFONTENCODING_KOI8, wxFONTENCODING_ISO8859_5, STOP},
421 /* windows */ {wxFONTENCODING_CP1251, STOP},
422 /* os2 */ {STOP},
423 /* mac */ {wxFONTENCODING_MACCYRILLIC, STOP}
424 },
425
426 {{STOP},{STOP},{STOP},{STOP}} /* Terminator */
427 /* no, _not_ Arnold! */
428 };
429
430
431 static bool FindEncoding(const wxFontEncodingArray& arr, wxFontEncoding f)
432 {
433 for (wxFontEncodingArray::const_iterator it = arr.begin(), en = arr.end();
434 it != en; ++it)
435 if (*it == f)
436 return true;
437 return false;
438 }
439
440 wxFontEncodingArray wxEncodingConverter::GetPlatformEquivalents(wxFontEncoding enc, int platform)
441 {
442 if (platform == wxPLATFORM_CURRENT)
443 {
444 #if defined(__WXMSW__)
445 platform = wxPLATFORM_WINDOWS;
446 #elif defined(__WXGTK__) || defined(__WXMOTIF__)
447 platform = wxPLATFORM_UNIX;
448 #elif defined(__WXOS2__)
449 platform = wxPLATFORM_OS2;
450 #elif defined(__WXMAC__)
451 platform = wxPLATFORM_MAC;
452 #endif
453 }
454
455 int i, clas, e ;
456 wxFontEncoding *f;
457 wxFontEncodingArray arr;
458
459 clas = 0;
460 while (EquivalentEncodings[clas][0][0] != STOP)
461 {
462 for (i = 0; i < NUM_OF_PLATFORMS; i++)
463 for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
464 if (EquivalentEncodings[clas][i][e] == enc)
465 {
466 for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
467 if (*f == enc) arr.push_back(enc);
468 for (f = EquivalentEncodings[clas][platform]; *f != STOP; f++)
469 if (!FindEncoding(arr, *f)) arr.push_back(*f);
470 i = NUM_OF_PLATFORMS/*hack*/; break;
471 }
472 clas++;
473 }
474
475 return arr;
476 }
477
478
479
480 wxFontEncodingArray wxEncodingConverter::GetAllEquivalents(wxFontEncoding enc)
481 {
482 int i, clas, e, j ;
483 wxFontEncoding *f;
484 wxFontEncodingArray arr;
485
486 arr = GetPlatformEquivalents(enc); // we want them to be first items in array
487
488 clas = 0;
489 while (EquivalentEncodings[clas][0][0] != STOP)
490 {
491 for (i = 0; i < NUM_OF_PLATFORMS; i++)
492 for (e = 0; EquivalentEncodings[clas][i][e] != STOP; e++)
493 if (EquivalentEncodings[clas][i][e] == enc)
494 {
495 for (j = 0; j < NUM_OF_PLATFORMS; j++)
496 for (f = EquivalentEncodings[clas][j]; *f != STOP; f++)
497 if (!FindEncoding(arr, *f)) arr.push_back(*f);
498 i = NUM_OF_PLATFORMS/*hack*/; break;
499 }
500 clas++;
501 }
502
503 return arr;
504 }
505
506 #endif // wxUSE_FONTMAP